METADATA
last updated: 2026-03-05 by BA
file_name: 2024-05-29_ACLA - LDT Complaint to FDA Court filing.md
file_date: 2024-05-29
title: ACLA - LDT Complaint to FDA Court filing
category: regulatory
subcategory: ldts
tags: 
source_file_type: pdf
xfile_type: NA
gfile_url: NA
xfile_github_download_url: NA
pdf_gdrive_url: https://drive.google.com/file/d/1IYnIE7-thqc7PAASkxy-L0YsyuQAp7en
pdf_github_url: https://github.com/FocusOnFoundationsNonprofit/floodlamp-archive/blob/main/regulatory/ldts/2024-05-29_ACLA%20-%20LDT%20Complaint%20to%20FDA%20Court%20filing.pdf
conversion_input_file_type: pdf
conversion: megaparse-v
license: Public Domain
tokens: 142259
words: 94535
notes: 
summary_short: The Complaint to FDA – ACLA LDT Complaint is a federal court filing by the American Clinical Laboratory Association and member laboratories challenging the FDA’s 2024 Final Rule that classifies laboratory-developed tests as medical devices under the FDCA. It argues the rule exceeds FDA’s statutory authority, conflicts with CLIA’s established regulatory framework for laboratory services, and is arbitrary and capricious under the Administrative Procedure Act. The complaint seeks declaratory and injunctive relief to vacate the rule and prevent FDA from enforcing device regulations on professional laboratory testing services.


CONTENT

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF TEXAS
SHERMAN DIVISION

AMERICAN CLINICAL LABORATORY ASSOCIATION; HEALTHTRACKRX INDIANA, INC.; and HEALTHTRACKRX, INC.,

Plaintiffs,

v.

U.S. FOOD AND DRUG ADMINISTRATION; U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES; XAVIER BECERRA, in his official capacity as Secretary of Health and Human Services; and ROBERT M. CALIFF, M.D., in his official capacity as Commissioner of Food and Drugs, United States Food and Drug Administration,

Defendants.

Case No.: 4:24-cv-479
COMPLAINT

## TABLE OF CONTENTS

PRELIMINARY STATEMENT ................................................................................ 1

PARTIES ..................................................................................................................... 9

JURISDICTION AND VENUE ................................................................................ 10

GENERAL ALLEGATIONS .................................................................................... 12
- A. Laboratory-developed tests are services carried out by highly skilled and trained laboratory professionals ..................... 12
- B. FDA’s statutory authority to regulate medical devices does not extend to professional services ............................................... 15
- C. Congress created a separate and distinct framework for regulating laboratory testing services .......................................... 21
- D. FDA has never broadly regulated laboratory testing services as medical devices ............................................................. 26
  1. FDA’s Many Years of Silence ..................................................... 27
  2. FDA’s Never-Finalized 1992 Guidance .................................... 29
  3. FDA’s Sporadic Claims of Authority in the 1990s and 2000s .................................................................................................... 30
  4. FDA’s Never-Finalized 2014 Guidance .................................... 31
  5. The 2020 Charrow Memo Questions FDA’s Authority to Regulate Laboratory Testing Services ............................ 35
  6. Congress Chooses Not to Enact Legislation ............................. 37
- E. FDA now seeks for the first time to classify virtually all laboratory testing services as medical devices .............................. 38
  1. The 2023 Proposed Rule .............................................................. 38
  2. ACLA’s Comments on the Proposed Rule ................................ 40
  3. The 2024 Final Rule .................................................................... 41
- F. HealthTrackRX and other ACLA members face irreparable harm from FDA’s final rule ............................................................ 50

CLAIMS FOR RELIEF ............................................................................................. 54

- COUNT 1 Violation of the Administrative Procedure Act—Contrary to Law, in Excess of Statutory Jurisdiction and Authority, and Contrary to Constitutional Right and Power 5 U.S.C. § 706(2) .......... 54
- COUNT 2 Violation of the Administrative Procedure Act—Arbitrary and Capricious and an Abuse of Discretion 5 U.S.C. § 706(2) ............ 57

PRAYER FOR RELIEF ........................................................................................... 58

## COMPLAINT FOR DECLARATORY AND INJUNCTIVE RELIEF

Plaintiffs American Clinical Laboratory Association (“ACLA”) and HealthTrackRX Indiana, Inc. and HealthTrackRX, Inc. (together “HealthTrackRX”) bring this action against the Food & Drug Administration (“FDA”), the Department of Health and Human Services, and the Secretary of Health and Human Services and the FDA Commissioner in their official capacities, challenging the final rule published on May 6, 2024, announcing FDA’s intent to regulate laboratory-developed tests as medical devices under the Federal, Food, Drug and Cosmetic Act. Because the final rule exceeds FDA’s lawful authority and is arbitrary and capricious and contrary to law, the rule should be set aside and vacated, and defendants should be enjoined from enforcing or implementing the rule. _See_ 5 U.S.C. § 706. Plaintiffs allege as follows:

### PRELIMINARY STATEMENT

1. The professional diagnostic testing services provided by clinical laboratories are an essential part of the nation’s healthcare system. These important testing services have long been relied on by healthcare providers to diagnose and develop appropriate treatments for patients who suffer from illness and disease. There are thousands of laboratories across the United States that offer tens of thousands of molecular and other types of high-quality diagnostic testing services to providers and patients. These testing services are a critical pillar of our nation’s health care system.

2. For decades, laboratory-developed testing services (often referred to as “LDTs”) have been regulated under a statutory and regulatory framework—the Clinical Laboratory Improvement Amendments Act of 1988 (“CLIA”)—that imposes numerous laboratory-specific standards to ensure the validity and reliability of laboratory diagnostic testing services, including the training and qualifications of the skilled professionals who perform, supervise, and interpret those tests. When creating and performing testing services, these laboratory professionals have not generally been required to comply also with the costly and burdensome pre-approval and clearance requirements that the Federal Food, Drug and Cosmetic Act (“FDCA”) authorizes FDA to apply to manufactured medical devices sold in interstate commerce. Nor has Congress ever granted FDA authority to regulate professional laboratory-developed testing services.

3. FDA’s final rule threatens to upend the nation’s entire laboratory profession by seeking to regulate all laboratory-developed tests as if they are medical devices under the FDCA. In asserting authority to transform the regulatory framework that has applied for decades, FDA cannot point to any new statutory authority granted by Congress. Nor can FDA contend that Congress has ever provided it with the resources that would be necessary to retain the personnel and build the expertise necessary to exercise sweeping authority over the thousands of testing services provided by the nation’s laboratories. To the contrary, Congress has recently entertained legislative proposals that would have granted FDA new authority to regulate laboratory-developed testing services, and it has declined to provide FDA that power.

4. FDA’s final rule relies on the extraordinary position that in 1976, when Congress expanded FDA’s authority to regulate medical devices, it also quietly intended to outlaw—and subject to substantial civil and criminal monetary penalties—any professional laboratory-developed testing services that were not first approved or cleared by FDA. The logic of FDA’s position is that tens of thousands of professionals across the country performing millions of diagnostic testing services every year, working with thousands of doctors and patients, have for decades done so in open and direct violation of the law. According to FDA, the only reason laboratories have not been civilly and criminally punished is because FDA has chosen to exercise unreviewable “enforcement discretion.” In short, FDA is taking the position that a “long-extant statute” grants it vast, “transformative” regulatory powers that it has not previously exercised—a position that courts have rightly approached with deep skepticism. _West Virginia v. EPA_, 597 U.S. 697, 724 (2022) (quoting _Util. Air Regul. Grp. v. EPA_, 573 U.S. 302, 324 (2014)).

5. If it is not vacated, FDA’s unprecedented final rule will have devastating and far-reaching consequences not only for the nation’s clinical
laboratories, but also for the nation’s entire healthcare system, including the millions of vulnerable patients who depend on the essential clinical testing services that laboratories provide. FDA’s final rule means that, in order to be _legally_ marketed, virtually all diagnostic laboratory tests will have to undergo costly and time-consuming administrative review through a regulatory process that was designed for evaluating manufactured medical devices, not professional testing services.

6. The final rule states that FDA intends to apply this onerous regulatory regime to new and modified laboratory-developed tests, which will dramatically increase research and development costs, hinder vital medical innovation, and hamper adaptation of existing tests to meet evolving patient needs. Indeed, FDA itself has recognized “significant regulatory changes” to the treatment of laboratory testing services “could have negative effects on the public health.” 62 Fed. Reg. 62,243, 62,249 (Nov. 21, 1997). With respect to unmodified existing tests, FDA states that as a matter of enforcement discretion it generally does not intend—at least not at this time—to enforce certain especially burdensome medical-device requirements, such as premarket review. But FDA’s final rule means that in the agency’s view all of those tests, including tests that physicians have relied on for decades, are being marketed illegally and are subject to FDA enforcement action at any time.

7. FDA does not have authority to regulate professional laboratory-developed testing services as medical devices. The text and structure of the FDCA make plain that FDA’s authority to regulate “devices,” which dates to 1938 and was expanded through the Medical Device Amendments of 1976, extends only to physical products that are sold and distributed by manufacturers in interstate commerce. The FDCA has never applied medical device regulation to laboratory testing services. And for good reason: Those tests are not physical products sold and distributed by manufacturers. Instead, they are professional healthcare services offered by highly skilled and trained laboratory professionals that are outside FDA’s regulatory expertise and are subject to different regulatory requirements. A laboratory-developed test is a process by which laboratory professionals use various tools—some of which may be individually regulated as devices—to derive diagnostic information that a patient and the patient’s physician may use in making health care decisions.

8. Nor has FDA provided any plausible interpretation of the statute that could support its approach. FDA’s assertion that laboratory testing services are devices just because the professionals performing those services _use_ devices is as unreasonable as calling a surgical procedure a “device” because the surgeon uses a scalpel, or calling a doctor’s physical examination a “device” because the doctor uses a stethoscope. The fact that a skilled professional may use physical tools, in addition to his or her professional expertise, training, and judgment, to perform a procedure does not mean that the procedure itself is a device.

9. Equally untenable is FDA’s contention that laboratory testing services are devices because they serve a similar function to in vitro diagnostic test kits, which FDA regulates as devices. An IVD test kit is a “device” because it is a packaged set of components manufactured and sold in interstate commerce as a single physical product, like an at-home COVID test. Such commercial test kits are fundamentally different from laboratory-developed tests, which are professional services performed by professional clinicians in a laboratory.

10. As noted above, the development and performance of laboratory-developed tests is regulated at the federal level under a separate statutory and regulatory framework—CLIA—that ensures the validity and reliability of laboratory tests and the training and qualifications of the skilled professionals who perform, supervise, and interpret those tests. Notably, when Congress enacted CLIA, it did not so much as hint that it had already granted FDA authority to regulate laboratory testing services as medical devices under the FDCA. If Congress had wanted to expand FDA’s authority so dramatically over an entire profession, it would have said so.

11. In its proposed rule, FDA initially contended that nearly all existing laboratory-developed tests would have to go through a burdensome
In the final rule, recognizing that its sweeping interpretation would be unworkable and have devastating consequences, FDA tried to rewrite the FDCA in the guise of dozens of pages of vague, non-binding “enforcement discretion policies” that are designed to mitigate (but not eliminate) those consequences. This “need to rewrite” the statute “should have alerted [FDA] that it had taken a wrong interpretive turn.” *Util. Air Regul. Grp.*, 573 U.S. at 328. “Agencies are not free to ‘adopt … unreasonable interpretations of statutory provisions and then edit other statutory provisions to mitigate the unreasonableness.’” *Id.* (quotation marks omitted).

12. The final rule repeatedly warns that FDA may change its enforcement discretion policy at any time and bring the hammer down on laboratories for unlawfully marketing existing tests. Even if FDA never takes that step, the rule creates enormous regulatory uncertainty for laboratories and places them in an impossible position: They must either (1) withdraw all their existing tests from the market (which FDA recognized would be devastating for patients and the public health); (2) incur massive costs to obtain FDA approval or clearance for their existing tests, which would divert resources from innovating and developing new tests and overwhelm FDA; or (3) continue serving patients by providing existing tests without FDA approval or clearance, even though FDA says that by doing so they are breaking the law and are subject to enforcement action at any time in the agency’s sole discretion.

13. In addition to casting a shadow over all existing tests, the final rule undermines innovation and threatens patient access to critical new diagnostic tests. FDA lacks the expertise or resources to timely and efficiently review and approve new and modified laboratory-developed testing services. Moreover, given the need for FDA approval or clearance, the rule will discourage laboratories from devoting scarce resources to research and development, which will impede the creation of new and improved tests for cancer, infectious disease, cardiovascular disease, and countless other diseases and conditions. Because many tests do not generate sufficient revenue to support the expense of seeking FDA approval or clearance, many important tests will never be developed—especially tests for rare diseases or that serve small patient populations, such as children or racial or ethnic minorities.

14. FDA has identified no genuine public-health justification for imposing these costs on laboratories and the physicians and patients who rely on them. The agency’s exercise of enforcement discretion for existing tests only underscores the lack of a valid public-health rationale for treating *any* laboratory-developed tests as medical devices.

15. For these reasons and those explained below, ACLA and HealthTrackRx seek declaratory and injunctive relief to vacate, set aside, and enjoin enforcement of the final rule.

### PARTIES

16. Plaintiff American Clinical Laboratory Association (“ACLA”) is a not-for-profit association with its principal place of business in Washington, D.C. ACLA is the national trade association representing leading laboratories that deliver essential diagnostic health information to patients and providers. ACLA’s members perform hundreds of millions of tests each year for patients across the country, and ACLA advocates for policies that expand access to the highest quality clinical laboratory services, improve patient outcomes, and advance the next generation of personalized care.

17. Plaintiff HealthTrackRx Indiana, Inc. is a corporation organized and existing under the laws of Indiana with its principal place of business in Denton, Texas. Plaintiff HealthTrackRX, Inc., is a corporation organized and existing under the laws of Texas with its principal place of business in Denton, Texas. HealthTrackRx is a leading national PCR-based infectious disease laboratory, providing services to over 10,000 clinicians nationwide. HealthTrackRx is also an ACLA member.

18. Defendant FDA, which has its principal office at 10903 New Hampshire Avenue, Silver Spring, Maryland 20993, is a federal agency headquartered in Maryland. It regulates drugs and medical devices under authority delegated by Congress and the Secretary of Health and Human Services.
19. Defendant U.S. Department of Health and Human Services, which has its principal office at 200 Independence Avenue, S.W., Washington, D.C. 20201, is a federal agency headquartered in the District of Columbia. It has authority over FDA.

20. Defendant Xavier Becerra is being sued in his official capacity as Secretary of Health and Human Services. As Secretary, Mr. Becerra has ultimate responsibility for the activities of the Department of Health and Human Services, including those actions complained of herein. Mr. Becerra maintains an office at 200 Independence Avenue, S.W., Washington, D.C. 20201.

21. Defendant Robert Califf, M.D., is being sued in his official capacity as Commissioner of Food and Drugs, FDA. As Commissioner, Dr. Califf is responsible for the activities of FDA, including those actions complained of herein. Dr. Califf maintains an office at 10903 New Hampshire Avenue, Silver Spring, Maryland 20993.

### JURISDICTION AND VENUE

22. This Court has original subject matter jurisdiction over this action pursuant to 28 U.S.C. § 1331 because it arises under the laws of the United States.

23. Plaintiffs have a right to bring this action pursuant to the Administrative Procedure Act (“APA”), 5 U.S.C. §§ 701–706, and the Declaratory Judgment Act, 28 U.S.C. § 2201.

24. Plaintiffs have standing because they or their members provide thousands of laboratory-developed tests that would be treated as devices under the final rule, making them direct objects of regulation under that rule. _See_ Dr. Reddy Decl. ¶¶ 6–7, 9–14, 24–37 (attached as Ex. A); Dr. Eisenberg Decl. ¶¶ 6, 8–10, 12, 18–20 (attached as Ex. B); Dr. Fesko Decl. ¶¶ 5–11 (attached as Ex. C); Dr. Genzen Decl. ¶¶ 12–13, 16–18, 22 (attached as Ex. D); Dr. Morice Decl. ¶¶ 19, 22, 25–30 (attached as Ex. E). Plaintiffs and their members also engage in research and development efforts to bring to market new and modified tests that would be treated as devices under the final rule. _See_ Dr. Reddy Decl. ¶¶ 20, 24, 32, 36–37; Dr. Eisenberg Decl. ¶¶ 6, 8, 9–10, 12, 18–19; Dr. Fesko Decl. ¶¶ 7, 9, 16, 19–21; Dr. Genzen Decl. ¶¶ 19–21, 25–42, 47–50, 58–59; Dr. Morice Decl. ¶¶ 9, 17–18, 22–26, 58.

25. There is currently an actual, justiciable controversy between the parties concerning whether FDA’s final rule is consistent with the requirements of the FDCA, 21 U.S.C. § 301 _et seq._, and the APA.

26. Venue is proper in this District pursuant to 28 U.S.C. § 1391(e) because this is a civil action in which the defendants are officers or agencies of the United States, plaintiff HealthTrackRx resides in this District, and no real property is involved in this action. _See_ Dr. Reddy Decl. ¶ 8.

### GENERAL ALLEGATIONS

#### A. Laboratory-developed tests are services carried out by highly skilled and trained laboratory professionals.

27. Laboratory-developed tests are procedures designed, developed, and performed by clinical laboratories certified to perform high-complexity testing to yield important clinical information about a patient that can be used to inform or guide patient care. _See_ Dr. Reddy Decl. ¶¶ 9–17, 23; Dr. Eisenberg Decl. ¶¶ 6–10, 15; Dr. Fesko Decl. ¶¶ 5–11; Dr. Genzen Decl. ¶¶ 11–13, 15–20, 23–26, 28; Dr. Morice Decl. ¶¶ 14. 27–28, 48–52.

28. Laboratories that develop and perform these tests are providing professional healthcare services; they are not acting as device manufacturers or distributing devices. _See_ Dr. Reddy Decl. ¶¶ 21–23; Dr. Eisenberg Decl. ¶¶ 14–15, 18; Dr. Fesko Decl. ¶¶ 14–15; Dr. Genzen Decl. ¶¶ 43–45; Dr. Morice Decl. ¶¶ 14, 48–56.

29. As an example, consider the steps associated with performing a mass spectrometry test offered by an ACLA member laboratory. Mass spectrometry is a chemical analysis technique with many uses, including helping manage hormonal disorders such as Cushing’s syndrome and measuring proteins with functions related to cancer and Alzheimer’s disease. After a physician orders the test, a blood specimen is obtained by a phlebotomist and sent to the laboratory. Laboratory staff then perform the

The following tasks:

a. *Pre-analytical steps.* The laboratory receives the blood sample and enters it into the laboratory information system. Laboratory staff then complete pre-analytical steps in accordance with the relevant standard operating procedures. That may include centrifuging the sample or aliquoting the sample into a separate tube for testing.

b. *Analytical steps.* A laboratory scientist prepares reagents, standards, and quality control materials, and retrieves the patient sample for testing. The scientist pipettes the applicable samples and reagents into a 96-well plate and extracts the analytes of interest using an automated liquid handling instrument. The scientist then enters relevant information into the instrument software and loads samples onto the testing system, which includes an automated sampler, liquid chromatography instrumentation, and a high-resolution mass spectrometer. When testing is complete, the scientist reviews the test both qualitatively and quantitatively (*e.g.,* reviewing chromatography and signal-to-noise ratios), including reviewing quality control to ensure the results are within parameters for acceptable performance. The scientist then reviews the patient results, uses software to determine the concentration of the analyte(s) being measured, and enters the results into the laboratory information system.

c. *Post-analytical steps.* A second laboratory scientist or lead scientist reviews the results to confirm they were accurately interpreted, quantitated, and entered into the laboratory information system. The reviewing scientist approves the results, sending them to the patient’s electronic medical record. The ordering physician then reviews the laboratory result produced by the test and uses it to inform patient care decisions.

30. This is a laboratory-developed testing service: a series of processes and tasks undertaken by trained laboratory professionals using instruments and other tools to derive information that may be useful to a treating physician. Under any reasonable interpretation, these procedures and the exercise of judgment that they require constitute a professional service, not a manufactured device. *See* Dr. Reddy Decl. ¶¶ 14, 22–23; Dr. Fesko Decl. ¶¶ 15, 20; Dr. Eisenberg Decl. ¶¶ 8, 14–15, Dr. Genzen Decl. ¶¶ 43–45; Dr. Morice Decl. ¶¶ 14, 48–56.

31. Laboratory-developed tests are a vital part of the U.S. healthcare system and make significant contributions to patient care. They have often been responsible for scientific innovations and breakthroughs—for example, testing for the BRCA1/BRCA2 genetic mutations that indicate susceptibility to breast and ovarian cancer—that have become part of the standard of care (and in some cases, have been incorporated into FDA-cleared or approved IVD test kits). They also play a critical role in responding to public health threats from rare or emerging pathogens and new synthetic drugs, such as fentanyl analogs.

32. Many important diagnostic tests are available *only* as laboratory-developed testing services because no FDA-cleared or approved test kit exists for a particular disease, condition, or patient population. And even when an approved or cleared test kit is available, laboratory-developed tests often perform better and are preferred by physicians. Unlike medical devices, which must always take their approved form, laboratory-developed tests can be updated and customized (under the supervision of a CLIA-qualified laboratory director) to take account of the latest scientific developments and the needs of particular patients and clinicians. *See* Dr. Reddy Decl. ¶¶ 17, 36–37; Dr. Genzen Decl. ¶¶ 26–27.

#### B. FDA’s statutory authority to regulate medical devices does not extend to professional services.

33. The FDCA was originally enacted by Congress in 1938. It authorized FDA to regulate “foods,” “drugs,” “devices,” and “cosmetics,” all of which were physical *products* that were mass-manufactured and commercially distributed. *See* Federal Food, Drug, and Cosmetic Act of 1938, Pub. L. No. 75-717, § 201(h), 52 Stat. 1040, 1041 (“The term ‘device’ … means instruments, apparatus, and contrivances, including their components, parts, and accessories, intended (1) for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals; or (2) to affect the structure or any function of the body of man or other animals.”); *see also id.* §§ 201(f), (g),
34. Congress greatly expanded FDA’s authority over devices in the Medical Device Amendments of 1976 (“MDA”), Pub. L. No. 94-295, 90 Stat. 539, which amended the FDCA to “impose[] a regime of detailed federal oversight” on medical devices. *Riegel v. Medtronic, Inc.*, 552 U.S. 312, 316 (2008).

35. The FDCA’s Medical Device Amendments classify medical devices into three categories based on the level of risk they present. Class I devices are subject only to “general controls” such as labeling requirements. Class II devices are also subject to “special controls” such as performance standards and postmarket surveillance measures. Class III devices are subject to “a rigorous regime of premarket approval.” *Id.* at 316–17 (quotation marks omitted); see 21 U.S.C. § 360c(a)(1).

36. There are a few statutory exceptions to these general rules. Class III devices that were marketed before the statute’s effective date in 1976 were allowed to remain on the market unless and until FDA promulgates a regulation requiring the submission of premarket approval applications. *Riegel*, 55 U.S. at 316–17; see 21 U.S.C. §§ 360c(f)(1), 360e(b)(1). Moreover, a new Class III device need not go undergo full premarket approval if FDA finds that the new device is “substantially equivalent” to a grandfathered device. 21 U.S.C. § 360c(f)(1)(A).

37. While the three device categories differ by level of risk, they all comprise tangible, physical products. For example, Class I devices include “elastic bandages and examination gloves,” Class II devices include “powered wheelchairs and surgical drapes,” and Class III devices include “replacement heart valves, implanted cerebella stimulators, and pacemaker pulse generators.” *Riegel*, 552 U.S. at 316–17.

38. The statutory definition of “device” makes clear that FDA’s regulatory jurisdiction under the FDCA is limited to physical products and does not encompass professional services. The statute provides:

   - The term “device” … means an *instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article, including any component, part, or accessory, which is—
   
   - (A) recognized in the official National Formulary, or the United States Pharmacopeia, or any supplement to them,
   
   - (B) intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals, or
   
   - (C) intended to affect the structure or any function of the body of man or other animals, and
   
   - which does not achieve its primary intended purposes through chemical action within or on the body of man or other animals and which is not dependent upon being metabolized for the achievement of its primary intended purposes.

   21 U.S.C. § 321(h)(1) (emphasis added).

39. All of the terms used in the FDCA’s definition of “device”—“instrument,” “apparatus,” “implement,” “machine,” “contrivance,” “implant,” and “in vitro reagent”—refer to tangible, physical objects. Moreover, the statute uses the term “article” as a catch-all to encompass all “devices,” and the plain meaning of “article” does not include intangible services. An “article” is a “particular material thing, esp. one belonging to a specified class; a commodity; an item of goods or property.” *Article*, Oxford English Dictionary (2023), [https://www.oed.com/search/dictionary/?scope=Entries&q=article](https://www.oed.com/search/dictionary/?scope=Entries&q=article). Consistent with this common definition, courts have consistently construed the term “article” to mean a “material thing” or a “tangible item.” See, e.g., *ClearCorrect Operating, LLC v. ITC*, 810 F.3d 1283, 1290–94 (Fed. Cir. 2015) (construing the term “articles” in the Tariff Act), reh’g en banc denied, 819 F.3d 1334 (Fed. Cir. 2016) (mem.).

40. Other provisions of the FDCA confirm that a “device” is a physical product, not a service. Several key provisions are triggered only when a device is shipped or received in interstate commerce, commercially distributed, or held for sale—actions that, in ordinary parlance, can be performed on a tangible article but not on an intangible professional service.

41. For example, section 510(k) of the FDCA requires a device manufacturer to file a premarket notification report with FDA at least 90 days before “the introduction or delivery for introduction into interstate commerce for commercial distribution of a device.” 21 U.S.C. § 360(k). “Commercial

distribution” is defined in an FDA regulation to mean “any distribution of a device intended for human use which is held or offered for sale.” 21 C.F.R. § 807.3(b). Similarly, section 301(k) prohibits various acts “with respect to” a device “if such act is done while such article is held for sale (whether or not the first sale) after shipment in interstate commerce and results in such article being adulterated or misbranded.” 21 U.S.C. § 331(k). None of these provisions can reasonably be applied to an intangible professional service performed within a laboratory.

42. Still other provisions of the FDCA discuss devices in ways that make sense only if applied to physical products. For example, an application for premarket approval for a device must include, among other things: (i) a description of “the components, ingredients, and properties” of the device; (ii) a description of the methods, facilities and controls used in “the manufacture, processing, and, when relevant, packing and installation” of the device; and (iii) “such samples of such device and of components thereof as the Secretary may reasonably require” or “information concerning the location of one or more such devices readily available for examination and testing.” 21 U.S.C. § 360e(c)(1)(B), (C), (E). Intangible professional services do not have components, ingredients, or properties; the services are not manufactured, processed, packed, or installed (even if providing them might entail using a manufactured product or products); and samples of a service cannot be submitted to FDA or made readily available for inspection. *See* Dr. Reddy Decl. ¶¶ 22–23; Dr. Eisenberg Decl. ¶¶ 14–15; Dr. Fesko Decl. ¶ 15; Dr. Genzen Decl. ¶ 45; Dr. Morice Decl. ¶¶ 14, 48–56.

43. So too, the statute provides that in certain circumstances FDA may order the manufacturer, importer, or distributor of a device to “repair the device” or “replace the device with a like or equivalent device.” 21 U.S.C. § 360h(b). Unlike physical products, intangible professional services cannot be repaired or replaced.

44. Several of FDA’s promulgated regulations for devices can similarly be understood only as applied to a manufactured product. For example, an FDA regulation requires the “label of every medical device” and “[e]very device package” to bear a unique device identifier. 21 C.F.R. § 801.20(a). “Label” is defined as “a display of written, printed, or graphic matter upon the immediate container of any article,” and “device package” is defined as “a package that contains a fixed quantity of a particular version or model of a device.” *Id.* § 801.3 (incorporating 21 U.S.C. § 321(k)). These requirements make sense in the context of manufactured devices, where the primary and expected means of communication between the manufacturer and any purchaser is through a standardized label. In sharp contrast, a laboratory scientist’s performance of the tasks comprising a laboratory-developed test cannot be “labeled” or “packaged” in compliance with these regulations. Nor can those professional services be summarized in a standardized label; instead, clinical laboratory services entail the exercise of professional judgment when interpreting testing results and often a consultation process between professional laboratory clinicians and doctors and other healthcare providers. *See* Dr. Eisenberg Decl. ¶ 14; Dr. Fesko Decl. ¶¶ 9, 15; Dr. Genzen Decl. ¶ 19; Dr. Morice Decl. ¶¶ 27, 48–56.

45. Viewed collectively, these provisions confirm what the statutory definition of “device” makes clear: A “device” under the FDCA is a physical product or manufactured good, not an intangible professional service.

#### C. Congress created a separate and distinct framework for regulating laboratory testing services.

46. Congress created a separate statutory and regulatory framework to regulate laboratory testing services: the Clinical Laboratories Improvement Act of 1967, Pub. L. No. 90-174, § 5, 81 Stat. 533, 536, which was significantly expanded by the Clinical Laboratory Improvement Amendments of 1988, Pub. L. No. 100-578, 102 Stat. 2903, codified at 42 U.S.C. § 263a. This statutory framework is commonly referred to as “CLIA.”

47. Congress’s enactment and expansion of CLIA in 1967 and 1988 confirms that it did not understand the Medical Device Amendments in 1976 as authorizing FDA to regulate laboratory testing services as medical devices.

48. CLIA establishes a framework for the regulation of laboratories
49. CLIA and its implementing regulations reflect that performing and interpreting laboratory tests requires significant scientific and technical knowledge, training, experience, and judgment, and is fundamentally different from manufacturing physical devices. *See* Dr. Reddy Decl. ¶ 22; Dr. Genzen Decl. ¶ 27; Dr. Morice Decl. ¶¶ 12, 49, 52, 56–58, 62; Dr. Eisenberg Decl. ¶¶ 15–16; Dr. Fesko Decl. ¶ 15.

50. Under CLIA, all laboratories that perform clinical tests on human specimens must be certified by CMS or accredited through certain CMS-approved accreditation organizations. 42 U.S.C. § 263a(b); *see* Dr. Reddy Decl. ¶ 12; Dr. Genzen Decl. ¶ 15. Both CMS and accreditation organizations issue standards to assure that laboratories’ performance is “consistent” and their tests are “valid and reliable,” including quality-control standards and standards for the qualifications of the personnel directing, supervising, and performing the tests. 42 U.S.C. § 263a(f)(1). The standards must take into account, among other things, the type of tests performed, the “degree of independent judgment involved,” “the amount of interpretation involved,” “the difficulty of the calculations involved,” and “the type of training required.” § 263a(f)(2).

51. The College of American Pathologists is the most prominent example of a CMS-approved accreditation organization, and to be accredited by that organization, a laboratory must be inspected initially and then every two years and must demonstrate that it complies with approximately 3,000 specific requirements, including validation of any clinical claims made by the laboratory for any laboratory-developed testing service. *See* College of American Pathologists, *CAP Advances Quality in Laboratory Medicine and Safeguards Patient Testing with Annual Release of Laboratory Accreditation Program Checklists* (Sept. 23, 2021), https://newsroom.cap.org/cap-in-the-news/cap-advances-quality-in-laboratory-medicine-and-safeguards-patient-testing-with-annual-release-of-la/s/88c2ad6c-72b4-4641-aaa7-3edb4954aac8.

52. The CLIA regulations ensure that laboratory testing services are performed only by highly skilled and trained laboratory professionals. Laboratories that perform high-complexity tests must be overseen by a laboratory director, who must either be a licensed physician or hold a doctoral degree in a chemical, physical, biological, or clinical laboratory science. 42 C.F.R. § 493.1443. The laboratory director is responsible for ensuring that the laboratory’s test methodologies are “capab[le] of providing the quality of results required for patient care,” that “[l]aboratory personnel are performing the test methods as required for accurate and reliable results,” and that “consultation is available to the laboratory’s clients on matters relating to the quality of the test results reported and their interpretation concerning specific patient conditions.” *Id.* § 493.1407(e)(3), (9).

53. The laboratory must also have a technical supervisor with appropriate training or experience for the types of tests performed by the laboratory, *id.* § 493.1449, and a clinical consultant qualified to “consult with and render opinions to the laboratory’s clients concerning the diagnosis, treatment, and management of patient care,” *id.* § 493.1455. The clinical consultant is responsible for providing “consultation regarding the appropriateness of the testing ordered and interpretation of test results.” *Id.* § 493.1457. And all laboratory personnel who perform high-complexity tests must either be licensed physicians or have appropriate training and experience in laboratory science or medical technology. *Id.* § 493.1489.

54. Under CLIA, laboratory testing services are subject to strict quality controls. When a laboratory introduces a new diagnostic test “not subject to FDA clearance or approval” (or when it modifies an FDA-cleared or approved IVD test kit purchased from a device manufacturer), it must, “before reporting patient test results,” establish “performance specifications” for the test—including specifications for accuracy, precision, analytical sensitivity, and other characteristics “required for test performance.” *Id.* § 493.1253(b)(2). Performance of the test is also subject to the laboratory’s CLIA-mandated
quality control system, which requires, among other things, establishment and performance of calibration and control procedures; maintenance and function checks for equipment, instruments and test systems; and ongoing quality monitoring. *Id.* § 493.1200–1299.

55. CLIA further requires laboratories to demonstrate proficiency in their tests multiple times a year. 42 U.S.C. § 263a(f)(3). For many of their tests, laboratories must enroll and participate in approved proficiency testing programs, which serve as external quality control checks for every test the laboratory performs. Proficiency testing requires that the laboratory test blinded samples according to its typical procedures and report the results back to the testing program for evaluation. 42 C.F.R. § 493.801. A laboratory that fails to achieve satisfactory proficiency scores may face sanctions, including suspension, limitation, or revocation of its CLIA certificate. *Id.* §§ 493.803(b), 493.1806.

56. CLIA-certified laboratories are subject to inspections, by the Department of Health and Human Services, state agencies, and authorized accrediting bodies. *See, e.g.*, 42 U.S.C. § 263a(g); 42 C.F.R. Part 493, Subpart Q.

57. Although CLIA was enacted nine years before the Medical Device Amendments and significantly expanded twelve years after those Amendments, neither CLIA nor its legislative history acknowledges any authority of FDA to regulate laboratory testing services as medical devices.

58. The Senate Report on the 1967 bill addressed concerns about possible overlap between regulation of clinical laboratories under CLIA and under the Medicare statute, but it did not mention any role for FDA. *See* S. Rep. No. 90-724 (1967), *reprinted in* 1967 U.S.C.C.A.N. 2076, 2084.

59. Likewise, the House Report on the 1988 bill described “[t]he Current Regulatory System” as involving federal regulation of laboratories “under two programs”—the Clinical Laboratory Improvement Act of 1967 and the Medicare statute—and did not so much as mention regulation by FDA. H.R. Rep. No. 100-899, at 11 (1988). The Report also states that the purpose of CLIA was to replace a “confusing” system where laboratories were regulated under “two separate and distinct statutes” with a single “unified regulatory mechanism”—a purpose that is at odds with subjecting laboratory testing services to regulation under both CLIA and the FDCA. *Id.* at 12.

60. In short, there is no indication that Congress, when it enacted CLIA, believed that clinical laboratories’ provision of testing services was already subject to regulation under the FDCA.

#### D. FDA has never broadly regulated laboratory testing services as medical devices.

61. In the nearly half-century since Congress enacted the Medical Device Amendments of 1976—not to mention the 86 years since Congress first gave FDA authority over medical devices in 1938—FDA has never before acted to broadly regulate laboratory-developed tests as “devices” under the FDCA. This lengthy history confirms that FDA lacks statutory authority to do so now. *See Util. Air Regul. Grp.*, 573 U.S. at 324 (an agency’s reliance on “a long-extant statute” to bring about a dramatic “expansion in [its] regulatory authority” should be met with “skepticism”); *Christopher v. SmithKline Beecham Corp.*, 567 U.S. 142, 157–58 (2012) (when an agency has responded to an industry’s “decades-long practice” with a “lengthy period of conspicuous inaction,” the likely explanation is that the industry practice was lawful).

##### 1. FDA’s Many Years of Silence

62. In the first 16 years following Congress’s enactment of the Medical Device Amendments to the FDCA, from 1976 through 1992, FDA did not claim any authority to regulate laboratory testing services as “devices.” Nor had FDA ever claimed such authority under the FDCA as enacted in 1938, even though that statute contained a similar definition of “device.”

63. Clinical laboratories thus reasonably understood that their services were not subject to regulation under the FDCA. *See Dr. Reddy Decl.* ¶ 24; Dr. Eisenberg Decl. ¶¶ 15–17; Dr. Fesko Decl. ¶ 14; Dr. Genzen Decl. ¶ 46. Congress acted on the same understanding when it enacted CLIA to create a single, unified, comprehensive system for the federal regulation of laboratory testing.
64. The first time FDA suggested that it might possess authority to regulate laboratory-developed tests as devices was in 1992—16 years after Congress enacted the Medical Device Amendments and 54 years after it first enacted the FDCA. But when that suggestion drew immediate and strenuous objections, FDA essentially backed down and announced a “policy” of not exercising jurisdiction over laboratory testing services that it adhered to for the next 30 years.

65. In its final rule, FDA cites a 1973 rulemaking as purported evidence that FDA treated tests as devices before Congress enacted the Medical Device Amendments. 89 Fed. Reg. 37,286, 37,328 (May 6, 2024) (citing 38 Fed. Reg. 7096 (Mar. 15, 1973)). That rulemaking defined “[i]n vitro diagnostic products”—not services—as diagnostic “reagents, instruments and systems.” 38 Fed. Reg. at 7098. (emphasis added). Although FDA now suggests that the term “systems” was intended to include testing services, the context makes clear that the only “systems” subject to the rule were finished products, not laboratory tests. For example, the 1973 rule included labeling provisions requiring that certain information be affixed to the “retail package” of the “article.” 38 Fed. Reg. at 7098. The 1973 rule thus confirms that FDA originally sought to regulate only physical products, not professional laboratory procedures or techniques. Moreover, when Congress amended the definition of “device” in 1976, it did not include the term “system” or even the term “in vitro product,” but only the narrower term “in vitro reagent.”

##### 2. FDA’s Never-Finalized 1992 Guidance

66. Sixteen years following the enactment of the Medical Device Amendments, in 1992, in a draft Compliance Policy Guide, FDA made the novel claim that it could regulate laboratory-developed tests as medical devices. This claim was made in passing in a document that generally addressed the marketing and distribution of IVD test kits. In a brief aside, FDA stated that “laboratories have been manufacturing ‘home brew’ products, either from products already on the market, or from components, and utilizing these unapproved products for diagnostic purposes,” and added that “[t]hese products are subject to the same regulatory requirements as any unapproved medical device.” FDA, Draft Compliance Policy Guide: Commercialization of Unapproved In Vitro Diagnostic Devices Labeled for Research and Investigation at 4 (Aug. 3, 1992).

67. The laboratory profession immediately objected to this abrupt and unexplained assertion of jurisdiction over professional laboratory testing services. For example, a law firm that represented clinical laboratories filed a citizen petition asking FDA not to assert jurisdiction over laboratories’ “in-house assays” and noting, among other concerns, that FDA’s authority over medical devices “does not extend to test methods, protocols, or services.” Citizen Pet. at 9, Hyman, Phelps & McNamara, P.C., Docket No. FDA-92-P-0405 (Oct. 22, 1992) (“1992 Citizen Petition”).

68. FDA did not immediately respond to the 1992 Citizen Petition. But following controversy over the 1992 draft guidance, FDA did not finalize that guidance or attempt to actively regulate laboratory testing services. Instead, FDA sought to calm the waters by announcing that it did “not intend to routinely exercise its authority over home-brew tests.” *IVD Policy Will Not Include Exemptions for “Standard-of-Care” Tests*, The Gray Sheet (Oct. 11, 1993).

##### 3. FDA’s Sporadic Claims of Authority in the 1990s and 2000s

69. FDA next asserted that it had jurisdiction over laboratory testing services in the non-binding preamble to a 1996 proposed rule regarding device classification levels for certain “active ingredients used in preparing in-house developed [laboratory] tests.” 61 Fed. Reg. 10,484, 10,485 (Mar. 14, 1996). In the preamble, FDA noted that it had previously regulated as devices only (i) “diagnostic tests that are traditionally manufactured and commercially marketed as finished products” (i.e., test kits), and (ii) tangible articles used as test “ingredients,” such as “laboratory apparatus” and “chemicals or antibodies,” that laboratories “purchase from biological or chemical suppliers.” *Id.* at 10,484.

70. In response, ACLA and other stakeholders filed comments.
challenging FDA’s assumption that it had authority to regulate laboratory testing services as medical devices.

71. In the preamble to the final rule, FDA stated that it “believes that clinical laboratories that develop such [in-house] tests are acting as manufacturers of medical devices.” 62 Fed. Reg. at 62,249. FDA recognized, however, that “the use of in-house developed tests has contributed to enhanced standards of medical care in many circumstances and that significant regulatory changes to this area could have negative effects on the public health.” *Id.* FDA therefore stated that it would continue to focus on regulating “ingredients … that move in commerce” and other tangible articles, not laboratory testing services. *Id.; see id.* at 62,250 (concluding that “regulation of all in-house developed tests” was not “appropriate at this time”).

72. Over the next 12 years, FDA continued to assert periodically and in draft non-binding guidance that it had statutory authority to regulate professional laboratory testing services as manufactured devices but was choosing not to. But FDA never took any final regulatory action backing up its non-binding statements. And the agency’s consistent policy of *not* treating laboratory-developed tests as devices made these occasional claims nothing but empty posturing.

##### 4. FDA’s Never-Finalized 2014 Guidance

73. FDA’s first real suggestion that it might put its posturing into practice came in 2010, when the agency announced its intention to “reconsider its policy of enforcement discretion” with respect to laboratory-developed tests. 75 Fed. Reg. 34,463–64 (June 17, 2010). FDA said it intended to “develop a draft oversight framework for public comment” that would “phase in … over time based on the level of risk” presented by various tests. *Id.*

74. In response, ACLA submitted comments reiterating that “laboratories are providers of testing services; they are not medical device manufacturers.” ACLA Supp. Comments on Oversight of Lab’y Developed Tests at 4, Docket No. FDA-2010-N-0274 (Sept. 15, 2010). ACLA explained that, while it might be appropriate for FDA to regulate as devices “the products used by clinical laboratories to perform tests,” including “commercially distributed *in vitro* diagnostic test kits,” FDA should not and cannot recklessly impose device regulation on the provision of “laboratory services.” *Id.* at 4, 6 (emphases added).

75. Two years later, with FDA still not having published any proposed oversight framework, Congress prohibited the agency from issuing “any draft or final guidance on the regulation of laboratory-developed tests” for five years unless the details of FDA’s plan were disclosed to the relevant congressional committees at least 60 days prior to such issuance. Pub. L. No. 112-144, § 1143(a), 126 Stat. 993, 1130 (2012).

76. A year after that, ACLA submitted a citizen petition asking FDA to acknowledge that laboratory testing services are not devices. ACLA’s petition explained that text and legislative history make clear that “devices” are tangible articles and do not include services or procedures. *See* ACLA Citizen Pet. at 7–9, Docket No. FDA-2013-P-0667 (June 4, 2013). The petition acknowledged that performing a laboratory-developed test “might involve use of” physical devices, such as “reagents,” “laboratory equipment,” or “IVD test kits.” *Id.* at 1, 8. But it stressed that a clinical service does not become subject to regulation as a device “simply because the service involves the use of tangible articles which may be subject to FDA regulation.” *Id.* at 8–9. Otherwise, it noted, “every surgical procedure or physical examination that is performed on a patient using tangible devices” would itself be a “device.” *Id.*

77. FDA denied ACLA’s citizen petition in 2014 and asserted that laboratory-developed tests “are ‘devices’ as defined in the FDCA.” FDA Denial of ACLA Citizen Pet. at 3, Docket No. FDA-2013-P-0667 (July 31, 2014). Eliding the distinction between professional services and physical products, FDA claimed that laboratory testing services are devices because they *make use of* various physical “articles,” such as “reagents,” “instruments,” and “equipment”—even though the testing services themselves are plainly not “articles.” *Id.* at 3–5, 23.

78. On the same day it denied ACLA’s citizen petition, FDA made similar assertions in response to two other citizen petitions, which had been
pending since 2006 and 2008, respectively. *See* FDA Denial of Wash. Legal Found. Citizen Pet. at 3–4, Docket No. FDA-2006-P-0149 (July 31, 2014); FDA Denial of Genentech, Inc., Citizen Pet. at 5–6, Docket No. FDA-2008-P-0638 (July 31, 2014). Also on that day, FDA notified Congress of its intent to issue draft guidance documents regarding laboratory-developed tests.

79. On October 3, 2014, FDA released the draft guidance documents it had promised in 2010, proposing to phase in new regulation of laboratory-developed tests as devices over a nine-year period. In announcing the draft guidance documents, FDA described laboratory-developed tests as “a subset of in vitro diagnostic devices that are intended for clinical use and designed, manufactured, and used within a single laboratory.” 79 Fed. Reg. 59,776, 59,777 (Oct. 3, 2014); *see also* 79 Fed. Reg. 59,779, 59,780 (Oct. 3, 2014).

80. ACLA submitted comments on the draft guidance documents. Among other points, ACLA’s comments explained once again that “a ‘device’ is a physical article or product” and “[l]aboratory-developed testing services are processes and methodologies that are qualitatively and categorically different from the tangible goods that FDA may regulate as ‘devices.’” ACLA Comments on Oversight of Laboratory Developed Tests and Reporting at 5, Docket Nos. FDA-2011-D-0357 & -0360 (Feb. 2, 2015). ACLA also reiterated that “[l]aboratory-developed testing services do not become medical devices merely because they sometimes utilize other medical devices,” such as reagents and laboratory equipment. *Id.* at 6. As ACLA noted, “every time a radiologist reads an x-ray, she is providing a service that depends on a medical device—the x-ray machine. However, the radiologist is rendering a service and is not subject to regulation under the FDCA” as a device manufacturer. *Id.*

81. On November 18, 2016, FDA backed down, announcing that it would not finalize the 2014 draft guidance documents. In a white paper published in January 2017, FDA noted that it had made this decision “to allow for further public discussion on an appropriate oversight approach, and to give our congressional authorizing committees the opportunity to develop a legislative solution.” FDA, Discussion Paper on Laboratory Developed Tests (LDTs) at 1 (Jan. 13, 2017).

##### 5. The 2020 Charrow Memo Questions FDA’s Authority to Regulate Laboratory Testing Services

82. In 2020, Robert Charrow, then-General Counsel of the Department of Health and Human Services, issued a memorandum regarding “Federal Authority to Regulate Laboratory Developed Tests.”

83. The Charrow memorandum is significant because it addresses and undermines key premises upon which FDA now relies. For example, it recognized that laboratory-developed tests “were never mentioned in the [Medical Device Amendments], in the House Report accompanying it, or during the floor debates.” Mem. from Robert Charrow, Gen. Counsel, to Stephen Hahn, M.D., Comm’r of Food & Drugs, at 3 (June 22, 2020) (attached as Ex. F). It further noted that Congress’s enactment of CLIA in 1988, and the Secretary’s issuance of “comprehensive rules governing clinical laboratories” pursuant to CLIA, “appeared to have occupied the field for regulating [laboratory-developed tests].” *Id.* at 3–4.

84. Contrary to FDA’s claims that it regarded laboratory-developed tests to be devices as far back as 1976, the Charrow memorandum acknowledges that FDA had “first suggested that [laboratory-developed tests] are subject to its jurisdiction” in 1992—16 years after the Medical Device Amendments were enacted—and that from 1992 until 2014, “FDA did little to regulate LDTs.” *Id.* at 4. Moreover, although FDA had proposed altering that status quo in 2014 when it published the draft guidance documents, it had subsequently declined to finalize those documents. *Id.* at 5.

85. Acknowledging the argument that laboratory-developed tests “are not physical embodiments, e.g. ‘contraptions,’ but rather are processes or services, and therefore not devices,” the Charrow memorandum observed that while “in vitro reagents are devices, … that does not necessarily lead to the conclusion that [laboratory-developed tests] fall within FDA’s jurisdiction.” *Id.* at 6. The memorandum explained that Congress’s enactments do not “lead[] to the conclusion that [laboratory-developed tests] are devices” and “the Secretary has issued rules implementing Medicare and CLIA that strongly
suggest that [laboratory-developed tests] are not devices and not within FDA’s jurisdiction.” *Id.* at 14.

86. The memorandum also recognized that laboratory-developed tests are not “goods or commodities” but rather “clinical laboratory services,” and are treated as such by Medicare. *Id.* at 10. The memorandum analogized “the development and use of” laboratory-developed tests to a “doctor’s development and use of a medical procedure.” *Id.*

##### 6. Congress Chooses Not to Enact Legislation

87. Reflecting the lack of statutory authority for FDA to regulate laboratory testing services under the FDCA, Congress considered legislative proposals that would have given FDA such authority. On March 5, 2020, the VALID Act was introduced in both houses of Congress. *See* Verifying Accurate Leading-edge IVCT Development Act of 2020, H.R. 6102, 116th Cong. (2020) (companion bill S.3404). The Act would have created a new regulatory pathway, separate from both drugs and devices, for FDA premarket review and regulation of “in vitro clinical tests,” including laboratory-developed tests. *See id.* § 2(a).

88. Commentators noted that “[e]arlier versions of the proposed VALID Act had been circulating in Washington for several years” following FDA’s “abortive” attempt to regulate laboratory-developed tests in 2014, which FDA had “abandoned” in 2016 “amid questions about [its] jurisdiction to regulate laboratory services.” Barbara J. Evans & Ellen Wright Clayton, *Deadly Delay: The FDA’s Role in America’s COVID-Testing Debacle*, 130 Yale L.J. Forum 78, 83–84 (2020)).

89. The VALID Act did not pass during the 116th Congress. It was reintroduced in the 117th Congress, where it again failed to pass. *See* VALID Act of 2021, H.R. 4128, 117th Cong. (2021) (companion bill S.2209). It was introduced again in 118th Congress, and once again it failed to become law. *See* VALID Act of 2023, H.R. 2369, 118th Cong. (2023) (companion bill S.2496).

#### E. FDA now seeks for the first time to classify virtually all laboratory testing services as medical devices.

##### 1. The 2023 Proposed Rule

90. With no congressional authorization forthcoming, FDA once again announced its intent to move forward with regulating virtually all laboratory-developed testing services as medical devices. FDA published its proposed rule in October 2023. *See* 88 Fed. Reg. 68,006 (Oct. 3, 2023).

91. In the proposed rule, FDA stated that it would amend a regulatory definition of “in vitro diagnostic products” to add the underlined language:

- [In vitro diagnostic products] are defined as “those reagents, instruments, and systems intended for use in the diagnosis of disease or other conditions, including a determination of the state of health, in order to cure, mitigate, treat, or prevent disease or its sequelae. Such products are intended for use in the collection, preparation, and examination of specimens taken from the human body.” ... These products are devices as defined in section 201(h)(1) of the Federal Food, Drug, and Cosmetic Act (the act) and may also be biological products subject to section 351 of the Public Health Service Act, *including when the manufacturer of these products is a laboratory.*

*See id.* at 68,017, 68,031 (proposed amendment to 21 C.F.R. § 809.3(a)). In the preamble, FDA made clear that it intended this amendment to signify that all laboratory testing services are “devices” and that whenever a laboratory scientist or technician performs a clinical laboratory test, he or she is engaged in “manufacturing” a “device.” *Id.* at 68,007–09, 68,017–19.

92. In the preamble to the proposed rule, FDA also stated its intent to “phase out its general enforcement discretion approach” so that most laboratory-developed tests would “fall under the same enforcement approach as other” medical devices within a few years. *Id.* at 68,007.

93. FDA also recognized that its rule would impose vast costs on the clinical laboratory sector (although again, its projections were low). It estimated that the up-front cost of preparing and submitting premarket approval applications, premarket notifications, and de novo classification requests *for existing tests alone* would exceed $35 billion and could be as high as $113 billion. FDA, Docket No. FDA-2023-N-2177, Laboratory Developed Tests Proposed Rule: Preliminary Regulatory Impact Analysis at 85 (Oct. 3, 2023). It also estimated that going forward, the annual compliance for affected laboratories would be more than $4 billion and could be as high as $14 billion.
94. FDA acknowledged that these costs would cause some existing tests to “come off the market” because laboratories would not be able to justify the high costs of obtaining the necessary approval or clearance for those tests. 88 Fed. Reg. at 68,014.

##### 2. ACLA’s Comments on the Proposed Rule

95. FDA received more than 6,000 comments on its proposed rule, a volume of public input that reflects the radical and transformative nature of FDA’s proposal.

96. ACLA submitted its comments on December 4, 2023. Among other critical points, ACLA’s comments explained, yet again, that FDA does not have legal authority to regulate laboratory-developed tests as devices—including because “devices” under the FDCA are physical products that are sold and distributed by manufacturers, whereas laboratory-developed tests are services offered by trained laboratory professionals that are regulated under CLIA’s distinct statutory and regulatory framework. _See_ ACLA Comments on Proposed Rule “Medical Devices; Laboratory Developed Tests” at 59–71, Docket No. FDA-2023-N-2177 (Dec. 4, 2023) (attached as Ex. G). ACLA’s comments also explained that FDA’s unlawful assertion of jurisdiction over clinical laboratory services would seriously harm patients by undermining diagnostic and medical innovation and limiting or eliminating access to critical tests. _Id._ at 7–18.

97. In addition, ACLA’s comments demonstrated that FDA had vastly underestimated the costs of regulating laboratory-developed tests as devices—including by underestimating the number of affected laboratories, the number of currently available tests that would require costly and time-consuming premarket submissions, the cost of preparing those submissions, and the cost of complying with other device regulations. _Id._ at 46–54. And conversely, ACLA showed that FDA had vastly _overestimated_ the benefits of its novel regulatory approach, including by using cherry-picked, anecdotal, and unverified “evidence” to paint an unfairly disparaging picture of laboratory testing services, while ignoring studies showing that laboratory-developed tests perform at least as well as FDA-approved or cleared IVD test kits. _Id._ at 36–46, 54–59.

##### 3. The 2024 Final Rule

98. FDA published the final rule on May 6, 2024.

99. As contemplated in the proposed rule, FDA amended the regulatory definition of “in vitro diagnostic products” in 21 C.F.R. § 809.3(a) to add the language, “including when the manufacturer of these products is a laboratory.” 89 Fed. Reg. at 37,286–87. And, as in the proposed rule, FDA made clear that it considers the provision of laboratory-based testing services a form of device “manufacturing.” _See id._ at 37,286–87, 37,289, 37,293, 37,328–32, 37,344.

100. In a major departure from the proposed rule, however, the preamble to the final rule states that FDA intends—for now and until it changes its mind—to exercise “enforcement discretion” for some or all requirements with respect to broad categories of laboratory-developed tests, including nearly all existing tests. _Id._ at 37,294–95. These non-binding “enforcement discretion policies” include the following:

- FDA will generally not enforce premarket review and Quality System (“QS”) requirements for existing tests that are not modified or that are “modified in certain limited ways.”
- FDA will generally not enforce premarket review requirements for tests approved by the New York State Department of Health’s Clinical Laboratory Evaluation Program.
- FDA will generally not enforce premarket review and QS requirements (except certain recordkeeping requirements) for tests “manufactured and performed” by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system.
- FDA will generally not enforce premarket review and QS requirements (except certain recordkeeping requirements) for non-molecular antisera tests for rare red blood cell antigens where such tests are “manufactured and performed” in blood establishments, including transfusion services and immunohematology laboratories, and where there is no alternative available to meet the patient’s need for a compatible blood transfusion.
- FDA will generally not enforce any requirements for “1976-Type LDTs” (tests with certain characteristics that FDA says were common among laboratory-developed tests offered in 1976).

- FDA will generally not enforce any requirements for Human Leukocyte Antigen tests that meet certain specified characteristics.
- FDA will generally not enforce any requirements for tests intended solely for forensic (law enforcement) purposes.
- FDA will generally not enforce any requirements for tests “manufactured and performed” within the Department of Defense or the Veterans Health Administration.

*Id.*

101. These extensive carveouts are necessary, the final rule acknowledges, because “expecting compliance with full [quality system] and premarket review requirements for all currently marketed” laboratory-developed tests “could lead to the loss of access to safe and effective” tests “on which patients currently rely.” *Id.* at 37,293; see Dr. Reddy. Decl. ¶ 6; Dr. Genzen Decl. ¶¶ 47, 56; Dr. Morice Decl. ¶¶ 15–16, 59–65.

102. In other words, faced with the impracticality and catastrophic impact of its novel interpretation of the law, FDA did not take that unworkability as a hint that its interpretation might be mistaken. Instead, to try to contain the damage, FDA effectively used a non-binding regulatory preamble to write a new statute on the fly, under the guise of “enforcement discretion policies.”

103. These broad carveouts undermine FDA’s legal rationale for the rule, which classifies all laboratory-developed tests as manufactured “devices” subject to the full suite of medical-device requirements regardless of whether the tests fall into the categories outlined in the enforcement discretion policies. For example, FDA does not identify any textual basis in the statute for subjecting new tests to a different regime than existing tests.

104. The broad carveouts are also inconsistent with FDA’s public-health rationale for the rule. For example, FDA cannot explain why, on the one hand, more limited regulation is sufficient for the tens of thousands of laboratory-developed tests in existence at the time of the final rule, but on the other hand, virtually every test developed *after* May 6, 2024, must run the full gauntlet of the medical-device requirements.

105. The final rule states that FDA will phase out its “general enforcement discretion approach” within a four-year period. 89 Fed. Reg. at 37,294. As a result, excepting the “enforcement discretion policies” described above, FDA will begin enforcing medical-device requirements with respect to laboratory-developed tests in five stages measured from the date of publication of the final rule:

- *After 1 year,* FDA will expect compliance with medical device reporting (“MDR”) requirements, correction and removal reporting requirements, and some QS requirements under 21 C.F.R. § 820.198.

- *After 2 years,* FDA will expect compliance with requirements not covered during other stages of the phaseout policy, including registration and listing requirements, labeling requirements, and investigational use requirements.

- *After 3 years,* FDA will expect compliance with other QS requirements under 21 C.F.R. § 820.198.

- *After 3½ years,* FDA will expect compliance with premarket review requirements for “high-risk IVDs offered as LDTs.”

- *After 4 years,* FDA will expect compliance with premarket review requirements for “moderate-risk and low-risk IVDs offered as LDTs.”

*Id.*

106. At the same time, the final rule emphasizes that both the “enforcement discretion policies” and the phased-in approach are merely matters of prosecutorial discretion and that laboratories are legally required to comply with all medical-device regulations *immediately.* The rule states that “the phaseout policy does not in any way alter the fact that it is illegal to offer” laboratory-developed tests “without complying with applicable requirements” and stresses that “[r]egardless of the phaseout timeline and enforcement discretion policies … FDA retains discretion to pursue enforcement action for violations of the FD&C Act at any time, and intends to do so when appropriate.” *Id.* at 37,295.

107. With respect to the “enforcement discretion policies,” FDA further cautions that “[a]s with any enforcement discretion policy, FDA may update any of these policies as circumstances warrant or if the circumstances that inform these policies change.” *Id.* at 37,297. FDA again emphasizes that “these enforcement discretion policies do not confer lawful marketing status on any [laboratory-developed tests] being marketed as described in the policies” and “do not in any way alter the fact that it is illegal to market [a laboratory-developed test] without complying with applicable requirements.”
The final rule also warns that FDA “intends to take action to enforce applicable requirements for [laboratory-developed tests] … as appropriate, taking into account any public health concerns as evaluated on a case-by-case basis.” *Id.* For example, “if FDA receives reports, or otherwise learns of information, that raise safety or effectiveness concerns with [a laboratory-developed test] that falls within an enforcement discretion policy, FDA generally intends to take action with respect to requirements applicable to that specific [test].” *Id.*

Again and again throughout the final rule, FDA declares that no laboratory is safe from enforcement merely because its conduct is consistent with FDA’s stated enforcement discretion policies. *See id.* at 37,301 (“[A]s noted elsewhere in this preamble, regardless of this or any other enforcement discretion policy, FDA retains discretion to pursue enforcement action at any time against violative [laboratory-developed tests] when appropriate.”); *id.* at 37,304 (same); *id.* at 37,307 (same).

As to existing tests, the final rule also states that FDA will expect compliance with premarket review and quality system requirements whenever the test is “changed in certain, more significant ways that could affect its basic safety and effectiveness profile,” such as “includ[ing] significantly different technology” in the test. *Id.* at 37,305. FDA does not explain how a laboratory might determine when a difference in technology is so “significant” as to trigger an expectation of compliance.

Under the final rule, FDA thus continues to assert comprehensive authority to regulate virtually all laboratory-developed tests as medical devices. FDA then tries to mitigate the fallout from that regulatory sea change by announcing vague, non-binding enforcement discretion policies in a 150-plus-page preamble to its final rule. But FDA takes the position that even laboratories acting within the scope of those vaguely defined policies are violating federal law and that FDA can decide to prosecute them at any time, leaving laboratories “at the mercy of [FDA’s] noblesse oblige.” *FCC v. Fox Television Stations, Inc.*, 567 U.S. 239, 255 (2012) (quoting *United States v. Stevens*, 559 U.S. 460, 480 (2010)).

In response to comments questioning FDA’s legal authority, FDA doubles down on its theory that professional laboratory testing services are medical “devices” just like pacemakers or test kits. “As an initial matter,” FDA says, “FDA does not read the definition of device to encompass only physical objects.” 89 Fed. Reg. at 37,331. And “[r]egardless” of that reading (i.e., even assuming the “device” definition is limited to tangible products), FDA explains, “a test system” developed by a laboratory “is a physical product and a material thing” because it involves “a set of components—such as reagents, instruments, and other articles—that function together to produce a test result.” *Id.* In other words, FDA’s position is that whenever laboratory professionals use multiple tangible articles together to perform a test, they are “manufacturing” a “device.” While FDA superficially disclaims that view, stating that its “position is not that laboratory services are articles but that in vitro diagnostic products used in laboratories (such as test systems) are articles,” FDA has effectively adopted a definition of “test systems” that conflates professional laboratory services with the articles used to perform those services.

Whereas the statutory definition of “device” refers to discrete objects or fixed assemblages of objects, which can typically be packaged and shipped, FDA’s approach treats as a “device” even a set of transient relationships between physical articles used by a skilled professional. For example, under FDA’s reductionist approach, if a surgeon uses multiple objects to perform a procedure, such as a scalpel and a set of sutures, the surgeon has “manufactured” a “device” by using those objects in combination.

In the final rule’s preamble, FDA also asserts that the statutory term “article” cannot be limited to tangible goods because, in FDA’s view, computer software can qualify as a medical device despite being “an intangible thing.” 89 Fed. Reg. at 37,331–32. Even assuming FDA is correct that software may sometimes qualify as a device, that does not support FDA’s assertion that
the “device” definition can be stretched to cover the intangible professional services provided by laboratory medical professionals, which are different from manufactured medical devices. As the Supreme Court has explained, while it is possible to conceive of “software in the abstract: the instructions themselves detached from any medium,” “[w]hat retailers sell, and consumers buy,” are “tangible,” “physical cop[ies] of the software” that, whether “delivered by CD-ROM” or “downloaded from the Internet,” are ultimately “contained in and continuously performed by” a piece of physical hardware such as a computer.  
*Microsoft Corp. v. AT&T Corp.*, 550 U.S. 437, 446–48, 449–51 (2007).

115. FDA acknowledges that the final rule will impose major burdens on laboratories, but as with the proposed rule, FDA underestimates the impact. FDA projects that the requirements in the final rule will initially affect about 79,114 existing tests offered by 1,181 existing laboratories, and that it will also affect about 10,013 new tests offered every year going forward. *See* FDA, Laboratory Developed Tests Final Rule: Final Regulatory Impact Analysis at 36, 54–55 (May 6, 2024) (“Final Impact Analysis”) (attached as Ex. H). As FDA notes, “most facilities that will be affected by this rule are defined as small businesses and the final rule is likely to impose a substantial burden on the affected small entities.” *Id.* at 6–7; 89 Fed. Reg. at 37,433.

116. Even under the generous assumption that FDA will adhere to its non-binding enforcement discretion policies, *see* Final Impact Analysis at 37–38, FDA estimates that it will need to review an additional 103 premarket applications, 1,090 premarket notifications, and 267 de novo classification requests each year—a vast increase in each category compared to the average from 2017 to 2021, including more than a doubling of the number of premarket applications, *see id.* at 57.

117. FDA estimates that the compliance costs for laboratories will total well over $1 billion per year. *See id.* at 2, 135, 178. Over the next two decades, FDA projects that total costs associated with the final rule will range from $12.57 billion to $78.99 billion, with a primary estimate of $28.61 billion. *Id.* at 125.

118. FDA acknowledges that the huge “increased cost to laboratories” may cause price increases for customers and “reduce the amount of revenue a laboratory can invest in creating and/or modifying” tests. *Id.* at 127.

#### F. HealthTrackRX and other ACLA members face irreparable harm from FDA’s final rule.

119. Under Fifth Circuit precedent, “the nonrecoverable costs of complying with a putatively invalid regulation typically constitute irreparable harm.” *Rest. L. Ctr. v. U.S. Dep’t of Lab.*, 66 F.4th 593, 597 (5th Cir. 2023) (collecting cases).

120. By FDA’s own admission, the final rule will impose significant nonrecoverable compliance costs on regulated laboratories, including HealthTrackRX and other ACLA members. *See* Final Impact Analysis at 125 (estimating compliance costs of about $101 million in year 1, $113 million in year 2, $386 million in year 3, and more than $1.6 billion every following year). These costs will be unrecoverable because FDA, like other federal agencies, enjoys sovereign immunity from monetary damages. *See* *Rest. L. Ctr.*, 66 F.4th at 598.

121. HealthTrackRX and other ACLA members will need to begin incurring these costs immediately. FDA has made clear that despite the “phaseout” timeline and “enforcement discretion” policies in the final rule, it “retains the authority to enforce any applicable requirements and pursue enforcement action *at any time*” against laboratories that offer laboratory-developed tests without complying with regulatory requirements applicable to medical devices. 89 Fed. Reg. at 37,372 (emphasis added). And even if FDA were to commit to not taking enforcement action before the dates set forth in the policy (and FDA has expressly disclaimed such a commitment), laboratories would still have to begin incurring compliance costs well in advance of those dates to ensure full compliance by the relevant deadline.

122. Although FDA greatly underestimates both the magnitude of unrecoverable compliance costs and how quickly laboratories will begin incurring those costs, even FDA suggests that laboratories’ costs of compliance in the first year after publication of the final rule will range from $47.85 million
to $216.75 million, with a primary estimate of $101.46 million. *See* Final Impact Analysis at 125. Those unrecoverable costs alone are sufficient to establish that laboratories face irreparable harm from the final rule.

123. Moreover, FDA wrongly assumes that laboratories will be able to defer certain compliance costs for several years. For example, FDA predicts that costs associated with preparing and submitting premarket approval applications, premarket notifications, and de novo classification requests for laboratory-developed tests—costs that FDA acknowledges will easily run to billions of dollars—will not occur until the third year after publication of the final rule. *See id.* at 124–25.

124. Contrary to FDA’s assumptions, laboratories cannot delay incurring these costs until just six to eighteen months before FDA says it will begin enforcing premarket review requirements. Indeed, to support premarket applications, laboratories will need to begin preparatory work immediately, including meeting with FDA reviewers to agree on analytical and clinical validation study protocols, running such validation studies, and otherwise compiling the voluminous material required to support FDA approval or clearance.

125. HealthTrackRx and other ACLA members have made substantial financial investments to maintain and expand their business—including opening new laboratories, acquiring assets, and hiring employees—in reliance on the understanding that laboratory testing services are not subject to FDA regulation as medical devices and are instead regulated under CLIA and applicable state law. Dr. Reddy Decl. ¶ 24; Dr. Eisenberg Decl. ¶ 18; Dr. Fesko Decl. ¶ 16; Dr. Genzen Decl. ¶ 22; Dr. Morice Decl. ¶ 66.

126. FDA’s final rule puts HealthTrackRx and other ACLA members in an untenable situation of regulatory uncertainty, which creates a serious risk of chilling investment in the maintenance of existing testing services and the development of new or modified testing services. Dr. Reddy Decl. ¶ 27; Dr. Eisenberg Decl. ¶ 17; Dr. Fesko Decl. ¶¶ 18–21; Dr. Genzen Decl. ¶¶ 47, 52; Dr. Morice Decl. ¶¶ 15–16, 58–65, 68.

127. ACLA members have already expended substantial time and capital to prepare for, and ensure that they are able to comply with, FDA’s final rule. *See* Dr. Reddy Decl. ¶ 26.

128. If FDA’s final rule is permitted to take effect, ACLA members will face even greater unrecoverable costs, often in the hundreds of thousands of dollars or more per test, in order to ensure that they can remain in compliance with federal law. *See id.* ¶ 29; Dr. Fesko Decl. ¶ 19; Dr. Genzen Decl. ¶¶ 50, 56.

129. If the final rule is allowed to remain in place, there is a substantial risk that some tests will no longer be available to help providers and patients because of the prohibitive costs of seeking FDA approval and clearance, especially for tests that are not high-volume. *See* Dr. Reddy Decl. ¶ 31; Dr. Genzen Decl. ¶¶ 56, 58; Dr. Morice Decl. ¶¶ 16, 59–61.

130. Given the unrecoverable costs of complying with the FDA’s medical device requirements, and the likelihood that device regulation will exacerbate an FDA-review bottleneck, the final rule will hinder innovation by making it more difficult for ACLA members to develop, and for patients to access, new and modified tests. *See* Dr. Reddy Decl. ¶¶ 32–37; Dr. Genzen Decl. ¶¶ 47–51; Dr. Morice Decl. ¶¶ 15–16, 59–65, 68; Dr. Fesko Decl. ¶ 21.

### CLAIMS FOR RELIEF

#### COUNT 1 Violation of the Administrative Procedure Act—Contrary to Law, in Excess of Statutory Jurisdiction and Authority, and Contrary to Constitutional Right and Power 5 U.S.C. § 706(2)

131. Plaintiffs reallege and incorporate by reference each of the preceding paragraphs as if set forth fully herein.

132. Under the Administrative Procedure Act, a court must set aside agency action that is not in accordance with law or in excess of statutory authority. *See* 5 U.S.C. § 706. An agency action is invalid and must be vacated if it exceeds the power conferred upon the agency by the statute. *See Perez v. Mortg. Bankers Ass’n*, 575 U.S. 92, 104–05 (2015).

133. FDA’s final rule is contrary to law and in excess of FDA’s statutory jurisdiction and authority because it treats laboratory testing services as medical devices that are subject to regulation under the FDCA, when in fact...
they are services performed by highly skilled healthcare professionals.

134. The text, structure, and history of the FDCA make clear that a device is a physical product, not a professional service. Treating laboratory testing services as devices would not only do violence to the statutory definition of “device,” but would also require distorting numerous other statutory and regulatory provisions that confirm that FDA’s device authority is limited to physical goods.

135. The text and history of CLIA provide further confirmation that laboratory services are not devices subject to regulation under the FDCA. In CLIA, Congress created a comprehensive framework for the regulation of professional laboratory testing services—including extensive personnel qualification requirements, quality controls, and proficiency testing—that is separate and distinct from the framework for regulation of manufactured medical devices under the FDCA. Congress first enacted CLIA in 1967, nine years before the Medical Device Amendments, and significantly expanded CLIA in 1988, twelve years after those Amendments. In doing so, Congress never so much as hinted at any existing authority of FDA to regulate laboratory testing services under the FDCA. And when Congress acted to create a distinct and uniform system of regulation for clinical laboratories, it clearly indicated that FDA had no such authority.

136. Although the clarity of the statutory text should put an end to the inquiry, FDA’s attempt to regulate laboratory testing services as devices also implicates the major questions doctrine. *See West Virginia, 597 U.S. at 724.* FDA’s rule would mean that the entire clinical laboratory sector, which is a significant part of the U.S. healthcare system, has been breaking the law for nearly 50 years, and possibly much longer. And it would mean that going forward, the entire profession is operating unlawfully and can be subject to civil and criminal penalties at any time, with its only protection coming from a policy of enforcement discretion that FDA insists it is free to revoke at any time. The rule would also wreak havoc on clinical laboratories and the doctors and patients they serve, imposing billions of dollars in immediate, unnecessary costs and preventing countless new tests from ever being developed. And it would produce a vast increase in the number of medical-device applications FDA must review every year.

137. An agency cannot impose massive costs and place an entire profession under its thumb in this manner without, at minimum, a clear statement from Congress. The Supreme Court has repeatedly warned that agencies should not attempt, and courts should not abide, such drastic expansions of the agency’s authority under a “‘long-extant statute”’—especially where, as here, Congress has “conspicuously and repeatedly declined to enact” such an expansion itself. *West Virginia, 597 U.S. at 724 (quoting Util. Air Regul. Grp., 573 U.S. at 324).*

#### COUNT 2 Violation of the Administrative Procedure Act—Arbitrary and Capricious and an Abuse of Discretion 5 U.S.C. § 706(2)

138. Plaintiffs reallege and incorporate by reference each of the preceding paragraphs as if set forth fully herein.

139. Under the Administrative Procedure Act, a court must set aside agency action that is arbitrary and capricious, an abuse of discretion, or inconsistent with the requirements of reasoned decision-making. *See* 5 U.S.C. § 706. An action is arbitrary and capricious if agency acts outside the reasonable scope of its lawful authority, fails to articulate a satisfactory explanation for its actions, or fails to respond adequately and reasonably to comments and objections.

140. FDA has not acted consistent with the requirements of reasoned decision-making because it has not adequately responded to objections, provided a reasoned justification for its rule, or reasonably explained its sweeping assertion of new regulatory authority. FDA’s decision to exercise enforcement discretion through the use of non-binding guidance in a preamble only underscores how unreasonable it is for FDA to outlaw an entire sector of professional services, especially given the reliance interests at stake.

141. Accordingly, even if Congress had granted FDA authority under the FDCA to regulate certain types of laboratory-developed tests as medical
devices, the agency has not exercised that authority consistent with the requirements of reasoned decision-making under the Administrative Procedure Act and the Constitution’s separation of powers.

142. FDA’s final rule cannot be allowed to stand because it is ultra vires, arbitrary and capricious, and an abuse of discretion.

### PRAYER FOR RELIEF

WHEREFORE, Plaintiffs request that the Court:

A. Enter a declaratory judgment that FDA’s final rule is contrary to law; in excess of statutory jurisdiction, authority, or limitations; and arbitrary or capricious, and that FDA is not authorized to regulate laboratory testing services as medical devices under the FDCA.

B. Enter an order that vacates FDA’s final rule and enjoins FDA from enforcing the final rule and regulating laboratory testing services as medical devices under the FDCA.

C. Order such other and further relief as the Court deems just and proper.

Dated: May 29, 2024

Respectfully submitted,

**/s/ Edward F. Fernandes**  
Edward F. Fernandes  
Texas Bar No. 06932700  
Christopher H. Taylor  
Texas Bar No. 24013606  
KING & SPALDING LLP  
500 W. 2nd Street  
Suite 1800  
Austin, TX 78701  
Tel: (512) 457-2000  
Fax: (512) 457-2100  
efernandes@kslaw.com  
ctaylor@kslaw.com  

Ashley C. Parrish*  
D.C. Bar No. 464683  
Paul Alessio Mezzina*  
D.C. Bar No. 999325  
Alexander Kazam*  
D.C. Bar No. 1708188  
KING & SPALDING LLP  
1700 Pennsylvania Avenue NW  
Suite 900  
Washington, DC 20006  
Tel: (202) 737-0500  
Fax: (202) 626-3737  
aparrish@kslaw.com  
pmezzina@kslaw.com  
akazam@kslaw.com  

Scott D. Danzis*  
D.C. Bar No. 481426  
COVINGTON & BURLING  
One CityCenter  
850 Tenth Street, NW  
Washington, DC 20001-4956  
Tel: (202) 662-6000  
sdanzis@cov.com  

*Counsel for Plaintiffs*  

*pro hac vice applications forthcoming*

## EXHIBIT A

AMERICAN CLINICAL LABORATORY ASSOCIATION;  
HEALTHTRACKRX INDIANA, INC.; and HEALTHTRACKRX, INC.,  
Plaintiffs,  

v.  

U.S. FOOD AND DRUG ADMINISTRATION; U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES; XAVIER BECERRA, in his official capacity as Secretary of Health and Human Services; and ROBERT M. CALIFF, M.D., in his official capacity as Commissioner of Food and Drugs, United States Food and Drug Administration,  
Defendants.  

**DECLARATION**

### DECLARATION OF JAY REDDY, Ph.D.

I, Jay Reddy, Ph.D., declare as follows:

1. I am a resident of Denton, Texas. I am over the age of eighteen, and I am competent to provide this declaration.

2. I am the Senior Vice President of Laboratory and Clinical Strategy at HealthTrackRx. I have held that position for a little more than 2 years. I was previously the company’s Vice President of Laboratory Operations.

3. I have worked at HealthTrackRx for more than 9 years.

4. I received my Ph.D. in Molecular Biology at Texas Women’s University.

5. As a result of my professional experiences and background, I am familiar with the laboratory testing services and procedures that HealthTrackRx offers. I am also familiar with the legal and regulatory requirements that apply to laboratory-developed tests.

6. I am deeply concerned about the final rule issued by the Food and Drug Administration (“FDA”) on May 6, 2024, and the enormous costs it will impose on clinical laboratories, such as HealthTrackRx. FDA’s final rule also creates serious risks for patients by threatening to reduce their access to important and safe testing services.

#### HealthTrackRx

7. HealthTrackRx is one of the nation’s premier infectious disease laboratories.

8. HealthTrackRx has its principal place of business in Denton, Texas.

9. For more than twenty years, HealthTrackRx has helped providers make informed clinical decisions by offering next-morning testing results to healthcare providers throughout the nation using accurate and targeted molecular diagnostic testing processes and procedures.

10. HealthTrackRx specializes in a particular type of testing service that employs a pathogen assay referred to as Real-Time Transcription Polymerase Chain Reaction (“RT-PCR”). The RT-PCR technology is used to detect bacteria and viruses on a molecular level. PCR testing services, which can be performed overnight, allow for faster and more accurate results than traditional culture assays, which can take 3 to 14 days (or even longer) to obtain results. By using advanced PCR testing, HealthTrackRx is able to help providers make better and more informed patient treatment decisions.

11. The company’s testing services cover a wide variety of different specialties, such as otolaryngology and ophthalmology, and target viral, bacterial, and fungal pathogens that cause a range of everyday infections, including respiratory infections, urinary tract infections, genitourinary infections, gastrointestinal infections, wound infections, and onychomycosis.

12. All of HealthTrackRx’s laboratories are certified by the United States Department of Health and Human Services under the federal Clinical Laboratory Improvement Amendments (“CLIA”) to perform clinical tests on human specimens. All HealthTrackRx laboratories are enrolled in proficiency
13. All of HealthTrackRx’s laboratories also participate in the rigorous accreditation program overseen by CAP, which is the “gold standard” accreditation program for molecular diagnostic laboratories.

14. HealthTrackRx has more than 300 employees and operates laboratories in Denton, Texas; Clarksville, Indiana; Marietta, Georgia; and Sherman Oaks, California. All testing personnel meet CLIA/CAP requirements, and HealthTrackRx employees in California are certified by the California Department of Public Health to perform testing. HealthTrackRx also employs professional “Certifying Scientists” to review quality control and raw data before releasing patient reports to providers. These professional scientists have a minimum of a Master’s degree in a biological or chemical science, and most also have certification by the American Society of Clinical Pathology (MB (ASCP) or MLS (ASCP)).

15. The testing services provided by HealthTrackRx are carefully validated according to federal guidelines before those services are offered to healthcare providers. Analytical and clinical performance evaluations are concluded before the launch of any new testing service, and tests are continuously re-evaluated throughout each year to ensure patient data integrity.

16. HealthTrackRx has assumed an active leadership role in educating the healthcare community about Antimicrobial Resistance, which is recognized by the Centers for Disease Control as one of the greatest threats to public health worldwide. Antimicrobial Resistance occurs when bacteria mutate over time and no longer respond to antibiotics, making infections more challenging to treat and increasing the risk of contagion, severe illness, and death.

17. HealthTrackRx helps providers by detecting antibiotic resistant genes and identifying pathogens that are difficult to culture. Because our best-in-class PCR tests meet demanding sensitivity and specificity standards, and because we are able to offer next-morning results that provide a personalized, patient-specific antibiotic summary, we are able to help providers identify appropriate antibiotics that work for their patients.

18. HealthTrackRx promotes antibiotic stewardship through active training on diagnostic stewardship. All new and existing accounts have ongoing training to help identify the appropriate tests for their patient population. Appropriate test ordering is key to reducing unnecessary prescribing of antibiotics.

19. HealthTrackRx has an Advisory Board that is made up of leaders in the healthcare industry. Members of the Board include:

   - **Barbara Alexander, M.D., FACP, FIDSA**, the company’s Chief Medical Advisor and Chair of the Advisory Board. Dr. Alexander is a Professor of Medicine and Pathology at Duke University, where she serves as Director of the Transplant/Immunocompromised Infectious Diseases Services.
   - **Josh M. Berlin, J.D.**, Chief Executive Officer of rule of three, LLC (“ro3”).
   - **Leah Binder**, President and Chief Executive Officer of Leapfrog Group.
   - **Elizabeth Canis**, former Vice President for Emerging Business and Partnerships at Anthem.
   - **Eric D. Hargan**, former Acting Secretary for the U.S. Department of Health and Human Services.
   - **David B. Nash, M.D., MBA**, Founding Dean Emeritus at the Jefferson College of Population Health.

20. HealthTrackRx also has a Clinical Advisory Board, composed of leading experts in infectious disease and antimicrobial resistance. Our Clinical Advisory Board helps guide the company’s development of new testing services and menus, seeking to improve patient outcomes and combat antimicrobial resistance. The members of our Clinical Advisory Board include:

   - **Barbara Alexander, M.D., FACP, FIDSA**.

- **Cornelius J. Clancy, M.D.**, Director of the Mycology Program at the University of Pittsburgh, Associate Chief of the VA Pittsburgh Health System and Chief of Infectious Diseases, and a Member of the Infectious Diseases Society of America’s Committee on Antimicrobial Resistance.

- **Barry Eisenstein, M.D.**, Chief Medical Officer and former Chair of the Scientific and Business Advisory Board of CARB-X.

- **Thomas M. File, Jr., M.D., MSC, MACP, FIDSA, FCCP**, Professor of Medicine and Chair of the Infectious Disease Section at Northeast Ohio Medical University.

- **Kimberly E. Hanson, M.D., MHS, FIDSA**, Professor of Medicine and Director of Transplant Infectious Diseases and Immunocompromised Host Service, and Section Head of Clinical Microbiology at the University of Utah.

- **Robin Patel, M.D.**, Elizabeth P. and Robert E. Allen Professor of Individualized Medicine, Professor of Microbiology and Medicine, Director of the Infectious Diseases Research Laboratory, and Co-Director of the Clinical Bacteriology Laboratory at the Mayo Clinic.

- **Adriana E. Rosato, SM (ASCP), MSC, Ph.D.**, Director of the Center for Molecular Medicine at the MaineHealth Institute for Research and Visiting Professor of Medicine at Tufts University.

#### The Consequences of FDA’s Final Rule

21. In treating laboratory-developed tests the same as medical devices, FDA’s final rule fundamentally misunderstands how laboratories, such as HealthTrackRx, perform professional testing services for the benefit of patients.

22. HealthTrackRx is not a manufacturer, and it does not manufacture products or articles of equipment. Its PCR testing services are not medical devices.

23. The molecular diagnostic services that HealthTrackRx’s professionals perform at its laboratories at the request of healthcare providers involve medical procedures, protocols, and processes that are used to analyze at the molecular level tissue, blood, and other patient specimens as part of the practice of laboratory medicine.

24. HealthTrackRx has made substantial financial investments — opening new laboratories, hiring employees, developing additional testing services, and expanding its business — in reliance on the understanding that laboratory testing services are not subject to FDA regulation as medical devices and are instead carefully regulated under CLIA and applicable state regulatory regimes.

25. I am concerned about the substantial costs that FDA’s final rule poses both for HealthTrackRx and for patients.

26. HealthTrackRx has already made substantial investments in order to prepare for, and ensure that it is able to comply with, FDA’s final rule. Because there is no ready ability to recover money damages from FDA, it is unlikely that HealthTrackRx will ever be able to recover those investments.

27. If FDA’s final rule is allowed to take effect, HealthTrackRx will face even greater unrecoverable costs and financial burdens in order to ensure that it is able to remain in compliance with federal law. Although FDA has announced carveouts that are supposed to reduce the consequences of FDA’s decision to regulate laboratory-developed tests as medical devices, the carveouts create intolerable regulatory uncertainty, as FDA’s rule states that the agency could change its mind at any time.

28. Notwithstanding the carveouts that FDA says it will observe as a matter of enforcement discretion, FDA’s position appears to be that all laboratory developed tests are illegal under federal law — and subject to civil and criminal penalties — unless they have been approved and cleared by FDA.

29. FDA’s own estimates recognize that obtaining FDA approval and clearance could cost hundreds of thousands of dollars for each separate test — even if there is a predicate FDA-approved medical device that can be used as a reference product to compare to any specific test service for which a manufacturer seeks approval or clearance. If there is no predicate device, the costs are likely to be substantially larger, as professional laboratories will be required to make significant investments and undertake substantial efforts to demonstrate to FDA’s satisfaction the safety and efficacy of tests.

30. FDA lacks the resources needed to oversee the prompt approval of potentially tens of thousands of laboratory-developed tests. I am therefore...
concerned that when HealthTrackRx submits a test for FDA approval and clearance, the company will be forced to wait many months (if not longer) for approval or clearance. Any delay in providing approval or clearance could cause substantial harm by making it more difficult for HealthTrackRx to provide the essential testing services that healthcare providers rely on in order to make appropriate treatment decisions.

31. Moreover, there are tests that HealthTrackRx includes on its testing menu that, because they are not high-volume tests, may not justify the costs associated with seeking FDA approval and clearance. If FDA’s final rule is allowed to remain in place, there are significant risks that some of these tests will no longer be available to help providers and patients.

32. FDA’s final rule will create significant disincentives for financial investment by laboratories in new testing protocols and processes. The impact will be especially significant for specialized testing services.

33. For example, at the request of a group of expert ophthalmologists, HealthTrackRx worked to develop a cutting-edge PCR test that quickly diagnoses whether a corneal ulcer — also known as infectious keratitis — is caused by a virus, bacterium, or fungus. There is no FDA-approved device that ophthalmologists are able to use for this purpose. Infectious keratitis is a major cause of visual impairment and blindness, often affecting marginalized populations. Because the proper treatment of keratitis differs depending on whether it is caused by a virus, bacterium, or fungus, quick and accurate diagnosis is important. With the right diagnosis, treatment can often cure the patient; a wrong diagnosis can often be debilitating. The PCR test that HealthTrackRx developed is critical because it allows providers to confirm quickly the source of a corneal ulcer and to identify appropriate treatment options.

34. Because this type of test is not used frequently or in large volumes, it is unlikely that it would ever have been developed if FDA’s final rule were in place. Given the substantial costs of seeking FDA approval and clearance, no laboratory would invest the months of effort to develop and validate a new test for the benefit of patients with corneal ulcers.

35. As another example, HealthTrackRx has stepped up during the last two public health emergencies to help with the testing burden for both COVID-19 and Monkeypox. The Centers for Disease Control needed laboratory professionals to validate high throughput tests to help with the increased testing burden. Because the FDA-approved tests for both diseases were low throughput and could not meet testing demands, professional laboratories, including HealthTrackRx, were essential in providing services and addressing unmet needs that manufactured medical devices could not address. The FDA’s final rule will stifle the very innovation that helped with testing demands in two public health emergencies in the last four years.

36. I am also concerned that FDA’s final rule will prevent laboratories from innovating and customizing tests when necessary. One of the important features of laboratory-developed testing services is the ability of individual laboratories to modify their testing protocols and procedures to adapt to specific requests made by providers (and their patients). If FDA’s final rule remains in place, however, I am concerned that the threat of FDA fines may discourage laboratories from customizing their professional testing services when needed by patients.

37. I have additional concerns that FDA’s final rule will negatively impact both ordering providers and patients. HealthTrackRx has moved away from “static” panels to customizable menus. That approach allows the physician to request targeted testing based on the symptoms presented by the patient as opposed to relying on a “one size fits all” approach, which is how medical devices work. FDA’s final rule could make customizable menus too expensive to pursue. That would force physicians to move back to relying on broad syndromic menus, which increase costs to patients. Increased costs reduce access to testing to those who cannot afford it, and in many cases, for those who need testing the most.

In accordance with 28 U.S.C. § 1746, I declare under penalty of perjury that the foregoing is true and correct.

Executed on this **23rd** day of May, 2024.

By:  
________________________  
Jay Reddy, Ph.D.

## EXHIBIT B

AMERICAN CLINICAL LABORATORY ASSOCIATION;  
HEALTHTRACKRX INDIANA, INC.; and HEALTHTRACKRX, INC.,  
Plaintiffs,  

v.  

U.S. FOOD AND DRUG ADMINISTRATION; U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES; XAVIER BECERRA, in his official capacity as Secretary of Health and Human Services; and ROBERT M. CALIFF, M.D., in his official capacity as Commissioner of Food and Drugs, United States Food and Drug Administration,  
Defendants.  

**DECLARATION**

### DECLARATION OF MARCIA EISENBERG, Ph.D.

I, Marcia Eisenberg, Ph.D., declare as follows:

1. I am a resident of Apex, North Carolina. I am over the age of eighteen, and I am competent to provide this declaration.

2. I have worked for Labcorp for more than 30 years, including in a variety of leadership positions. I am currently a Senior Vice President and the Enterprise Chief Scientific Officer at Labcorp.

3. I oversee research and development and science and technology for the company, which includes test development, optimization, and automation. I have been recognized and honored for my contributions to the advancement of forensic DNA testing, and I have been involved with the development and validation of hundreds of clinical assays used for patient care.

4. I earned a B.S. in biology, a B.A. in psychology, and an M.S. in molecular biology from the State University of New York at Albany. I earned a Ph.D. in molecular biology from the University of Kentucky.

5. My experience includes work with the National Institute of Environmental Health Sciences. I have been a member of the FBI’s Technical Working Group on DNA Analysis Methods, and I was an appointed member of the National DNA Advisory Board during its lifespan.

#### Labcorp

6. Labcorp is a global life sciences and healthcare company. With nearly 100 laboratories in dozens of countries, Labcorp performs professional testing services for providers and patients across the world. Our services include doctor-requested testing, consumer-initiated testing, research and development laboratory testing, and clinical trial laboratory testing.

7. In 2019, Labcorp celebrated its 50th anniversary and the company’s transformation from a local laboratory operating in a former hospital to a leading global life sciences company that is deeply integrated in guiding patient care.

8. Labcorp prides itself on leveraging cutting-edge science, technology, and innovations to find healthcare answers for patients. We pursue scientific advancements in clinical diagnostic testing and breakthrough treatments, relying on a professional team of more than 2,500 M.D.s and Ph.D.s.

9. Labcorp works to provide healthcare providers and patients with the highest quality and most comprehensive menu of testing services available. We have a growing list of more than 6,000 different routine and esoteric tests.

10. Our scientific experts and specialists focus on helping patients in areas of pressing need, including oncology, Alzheimer’s, liver, and autoimmune diseases, as well as helping people prevent serious illness and remain healthy. The world-class diagnostic testing services that we provide help enable and accelerate patient care and access to innovative treatments, medicines, and new technologies that can change outcomes and lives.

11. The Labcorp Charitable Foundation, which was founded in 2020, has given out more than 341 grants to promote greater access to healthcare and education. These grants have focused on supporting food pantries and summer meal programs for children, providing access to healthcare and patient support services for the underserved, broadening access to STEM education programming, and providing ongoing support for medical research, screenings, and programs that promote a healthy lifestyle.

12. Labcorp is constantly innovating and developing new testing services and updating older tests. In April and May 2024 alone, we developed numerous new LDTs, including for early Sjogren’s Syndrome, Candida auris colonization screening, and others. We also updated dozens of other existing LDTs.

#### FDA’s Final Rule

13. The new final rule issued by the Food & Drug Administration (“FDA”) poses a significant threat to the nation’s healthcare system and the longstanding practices and regulatory requirements that Labcorp has relied on to build and expand its business.

14. FDA is wrong to suggest that laboratory developed tests are equivalent to manufactured medical devices sold in interstate commerce. The professional testing services that laboratories provide are not instruments, machines, or physical objects that qualify as medical devices subject to FDA approval and clearance. Instead, they are an essential healthcare service that involves the exercise of professional medical expertise and judgment to provide clinical information to physicians and patients. Laboratory developed testing
The professionals at Labcorp are expert technicians, M.D.s, and Ph.Ds; they are not manufacturers.

15. During my decades in the clinical laboratory industry, professional laboratory testing services have been regulated under the federal Clinical Laboratory Improvement Amendments Act of 1988 and by individual states under state law, not the FDA. Those regulations are designed to take account of the processes and procedures that are involved when professional clinicians engage in laboratory testing. In contrast, the requirements that apply to medical devices are a poor fit for regulating professional services. If the government wants to change the regulatory requirements that apply to clinical laboratories, that should be accomplished through legislation, not by re-tooling a statute that was never designed for that purpose.

16. The suggestion made by FDA in its final rule that Congress in 1976 somehow banned laboratory developed testing services unless and until they are approved and cleared by FDA would mean that thousands of laboratories have violated the law for more than half a century. That makes no sense.

17. It is also unfair. In its final rule, FDA has said that it does not intend to enforce the rules that apply to medical devices to existing laboratory developed tests, and it has suggested that it intends to limit when the agency will exercise its enforcement discretion against laboratories. But FDA has made clear that it believes those testing services are subject to the medical-device requirements. FDA has suggested that it could change its mind about enforcement at any time. That creates an untenable situation of regulatory uncertainty.

18. Labcorp has made substantial investments to expand its professional testing capabilities in reliance on the existing regulatory system and the understanding that our large team of M.D.s and Ph.Ds provide professional testing services that are not subject to the same regulatory requirements as manufactured medical devices. Since the early 2000s, Labcorp has acquired assets and made substantial investments that have allowed the company to expand its testing capabilities and expertise in the following areas: (1) ultra-sensitive hepatis C testing, (2) molecular microbial testing, (3) anatomic pathology, (4) esoteric oncology, (5) genetic and oncology services, (6) forensics, (7) specialized toxicology, and (8) cardiovascular and metabolic disorders.

19. Under FDA’s rule, for laboratories to avoid the risk of FDA changing its position and bringing future enforcement actions, they will need to seek FDA clearance or approval for the many tests that are already on the market and for every new testing protocol they might develop in the future. FDA does not have the personnel or financial resources necessary to review and approve all those testing protocols, which could take decades to complete.

Nor would it make sense for laboratories to undertake the cost and expense of seeking approval for all the different types of testing services they provide.

20. FDA’s rule threatens either to force laboratories to provide services with an uncertain risk of future enforcement or to deprive patients of essential testing services. Neither option is in the interests of laboratories or the nation’s healthcare system.

In accordance with 28 U.S.C. § 1746, I declare under penalty of perjury that the foregoing is true and correct.

Executed on this 24th day of May, 2024.

By:  
Marcia Eisenberg, Ph.D.

## EXHIBIT C

AMERICAN CLINICAL  
LABORATORY ASSOCIATION;  
HEALTHTRACKRX INDIANA,  
INC.; and HEALTHTRACKRX,  
INC.,  
Plaintiffs,  

v.  

U.S. FOOD AND DRUG  
ADMINISTRATION; U.S.  
DEPARTMENT OF HEALTH AND  
HUMAN SERVICES; XAVIER  
BECERRA, in his official capacity as  
Secretary of Health and Human  
Services; and ROBERT M. CALIFF,  
M.D., in his official capacity as  
Commissioner of Food and Drugs,  
United States Food and Drug  
Administration,  
Defendants.  

**DECLARATION**

### DECLARATION OF YURI A. FESKO, M.D.

I, Yuri A. Fesko, M.D., declare as follows:

1. I am a resident of North Carolina. I am over the age of eighteen, and I am competent to provide this declaration.

2. I have worked for Quest Diagnostics (“Quest”) for more than seven years. I currently serve as Quest’s Chief Medical Officer.

3. I am a board-certified physician in the areas of oncology, hematology, and internal medicine. In my current role, I lead Quest’s medical affairs organization and oversee the Company’s medical team of approximately 700 MDs and PhDs. Prior to joining Quest, I was medical director of oncology for Duke Cancer Center of Wake County, where I was a member of the oncology faculty.

4. Because of my experience with Quest, I have a keen understanding of how laboratory testing services are developed and provided and of the meaningful differences between professional laboratory services and manufactured medical devices.

#### Quest Diagnostics

5. Quest is one of the world’s leading providers of professional diagnostic testing services. It serves customers globally, and each year it provides testing services to approximately 1 in 3 adults and half the physicians and hospitals in the United States.

6. Quest’s diagnostic testing portfolio is focused on three areas:

   - **General Diagnostics**: Routine and non-routine testing and consultation services essential to healthcare delivery.
   - **Advanced Diagnostics**: Genetic and advanced molecular testing and consultation services based on rich clinical, scientific, and medical expertise and innovation.

- **Diagnostic Services.** A wide range of capabilities that efficiently provide healthcare insights to individuals, employers, and institutions to support population health.

7. Quest is a leader in developing laboratory testing services, with clinical experts, resources, and other professional services all focused on helping to meet the needs of providers and patients. Its broad and deep menu of testing services includes 3,500 routine, esoteric, and genetic tests.

8. Quest maintains a nationwide network of clinical laboratories, including advanced laboratories as well as rapid response laboratories (smaller facilities where we can quickly perform an abbreviated menu of routine tests for customers who require rapid turnaround times). The company operates 24 hours a day, 365 days a year, with a nationwide network of patient service centers, phlebotomists in physician offices, and a range of connectivity resources, including call centers and mobile paramedics, nurses, and other health and wellness professionals.

9. Quest employs approximately 700 MDs and PhDs and approximately 23,000 phlebotomists, paramedics, nurses, and health and wellness professionals. It has a large in-house staff of medical and scientific experts, including medical directors, scientific directors, genetic counselors, and board-certified geneticists. These professionals provide medical and scientific consultation to healthcare providers and patients regarding the company’s testing services, helping to improve health outcomes. The company’s professionals also publish original research in peer-reviewed publications, at medical and scientific conferences, and as a public service.

10. Quest is a leading provider of diagnostic information services for infectious disease, such as COVID-19 (including molecular diagnostic and serology antibody offerings), tuberculosis, and tick-borne disease. It also provides diagnostic solutions for emerging infectious diseases, such as Zika, West Nile Virus, SARS, and Influenza A H1N1. It has leading positions in drug monitoring and toxicology, in neurology diagnostics, in advanced cardiovascular diagnostic services, and in cancer diagnostics.

11. Quest is also a leading provider of workplace drug testing. It is certified by the U.S. Department of Health and Human Services to perform federally mandated drug testing using electronic custody and control forms for safety-sensitive workers. In addition, Quest is a leading provider of employer population health services, including biometric screenings, flu shots, and related preventative services that leverage clinical data to improve population health outcomes and reduce healthcare costs. The solutions provided by Quest enable employers to leverage screening insights to identify chronic disease risks, connect employees to needed in-network care, and empower better health.

12. As an organization dedicated to improving our patients’ health, Quest aims to support a culture that is guided by “the 5Cs”:

   - **Customers first:** Every decision made by Quest Diagnostics starts with a patient or customer in mind. The quality of work is vital because the answers it delivers are a matter of life.

   - **Care:** Quest is in the healthcare business, with care at the core of everything it does. Its professionals do the right thing with empathy, integrity, and respect to show each patient, customer, and colleague they matter.

   - **Collaboration:** Creating a healthier world is a monumental task. The professionals at Quest work as a team, internally across departments and externally throughout the healthcare ecosystem and in the communities where its professionals work and live.

   - **Continuous improvement:** Quest recognizes that delivering superior quality requires intention and innovation. It is committed to being better today than yesterday and even better tomorrow.

   - **Curiosity:** The professionals at Quest are constant learners. To do their jobs, they must be relentlessly curious, because that is what it takes to move healthcare forward.

#### FDA’s Final Rule

13. The final rule issued by the Food & Drug Administration (“FDA”) on May 6, 2024, seeks to treat professional laboratory testing services as if they are manufactured medical devices and asserts sweeping authority to subject those professional services to the same regulatory requirements as medical devices under the Food, Drug and Cosmetic Act.

14. FDA’s final rule is premised on the extraordinary view that in 1976, Congress prohibited the development and use of any laboratory tests that were not approved and cleared by FDA, and that in the ensuing five
decades, thousands of professional laboratory clinicians across the nation have been violating federal law by offering tests to physicians and patients without first seeking FDA approval and clearance. That makes no sense and is unfair to the thousands of Quest employees who are dedicated to helping patients by providing essential testing services.

15. FDA’s final rule ignores the fundamental differences between professional laboratory testing services and manufactured medical devices. Quest is not a manufacturer. It is a healthcare organization dedicated to and focused on providing empathetic, high-quality, and individualized patient care. The professional diagnostic services that Quest provides are not physical objects or articles that qualify as medical devices. Instead, the information and consultation services that Quest provides require the exercise of medical expertise and professional judgment.

16. Quest has made substantial investments in reliance on the understanding that Congress has never granted FDA authority to treat professional diagnostic testing services the same as manufactured medical devices. Those services are instead regulated under a specialized, laboratory-specific framework set forth in the Clinical Laboratory Improvement Amendments of 1988 (“CLIA”). That framework was carefully designed and tailored by Congress to ensure the validity and appropriateness of the processes and procedures that are followed when laboratory professionals develop and perform testing services.

17. Congress has recently considered legislation that would change the regulatory requirements for laboratory testing services. Congress has so far rejected calls to grant FDA authority to regulate laboratory developed testing services and has instead decided to keep the existing regulatory framework in place.

18. Even though Congress has never granted FDA authority to regulate laboratory professional services, FDA takes the position that all laboratory developed tests are subject to the statutory requirements that apply to manufactured medical devices (in addition to the laboratory-specific requirements that apply under CLIA). In an attempt to reduce some of the immediate negative consequences that this position would otherwise have on the nation’s healthcare system, FDA has said that, at least for now, it does not intend to enforce certain regulatory requirements with respect to existing testing menus, and it has said that the agency will exercise its “enforcement discretion” not to enforce the statute in a variety of contexts. This attempt to soften the blow of its final rule does not solve the rule’s many problems and only creates more regulatory uncertainty.

19. Most notably, FDA has emphasized that it could change its mind and end its supposed “enforcement discretion” at any time. That puts Quest (and other laboratories) in an impossible position — either (1) expend substantial resources to go through FDA’s costly approval and clearance process and have fewer resources available to invest in continued innovation and development, or (2) continue to provide the essential testing services relied on by one third of the nation’s adult population and risk being subject to future enforcement action by FDA. No company should be put in that impossible position.

20. It is also unclear how the nation’s laboratories are likely to respond to the uncertainty created by FDA’s new rule. FDA’s attempt to assert authority over an entire profession is a massive undertaking, and when laboratories start seeking FDA approval and clearance for their testing services the problems are only likely to increase. The costly requirements imposed by the medical device regulations are not calibrated to account for the professional and expert judgments made by laboratory professionals when developing and performing diagnostic testing services.

21. Moreover, FDA lacks the resources and experience to review and approve the many thousands of different laboratory tests that are available today and the many more tests that are likely to be developed in the future. It would take years (if not decades) for FDA to review, approve, and clear all of the laboratory tests that are offered by Quest and other laboratories. But no one knows how long FDA will continue to exercise enforcement discretion or how it will respond if laboratories start seeking FDA approval and clearance. The risk of regulatory delay is significant, which could chill the incentives for laboratories to innovate and develop new testing processes and procedures needed to keep up with the pace of change in multiple medical specialties.

In accordance with 28 U.S.C. § 1746, I declare under penalty of perjury that the foregoing is true and correct.

Executed on this 24th day of May, 2024.

By:  
Yuri A. Fesko, M.D.

## EXHIBIT D

AMERICAN CLINICAL  
LABORATORY ASSOCIATION;  
HEALTHTRACKRX INDIANA,  
INC.; and HEALTHTRACKRX,  
INC.,  

Plaintiffs,  

v.  

U.S. FOOD AND DRUG  
ADMINISTRATION; U.S.  
DEPARTMENT OF HEALTH AND  
HUMAN SERVICES; XAVIER  
BECERRA, in his official capacity as  
Secretary of Health and Human  
Services; and ROBERT M. CALIFF,  
M.D., in his official capacity as  
Commissioner of Food and Drugs,  
United States Food and Drug  
Administration,  

Defendants.  

**DECLARATION**

### DECLARATION OF JONATHAN GENZEN, M.D., Ph.D.

I, Jonathan Genzen, M.D., Ph.D., declare as follows:

1. I am a resident of Salt Lake City, Utah. I am over the age of eighteen, and I am competent to provide this declaration.

2. I am the Chief Medical Officer and Senior Director of Government Affairs at ARUP Laboratories and have served in these roles since July 2022. I am also the Medical Director of Laboratory Automation for ARUP Laboratories and a Co-Medical Director of Automated Core Laboratory at ARUP Laboratories.

3. I have worked at ARUP Laboratories for more than 11 years.

4. In addition to my work at ARUP Laboratories, I am a clinical Professor in the Department of Pathology at the University of Utah School of Medicine. I am licensed to practice medicine in both Utah and New York.

5. I received my Ph.D. in Biological Sciences at the University of Chicago and my M.D. in Medicine at the University of Chicago Pritzker School of Medicine.

6. I completed my clinical pathology residency training at Yale University / Yale New Haven Hospital, and I was a post-doctoral Research Fellow at Yale University in the Department of Laboratory Medicine.

7. I specialize in clinical pathology, laboratory medicine, clinical chemistry, medical diagnostics, in vitro diagnostics, and laboratory automation.

8. I am a fellow of the American Society for Clinical Pathology and the College of American Pathologists. I am also a member of the Association for Diagnostics and Laboratory Medicine and the Academy of Clinical Laboratory Physicians and Scientists.

9. As a result of my professional experiences and background, I have significant knowledge of clinical diagnostic laboratory services. I am familiar with the legal and regulatory requirements that have long applied to laboratory-developed tests, including the regulations that apply to the conduct of laboratories and the validation of diagnostic testing services.

10. I am also familiar with the final rule issued by the Food and Drug Administration ("FDA") on May 6, 2024, and the analysis that the FDA has relied on to justify its new rule. I was directly involved in preparing the comments submitted on the FDA’s proposed rule by ARUP Laboratories.

11. As described in greater detail below, I am deeply concerned about the FDA’s final rule, the enormous costs it will impose on clinical laboratories, and the harm it will cause to patients across the nation. The FDA’s final rule is based on a flawed understanding of how laboratories develop and provide professional testing services to help healthcare providers treat and diagnose patients. The rule poses serious risks to patients by threatening to reduce access to safe testing services over time, which will disproportionately harm patients with rare diseases, underserved patient populations, patients with cancer, and children.

#### ARUP Laboratories

12. ARUP Laboratories was founded 40 years ago in 1984 as an enterprise of the University of Utah’s Department of Pathology. ARUP Laboratories operates the hospital and outpatient clinical laboratories for one of the nation’s most respected academic medical centers — University of Utah Health.

13. ARUP Laboratories is an academic, non-profit institution dedicated to providing hospitals and health systems with unparalleled quality testing services, particularly for otherwise unmet needs, while continuously adapting to the ever-changing needs of the healthcare industry.

14. ARUP Laboratories is a member of the American Clinical Laboratory Association ("ACLA"). Our Chief Executive Officer, Andy Theurer, is a member of the ACLA's Board of Directors.

15. ARUP Laboratories participates in the leading certification programs for clinical laboratories. It is accredited by the College of American Pathologists ("CAP-accredited") and certified by the International Organization for Standardization, which establishes the international standards for quality and competence in medical laboratory environments ("ISO 15189-certified"). It has also received certification by the Centers for Medicare and Medicaid Services under the requirements of the Clinical Laboratory Improvement Amendments of 1988 ("CLIA") ("CLIA-certified").

16. ARUP Laboratories is focused on providing hospitals and healthcare systems with unmatched professional clinical laboratory testing services, helping them to remain cost-effective and improve patient care. We believe in...
collaborating, sharing knowledge, and contributing to laboratory science in ways that provide the best value for patients.

17. **ARUP Laboratories** is the nation’s largest non-profit clinical reference laboratory. With more than 70 laboratory sections jointly located on its 700,000-square-foot main campus in Salt Lake City, Utah, ARUP Laboratories provides all types of comprehensive laboratory testing services, from routine screening tests to esoteric molecular and genetic assays. We provide professional services to more than 2,000 community hospitals and academic medical centers across the nation.

18. ARUP Laboratories has more than 4,000 employees, with a testing menu that offers hospitals and health centers access to more than 3,000 tests and test combinations. It processes an average of 70,000 specimens of blood, body fluid, and tissue per day, impacting more than 8 million patients each year.

19. ARUP Laboratories has more than 100 nationally and internationally recognized board-certified medical directors — including pathologists, subspecialty-qualified clinicians, and board-certified clinical scientists. Our MD and PhD certified professionals have extensive medical and scientific expertise to ensure that our testing services meet the ongoing clinical needs of health care providers and patients, while also providing clinical consulting services by telephone and secure email communications.

The expertise of our medical directors and scientists ensures an exceptionally high standard of quality, with each assay that ARUP Laboratories develops undergoing a rigorous and scientific validation process before it is added to the menu of testing services that ARUP Laboratories provides.

20. The **ARUP Institute for Clinical and Experimental Pathology** is the research and development arm of the organization, with over 60 scientists actively engaged in test development and optimization in collaboration with experts in the University’s medical facilities. Over several decades, the Institute has developed, validated, verified, improved, and maintained at least 1,500 laboratory-developed tests.

21. Consistent with ARUP’s academic foundations and its commitment to sharing knowledge with the clinical laboratory community, ARUP scientists and medical directors publish over 130 peer-reviewed articles each year. Collectively, we have published more than 3,400 scientific and clinical manuscripts in peer-reviewed journals to date.

22. In 2021, ARUP Laboratories completed a new, state-of-the-art 220,000-square-foot laboratory facility that spans four floors — the result of a substantial investment that ARUP Laboratories made, in reliance on the existing regulatory framework, to expand its ability to provide the highest quality services. The facility is designed to optimize quality laboratory testing, featuring world-class automation to provide efficient, large-scale laboratory operations with the ability to reconfigure quickly. In addition to other improvements, the facility includes new space for ARUP’s mass spectrometry laboratory, its automated core chemistry and immunology laboratories, and its specimen processing teams.

#### Professional Laboratory-Developed Testing Services

23. Professional laboratory-developed testing services are used by licensed practitioners in making health care decisions with their patients by providing diagnostic and other information that is used to monitor patients; influence medical, surgical, dietary, and other potential patient interventions; and inform future clinical advancements.

24. The testing services developed and performed by clinical laboratories have played a critical role in helping to diagnose and treat patients and are often at the forefront of innovation, particularly in academic and university clinical laboratory settings. For example, testing services that employ molecular diagnostics are routinely used in the diagnosis of malignancy, in the identification of genetic variants that suggest additional therapeutic interventions, in the characterization of genetic variants found in inheritable diseases, and in the diagnosis and treatment of infectious diseases. In many of these cases, no FDA-cleared or -approved medical devices are available, and physicians and patients rely on the professional testing services provided by laboratories to meet otherwise unmet needs, commonly when that

25. Laboratory-developed tests are frequently developed in academic clinical laboratories and in reference (e.g., referral) laboratories. For reference laboratories, requests for access to esoteric tests and services to diagnose rare disorders are relatively common, as specimens are received from clinics and hospital facilities extending over wider geographic areas or networks. Given the high costs of obtaining premarket approval or clearance from the FDA, as well as the limited financial incentives for medical device manufacturers to develop esoteric tests or tests for rare diseases, reference laboratories address unmet clinical needs by developing and offering professional testing services that are performed by clinical laboratory professionals in a single laboratory location.

26. As experts in clinical laboratory testing operations, clinical pathologists, doctoral-level clinical laboratory scientists, and laboratory personnel become aware of the strengths and limitations of different assays and testing platforms. This awareness comes from direct experience with assay and instrument operation, as well as peer-to-peer information sharing within the clinical laboratory community, scientific literature, and national and international conferences. When needed and appropriate, experts exercise their professional judgment in seeking to modify, update, and validate testing procedures to address specialty care needs.

27. Test modifications are particularly important in clinical laboratory settings. For example, important issues include alternative specimen types (e.g., when the specimen type listed in an FDA-approved test is not the appropriate matrix for clinical evaluation), extension of specimen stability parameters, or automation of otherwise manual tests to improve throughput, quality, and efficiency of testing and to minimize risks of repetitive motion injuries to laboratory professionals. There are also times when modifying a test is necessary to adapt to specific patient needs or to adapt to urgent reagent shortages. Under CLIA’s regulatory framework, the medical director of each laboratory is responsible for exercising professional judgment in deciding when to modify testing procedures to ensure that clinical testing is conducted appropriately, including validating the acceptability of the specimen used for testing. Under the FDA’s final rule, many such modifications would require FDA premarket review, thus delaying and/or preventing service improvements to meet clinical and public health needs.

28. ARUP Laboratories provides a wide range of professional testing services to assist hospitals and health centers in deciding how best to diagnose and treat patients.

29. **Genetic testing.** ARUP’s Institute for Clinical and Experimental Pathology was founded to foster the academic research of ARUP’s medical directors, while also advancing the science of diagnostic laboratory medicine to improve patient care.

30. Under the leadership of its professional medical directors, ARUP Laboratories was one of the first laboratories to use DNA sequencing as part of the routine testing services provided to hospitals and health centers. ARUP Laboratories leveraged DNA sequencing technologies developed by Frederick Sanger that emerged in the 1980s and ’90s and, more recently, ARUP has relied on next generation sequencing (known as massively parallel sequencing) technologies, which can be used to rapidly sequence whole genomes.

31. Building on these technological advances, many of ARUP’s laboratory-developed genetic tests have resulted in successful innovations in the diagnosis and treatment of diseases.

32. One of ARUP’s first genetic sequencing tests was a quantitative hepatitis C virus assay. Another notable test is ARUP’s genetic panel for myeloid malignancy variants, which at the time it was developed in 2014, was one of the first tests available to detect and treat patients with diseases resulting from stem cell variants. In addition, ARUP Laboratories has developed rapid molecular tests for certain immunodeficiency disorders that use state of the art tools, such as next generation DNA sequencing, to find new
causes of primary immunodeficiencies, which can present from the newborn period until mid-to-late adulthood. ARUP Laboratories was also the first clinical reference laboratory to offer unique assays that measure the ability of a drug to inhibit tumor necrosis factor ("TNF") and detect the presence of antibodies that neutralize TNF antagonist drug activity which can lead to treatment failure.

33. **Clinical toxicology.** More than a decade ago, ARUP Laboratories introduced a new and innovative approach to panel-based drug testing that focused on improving efficiency and specificity of results, and at the same time, reducing costs. Instead of using exclusively immunoassays to screen specimens, ARUP Laboratories developed testing procedures that take advantage of the benefits of mass spectrometry. An immunoassay screen detects the presence of a targeted compound or similar compound by the signal that changes when the compound (or compounds) react with specifically formulated reagents. In contrast, mass spectrometry identifies each targeted compound that is present in a specimen individually based on mass-to-charge ratio, in combination with other unique chemical and physical characteristics. Tests that employ mass spectrometry can yield results with higher specificity than immunoassays, which translates to a lower risk of false-negative and false-positive results.

34. ARUP Laboratories is proud of the work that it does to support clients across the nation in cases of unknown drug exposures. For example, ARUP Laboratories developed a mass spectrometry test that targets detection of over 100 different compounds cited in data from the American Association of Poison Control Centers on the most common accidental exposures.

35. There are many ways clinical toxicology testing and therapeutic drug monitoring can be used to help patients. For example, laboratory-developed tests are often used to help cancer patients who are receiving different forms of chemotherapy. Similarly, laboratory testing is important when a patient receives a kidney transplant and is required to take immunosuppressant drugs to reduce the risk of rejecting the new organ. Therapeutic drug monitoring allows the physician to calibrate dosing correctly and help the patient avoid harmful side effects, such as infection. Laboratory-developed testing is also important when identifying and monitoring trace and toxic elements with industrial exposures.

36. **Pharmacogenomics.** ARUP Laboratories has made significant advances in the field of pharmacogenomics, an emerging medical specialty that employs genetic and phenotype testing to predict or explain patient response to certain medications. Laboratory-developed testing in this area can be used to guide selection of drug options and doses for individual patients while avoiding adverse drug effects.

37. As with many laboratory testing services, choosing the appropriate test and interpreting the results can be complicated and requires professional expertise. As a result, ARUP Laboratories often assists physicians in understanding the complexities associated with testing, including pharmacogenomics.

38. **Maternal / pediatric health.** Another area where ARUP Laboratories has advanced healthcare is in the areas of maternal and pediatric health. In parallel with the opioid epidemic, there has been a significant increase in newborns experiencing neonatal abstinence syndrome. The professionals at ARUP Laboratories spent years developing a new mass spectrometry test, using either meconium or umbilical cord tissue as the specimen type, that identifies almost 50 different types of compounds and can be used to assess in utero drug exposure.

39. ARUP Laboratories has collaborated with hospital delivery units as well as representatives from children and family services agencies across the nation to understand when new drugs should be added to its screening panels. For example, collaboration with clinicians and caregivers led ARUP to be among the first clinical laboratories to add gabapentin to its umbilical cord drug screening panel. That drug is often prescribed as an alternative to opioids or in combination with opioids for pain management but has been increasingly

40. **ARUP Laboratories’ expertise in newborn drug testing** is having a positive impact on the development of public health programs. The close collaboration between ARUP’s medical directors and professors at the University of Utah Department of Obstetrics and Gynecology has allowed researchers to identify the prevalence and trends in drug exposures. From a public health standpoint, this research has allowed more targeted interventions, including educating clinicians to talk with patients about substance use and pregnancy in affected regions and finding ways to link patients to multidisciplinary care and addiction services. Data collected has also been used in a successful grant application for resources to reduce morbidity and mortality from substance use disorders during pregnancy in Utah.

41. **The Penelope Program.** ARUP Laboratories is a key partner in the “Penelope Program” at University of Utah Health, which is a collaboration between ARUP Laboratories, the Department of Pediatrics, the Department of Human Genetics, the Utah Center for Genetic Discovery, the Center for Genomic Medicine, and the Primary Children’s Hospital.

42. The program was launched in 2015 to address the challenge of undiagnosed diseases, recognizing that families often spend years — if not decades — searching for answers to the unidentified illnesses affecting their children. The program brings together a team of experienced clinicians from multiple specialties, molecular geneticists, data scientists, and researchers who pool their knowledge and expertise to help evaluate pediatric patients from different angles and perspectives, to identify potential diagnoses, and to develop an appropriate diagnostic plan. The team has access to advanced technologies and diagnostic tools to look for diagnoses that may have been missed. New testing, such as whole genome sequencing and RNA sequencing, can unmask genetic causes hidden in the depths of a complex genome. The program is also actively engaged in reducing disparities in access to advanced diagnostics.

#### FDA’s Final Rule Threatens Patient Health

43. The FDA’s final rule improperly treats laboratory-developed tests as if they are manufactured medical devices. The rule reveals the FDA’s fundamental misunderstanding of how laboratories perform professional testing services.

44. Laboratories are not manufacturers. And the tests they perform are not medical devices or other types of equipment. Instead, a laboratory-developed test is a professional service that reflects a series of procedures, medical protocols, and processes involved in analyzing tissue, blood, and other specimens as part of the practice of laboratory medicine. Those processes are validated and overseen by experts who must exercise informed clinical judgment in assembling the technical steps involved in conducting a test, understand how those steps interact, and determine how data should be interpreted.

45. Mass spectrometers and other manufactured equipment used by healthcare professionals are only tools used in performing a laboratory-developed test. The test itself entails procedures, methodologies, and processes that do not qualify as instruments, apparatuses, machines, contrivances, implants, in vitro reagents, or other related articles subject to FDA regulation. When laboratory clinicians develop the processes and procedures necessary to perform laboratory-developed testing services, they are no different than other health care professionals who develop protocols or methodologies for treating patients or diagnosing diseases.

46. I am concerned that FDA’s final rule takes the position that laboratory tests not cleared or approved by FDA are illegal and that laboratories have been violating the law for decades. Although the preamble to FDA’s final rule says that the agency intends to exercise enforcement discretion — to allow laboratory-developed tests to remain on the market until it decides otherwise — the notion that laboratories and the professionals who run them are all engaged in unlawful conduct is absurd and, in my mind, shows...

that the FDA itself is not acting reasonably and within the scope of any lawful authority granted by Congress.

47. I am also deeply concerned that, because the FDA lacks the resources to oversee tens of thousands of laboratory-developed tests, laboratories will face significant regulatory uncertainty and patients will face the risk of being denied access to the essential medical services they depend on clinical laboratories to provide. In short, the FDA’s attempt to regulate laboratory testing services as medical device products will undermine the provision of health care and stifle innovation in a critical sector of our health care ecosystem.

48. FDA clearance and/or approval requirements will also have significant negative consequences for the innovation that occurs when professional laboratory clinicians modify existing testing procedures, or tailor them to address unmet patient needs.

49. Laboratory-developed tests can be modified to address a patient’s specific circumstances, which can lead to the discovery of new and improved diagnostic approaches and testing protocols. This ability to innovate is likely to be curtailed under the FDA’s new rule, which is likely to result in treating many of these modifications as creating a new “test” subject to separate FDA premarket review. This appears to also apply to minor modifications, such as adding manual immunoassays or PCR-based assays to simple liquid handlers, for example. Through these requirements, quality improvements through automation are paradoxically disincentivized by the FDA.

50. Under the final rule, ARUP Laboratories and other clinical laboratories will have to devote significant resources to developing FDA-centric quality system processes and adhering to device submission requirements, including premarket submissions for modified and new laboratory-developed tests. But there are not enough laboratory professionals to support compliance with FDA’s final rule while maintaining current testing levels.

51. Patient access to innovative tests will also be harmed because device regulation is likely to cause an FDA-review bottleneck going forward. That can only slow patient access to innovative tests as a result of extended review times, inadequate FDA resources to engage with applicants and developers, and clinically beneficial tests that are discontinued by laboratories due to excessive compliance costs over time.

52. FDA’s final rule appears to recognize this concern by suggesting that it will not enforce certain requirements of federal law against laboratories that comply with New York requirements for laboratory testing. But that does not change the reality that tests not approved or cleared by FDA will be considered unlawful, and that FDA has made clear that it could change its enforcement guidelines at any time.

53. Even with the enforcement discretion announced in the final rule, the number of future premarket approval applications for laboratory-developed testing services will likely increase significantly. The FDA lacks the resources to deal effectively with these submissions, and, even if the FDA had more resources, there are not enough trained scientists and regulatory professionals for it to hire. The FDA will be competing with laboratories that also would need to increase hiring of the same professionals to deal with the new regulatory system.

54. The FDA’s final rule also vastly overestimates the benefits of regulating professional testing services as the equivalent of medical devices. In suggesting that a large percentage of errors are attributable to laboratory tests, the FDA misapplies a study and reaches conclusions that are inconsistent with the diagnostic literature. A review of that literature suggests that only one to four percent of diagnostic errors may be attributable to faulty test results. As a result, and as explained in more detail in ARUP’s public comment letter, it appears that in its initial regulatory impact analysis the “FDA has made, at a minimum, an approximately 250-fold overestimate in its assessment of financial benefit,” failing to consider relevant data and applying only superficial assumptions. Unfortunately, it carries erroneous assumptions into calculations used in its final regulatory impact analysis, therefore still overestimating the purported financial benefits to society in the final rule. For
example, while the FDA cited our 2023 research manuscript [Rychert et al. Am J Clin Pathol. 2023. 160(3):297-302] in its final regulatory impact analysis when describing the percentage of clinician test orders that are laboratory-developed tests (3.9%), it more than doubled this percentage in its revised calculations of financial benefit, claiming that our data was based on "information for one single laboratory." This arbitrary increase in percentage, however, is in direct contradiction to the discussion of results in our manuscript, which states that "the presence of a national reference laboratory as part of the university health system may also have contributed to more LDTs being available for ordering than at other institutions. If this were the case, the present study’s finding may overrepresent LDT orders vs those placed at other institutions." I therefore believe that the FDA has misapplied our study findings in its final regulatory impact analysis in a manner that overestimates the financial benefit to society.

55. The FDA’s final rule understates the costs of treating professional laboratory services the same as medical devices. The rule is particularly problematic because it assumes — without sufficient or adequate analysis — that added FDA oversight would improve the safety and effectiveness of laboratory-developed tests, and yet it fails to take into account the reductions in access to safe testing that will occur if FDA’s final rule remains in place.

56. The FDA’s final rule will reduce access to safe testing because the staggering costs of seeking FDA approval threatens to force clinical laboratories over time to reduce the range of testing services they provide. In turn, that will disproportionally affect patients with rare diseases, underserved populations, patients with cancer, and children. For many laboratory-developed tests, market prices would increase due to reduced competition, and patients might lose timely access to diagnostic and treatment.

57. Hospitals and health centers trust the testing services provided by ARUP Laboratories because of our decades of experience in developing and performing tests, using them in our laboratories, and using them in consultation with expert clinicians for the care of their patients. Nearly all of our customer health systems are under different corporate ownership than ARUP Laboratories. Therefore, as written, the FDA’s unmet needs exemptions in the final rule do not apply in our setting, and ongoing development of testing for unmet needs for the patients we serve will be more difficult under the final rule.

58. Under the FDA’s final rule, new, and many modified existing, laboratory-developed tests would need to be submitted to the FDA for premarket review. But it is unrealistic to expect laboratories to be able to afford the massive costs involved in obtaining FDA clearance or approval of every test that they might develop in the future. Many laboratory-developed tests are low-volume tests that are used infrequently. These are clinically essential tests for patients suffering from rare diseases or difficult-to-diagnose conditions, and they are often essential in developing effective treatment for patients who have unmet medical needs. But these types of tests often fail to generate sufficient revenues to justify going through the very expensive FDA-clearance or approval process, which is why testing is often performed in reference settings.

59. ARUP Laboratories is concerned about the negative impact of new regulatory requirements and premarket reviews on the clinical laboratory industry. ARUP Laboratories is also concerned that it may not be cost effective for many clinical laboratories to continue to develop new, low-volume tests that can be used to diagnose or monitor rare diseases, particularly given the restrictions on testing for unmet needs in patients seen in facilities outside a laboratory’s corporate ownership.

60. Diverting resources away from helping patients and toward seeking FDA clearance or approval of testing services that FDA lacks the resources and expertise to evaluate would not benefit patients or the public interest. Indeed, the most alarming consequence of FDA’s rule is to declare all existing laboratory-developed tests to be unlawful and to make it more difficult for patients to continue to obtain the essential testing services they need.

In accordance with 28 U.S.C. § 1746, I declare under penalty of perjury that the foregoing is true and correct.

Executed on this ___23___ day of May, 2024.

By:  
 
Jonathan Genzen, M.D., Ph.D.

## Exhibit E

AMERICAN CLINICAL  
LABORATORY ASSOCIATION;  
HEALTHTRACKRX INDIANA,  
INC.; and HEALTHTRACKRX,  
INC.,  

Plaintiffs,  

v.  

U.S. FOOD AND DRUG  
ADMINISTRATION; U.S.  
DEPARTMENT OF HEALTH AND  
HUMAN SERVICES; XAVIER  
BECERRA, in his official capacity as  
Secretary of Health and Human  
Services; and ROBERT M. CALIFF,  
M.D., in his official capacity as  
Commissioner of Food and Drugs,  
United States Food and Drug  
Administration,  

Defendants.  

**DECLARATION**

### DECLARATION OF WILLIAM MORICE II, M.D., Ph.D.

I, William Morice II, M.D., Ph.D., declare as follows:

1. I am a resident of Rochester, Minnesota. I am over the age of eighteen, and I am competent to provide this declaration.

2. I have worked for Mayo Clinic for more than 24 years, including in a variety of leadership positions.

3. Since 2015, I have served as the President of Mayo Clinic Laboratories. Since 2023, I have served as the Chief Executive Officer and President of Mayo Collaborative Services, which includes Mayo Clinic Laboratories. Mayo Clinic Laboratories is Mayo Clinic’s outreach reference laboratory service for external customers.

4. From 2015 to 2022, I served as the Chair of Mayo Clinic’s Department of Laboratory Medicine and Pathology, Rochester, Minnesota, and I have long been affiliated with the Department’s Division of Hematopathology, serving as a physician consultant on Mayo Clinic medical staff since 2000.

5. I am a Professor of Laboratory Medicine & Pathology at the Mayo Clinic College of Medicine.

6. I also serve on the College of American Pathologists, the leading accreditation organization for clinical laboratories under the program established by the Clinical Laboratory Improvement Amendments of 1988 (“CLIA”).

7. Since 2017, I have served as a member of the Research and Innovation Board for the Brussels Academic Hospitals Laboratory, which is an international group of experts from different institutions dedicated to identifying areas of worldwide need for diagnostic services. I am also a founding member of the Bone Marrow Pathology Group, which is a group of academic pathologists focused on clinical and laboratory research to identify and address diagnostic issues associated with bone cancer.

8. I earned my Bachelor of Science degree in biochemistry from Indiana University, Bloomington, as well as combined medical and doctoral degrees in biomedical sciences and immunology from the Mayo Clinic College of Medicine.

9. As a Mayo Clinic faculty member, my research has focused on the diagnoses of blood and bone marrow-derived cancers, including lymphoproliferative disorders of T cells and natural killer (NK) cells, plasma cells, and B cells. I have studied these disorders using a variety of methods, with a particular emphasis on flow cytometry, and with the aim to develop practical applications for clinical diagnosis.

10. I am a member of the Board of Directors of the American Clinical Laboratory Association (“ACLA”), and I am currently serving as the Board’s Chair. ACLA is the national trade association representing leading laboratories that deliver essential diagnostic health information to patients and providers by advocating for policies that expand access to the highest quality clinical laboratory services, improve patient outcomes, and advance the next generation of personalized healthcare.

11. As a result of my professional experiences and leadership roles, I have significant knowledge of clinical diagnostic laboratory services. I engage frequently with medical device manufacturers, including manufacturers of in vitro diagnostic (“IVD”) test kits, and understand the substantial and meaningful differences between medical device products and laboratory-developed testing services. I am also familiar with the significant time and expense involved in seeking federal approval to market and sell medical devices to third parties.

12. I am generally familiar with the legal and regulatory requirements that have long applied to laboratory-developed tests, including the CLIA requirements that regulate the conduct of laboratories and the validation of laboratory diagnostic testing services. I am familiar with the state regulations for all states in which Mayo Clinic Laboratories does business. I am also familiar with the final rule issued by the Food and Drug Administration (“FDA”) on May 6, 2024.

13. As described below, I am deeply concerned about FDA’s final rule. FDA’s position — that laboratory-developed tests are and have always been subject to all of the requirements of the federal Food, Drug, and Cosmetic Act, including premarket approval requirements for medical devices — is an implausible position that is contrary to decades of settled expectations in the healthcare profession.

14. FDA’s final rule fails to appreciate that laboratory-developed tests are not manufactured products or devices; instead, they are services provided.
by trained laboratory and medical professionals. When laboratorians develop a methodology to perform laboratory-developed testing services, they are no different from other health care professionals that develop protocols or methodologies for treating patients or diagnosing diseases.

15. FDA lacks the expertise to regulate the important professional healthcare services that clinical laboratories provide, and the radical change in regulation that the agency seeks to impose will stymie innovation and undermine laboratories’ ability to help diagnose and treat diseases, especially those that are rare.

16. I am also deeply concerned that, because FDA lacks the resources to oversee the prompt approval of laboratory-developed tests, laboratories will face significant regulatory uncertainty and patients may be denied access to the essential medical services they depend on clinical laboratories to provide. In short, FDA’s choice to regulate laboratory testing services as medical device products will undermine the provision of health care and stifle innovation in a critical sector of our health care ecosystem.

#### Mayo Clinic Laboratories

17. Mayo Clinic is the largest integrated, not-for-profit medical group practice in the world. It is committed to clinical practice, education, and research, and its experts work together to diagnose and treat the toughest medical challenges and unmet needs of patients. It has more top rankings for high-quality patient care than any other healthcare organization.

18. Mayo Clinic’s unwavering drive to create better medical care and its relentless pursuit of research helps patients by making earlier diagnoses possible and developing new cures. More than 1.3 million patients, from nearly 130 counties, visit Mayo Clinic campuses each year.

19. Mayo Clinic physicians are supported by the Department of Laboratory Medicine and Pathology, which performs testing services for diagnostic and therapeutic evaluations. More than 2,350 employees, working in numerous specialty laboratories, staff the Department. They form one of the largest clinical laboratories in the world, which performs more than 26 million tests annually. The laboratory’s professional teams consist of highly experienced physicians, scientists, medical technologists, medical technicians, histotechnologists, cytology technologists, pathologist’s assistants, phlebotomists, lab assistants, biologists, chemists, microbiologists, geneticists, genetic counselors, and other specialists.

20. Mayo Clinic’s most extensive laboratory service facilities are located in Minnesota and are provided by more than 190 physicians and Ph.D. scientists. Besides serving Mayo Clinic patients, the laboratories also undertake testing for clinics and hospitals, both in the United States and in more than 50 countries worldwide.

21. Mayo Clinic’s laboratory specialty areas include anatomic pathology, clinical biochemistry and immunology, clinical core laboratory services, clinical microbiology, community laboratory medicine and pathology, dermatopathology, experimental pathology and laboratory medicine, hematopathology, laboratory genetics, and transfusion medicine.

22. Mayo Clinic Laboratories is a reference laboratory that specializes in esoteric laboratory testing — testing for the most unusual, complex, and difficult cases. The laboratory was launched more than 50 years ago, in March 1971 and has always operated as a highly specialized laboratory housed within an academic medical center.

23. As part of providing medical diagnostic services to patients, Mayo Clinic Laboratories has always reflected Mayo Clinic’s core mission — what are referred to as the “three shields”: (1) integrated clinical practice, (2) research, and (3) education.

24. Mayo Clinic Laboratories has partnered since its inception with Mayo Clinic and its high-caliber faculty and state-of-the-art laboratories to help patients. For example, Mayo Clinic Laboratories has long provided second opinions on surgical pathology. When a pathologist or physician at another hospital, health system, or practice has a difficult diagnosis, they call Mayo Clinic Laboratories, which works with them and connects them to a physician or surgeon from Mayo Clinic to discuss the case. Research has also long been

a distinguishing mark of Mayo Clinic Laboratories, particularly in the context of developing new diagnostic tests and other groundbreaking medical services. In addition, Mayo Clinic Laboratories is well known for its educational programs offered to other healthcare providers on topics such as endocrine diseases, infectious diseases, surgical pathology, and liver pathology.

25. Fueled by the caliber of Mayo Clinic’s academic medical faculty and its state-of-the-art laboratories, Mayo Clinic Laboratories is an internationally respected laboratory that provides important and cutting-edge testing services to help diagnose and treat patients with some of the most complex conditions.

26. Mayo Clinic Laboratories currently serves more than 3,400 national and international clients and, in ordinary circumstances, performs approximately 35,000 testing services per day. Together with Mayo Clinic, Mayo Clinic Laboratories relentlessly innovates on behalf of patients, building the diagnostics ecosystem of the future to help physicians save and improve more lives.

#### Laboratory-Developed Testing Services

27. When a patient visits Mayo Clinic with a difficult or complex medical challenge, or when its physicians and scientists are asked to consult with physicians outside of Mayo Clinic, the patient’s caregivers will turn to Mayo Clinic Laboratories and its expert laboratorians to help diagnose and evaluate the patient. In these situations, existing or routine tests — including available FDA-cleared or approved tests may be incapable of providing the information required. Instead, Mayo Clinic physicians and scientists have to rely on combining medical expertise with research and diagnostic expertise to ensure that it can properly evaluate and analyze the patient’s condition and then use that evaluation to diagnose the patient and develop effective treatment options.

28. Mayo Clinic Laboratories follows a rigorous process when developing new testing services. Laboratory teams focus on developing new tests when there is an unmet patient need and when it appears that new testing services could improve patient diagnosis, treatment, or care. Because the laboratory that develops a test must be knowledgeable about how the test will be used and understand the specimen type, range of detection, turnaround time requirements, and potential complementary tests, Mayo Clinic Laboratories has a large team of highly experienced physicians, scientists, medical technologists, medical technicians, histotechnologists, cytology technologists, pathologist’s assistants, phlebotomists, lab assistants, biologists, chemists, microbiologists, geneticists, genetic counselors, and others.

29. When developing the necessary protocols and methodologies for a new test, interdisciplinary experts and specialists consult with other medical experts, educators, and researchers at Mayo Clinic. Developing a new test requires a research and development process that documents standard operating procedures for each assay and includes basic analytical validation studies to show that assays satisfy clinical requirements. In addition, a variety of quality and process controls must be in place to verify that all steps of the assay are working appropriately, including extraction controls, controls to assess clinical accuracy, hybridization controls, external analyte controls, and/or internal controls for analytes expected to be presented in each test sample.

30. There are many examples of areas where Mayo Clinic Laboratories has developed groundbreaking approaches to providing the types of professional diagnostic services that are essential to helping patients. A few examples follow:

31. **Cardiovascular testing.** Mayo Clinic Laboratories has for more than two decades been at the forefront of cardiovascular genetic testing. The Laboratory’s current test menu features 24 different panels that collectively include more than 300 genes linked to inherited cardiovascular disorders, many of which are rare and challenging to diagnose. Disorders can involve multiple genes and may present similarly to other disorders with distinct genetic causes.
32. When providing genetic specimen or similar testing services, Mayo Clinic Laboratories takes an expansive approach. A team of technicians, genetic counselors, Ph.D. geneticists, and physicians work together to analyze the genetic data and determine how the data correlates with the clinical picture of each individual patient. This correlation entails both review of each patient’s family history and medical records and discussions with the treating clinicians.

33. There is significant professional medical expertise involved in developing well-thought-out gene panels to ensure that they are clinically useful and valid. In addition, professional expertise is required to interpret the results to ensure that the findings are accurate and consistent with the patient’s overall clinical presentation. Mayo Clinic Laboratories’ testing is part of Mayo Clinic’s integrated clinical practice.

34. For example, I am aware of the services Mayo Clinic Laboratories provided to a Mayo Clinic patient who was under the care of a preventative cardiologist. The patient had both high cholesterol and stage 1 hypertension, but current risk calculators and traditional blood tests were not helpful in reaching an appropriate diagnosis because they did not take into account the patient’s family history and because coronary artery disease and endothelial plaque formation are complex and multifactorial processes. As a result, the patient’s treating cardiologist recommended to the patient that a sample be sent to Mayo Clinic Laboratories for specialized ceramide testing. Ceramide testing uses liquid chromatography-mass spectrometry technology, which sorts through complex molecular compounds. By testing patients, the Laboratory is able to reveal more about heart disease and stroke risk than standard lipid tests, as ceramides are involved in plaque formation and can reflect inflammation, bad cholesterol, or coagulation and thickening of the blood. By evaluating the results of these tests, the treating physician was able to develop a tailored treatment plan for the patient that was appropriate to his risk level.

35. **Cancer testing.** Another important area where Mayo Clinic Laboratories is a leader is in the area of cancer diagnosis and treatment, where laboratory testing services are relied on to provide crucial information to guide appropriate care in real time. Because cancer is such a complex disease about which our medical understanding continues to evolve, it is important to be able to account for emerging new medical information on discoveries and technological advancements.

36. Mayo Clinic has one of the most robust laboratory and pathology practices for cancer care in the world, offering testing services that span diagnostic, theranostic, and prognostic approaches. The comprehensive test menu at Mayo Clinic Laboratories includes numerous laboratory-developed tests that aid in providing both diagnostic and prognostic information and treatment selection guidance across the full spectrum of malignancies.

37. Mayo Clinic Laboratories has developed a suite of hematology and oncology, next-generation sequencing panels that are used to help diagnose patients depending on the genetic associations of an individual patient’s cancer. Mayo Clinic Laboratories uses these panels to evaluate several gene mutations, rearrangements, and amplifications. The results of these analyses provide information that, along with correlation with the patient’s clinical features, inform clinicians regarding the appropriate diagnosis and are used to guide treatment. This all helps to ensure that patients receive the best cancer care and are able to follow the best treatment strategies.

38. The laboratory-developed tests developed by Mayo Clinic Laboratories include those used to diagnose sarcoma, melanoma, lung, and colorectal cancers, including gastrointestinal stromal tumors. New panels have recently been developed to help evaluate patients for renal cell carcinoma, bladder and prostate cancer, gynecological cancer, endometrial cancer, and ovarian, fallopian tube, and peritoneal cancer. In addition, panels have been developed to test for B-cell lymphoma, chronic lymphocytic leukemia, T-cell lymphoma, plasma cell myeloma, and histiocytic neoplasms.

39. These tests are not just a question of chemistry. They involve analyzing actionable genes that are carefully selected by a multi-disciplinary team of professional clinicians, geneticists, genetic counselors, and laboratory testing experts. The genes included on the testing panels are recognized as
clinically significant and are included in consensus group testing guidelines put forth by the World Health Organization, the National Comprehensive Cancer Network, the Intercultural Cancer Council, and European LeukemiaNet, among other governing bodies.

40. Because the experts who design the hematology panels are themselves hematopathologists, they are intimately familiar with the underlying disease. When questions arise, testing specialists collaborate with Mayo Clinic physicians to obtain their expert assistance with accurately interpreting the results. Moreover, each and every case is reviewed by a team of genetic experts, including technical experts, genetic counselors, and physician scientists, all of whom help to interpret the test results and ensure that they are accurate.

41. The interpretive report that accompanies the testing results describes in detail the gene variant detected and contextualizes the clinical significance from a diagnostic, prognostic, and/or therapeutic perspective. The report also may include information concerning relevant clinical trials. Even after the clinical report is provided to the treating physician(s), ongoing consultation between the laboratory professionals and the treatment team is common.

42. I was personally involved in developing standards for classifying and identifying different types of leukemia and, in particular, T cell, large granular, lymphocytic leukemia and NK-cell, large granular, lymphocytic leukemia, which are rare blood cancers of cytotoxic effector lymphocytes. Diagnosis involves evaluating blood and bone marrow samples for clonal abnormal T- and NK-cells through a variety of methods, all of which are synthesized by the pathologist and correlated with the clinical features to render the correct diagnosis.

43. Early diagnosis of this rare disease is often crucial for appropriate management and optimizing the outcomes for patients. That is especially true because patients with T cell and NK cell disorders often present with symptoms that are similar to other diseases and other types of leukemia. Early detection can ensure that the disease is properly treated and allow patients to live long and fulfilling lives.

44. The standards I helped develop to support the practice of medicine in this important area of pathology are reported in guidelines published by the World Health Organization and are part of the standard of care for classification of patients with this rare disease. _See_ Rita Alaggio, et al., _The 5th Edition of the World Health Organization Classification of Haematolymphoid Tumours: Lymphoid Neoplasms_. Leukemia. 2022 Jul;36(7) 1720-1748. doi: 10.1038/s41375-022-01620-2. Epub 2022 Jun 22. Erratum in: Leukemia. 2023 Sep;37(9):1944-1951. PMID: 35732829; PMCID: PMC9214472.

45. As this publication recognizes, evidence-based classification of disease is fundamental for the treatment of individual patients, monitoring global disease incidence, and investigating all aspects of disease causation, prevention, and therapy.

46. **Infectious diseases.** Mayo Clinic Laboratories is also internationally renowned for offering a broad selection of laboratory-developed tests designed for the rapid identification and in-depth characterization of the causative agents of infectious diseases. The Laboratory has developed culture techniques, immunoassays, antimicrobial susceptibility testing, and molecular methods for rapid detection, identification, and characterization, as well as providing treatment guidance for microbial pathogens.

47. For example, Mayo Clinic Laboratories has developed a test that assists with identifying nervous system infections. The assay uses shotgun metagenomic sequencing to apply complex bioinformatic analysis to all DNA and RNA sequences in a cerebrospinal fluid specimen. The sequences deriving from humans are removed, and the remaining sequences are then analyzed by expert technicians for a potentially pathogenic microorganism.

#### The Material Differences Between Laboratory-Developed Testing Services and FDA-Approved Medical Devices

48. Laboratory-developed testing services are materially different from the types of instruments, machines, or physical objects that qualify as and are properly considered to be medical devices.

49. Laboratory-developed testing services are essential professional healthcare services that are provided in connection with the practice of medicine. The types of tests developed by federally-certified CLIA laboratories reflect unique assays that yield important clinical information about a patient that can be used by physicians and other medical professionals to inform and guide that patient’s care and treatment.

50. When Mayo Clinic Laboratories develops and performs laboratory-developed testing services, it is not acting as a device manufacturer. Mayo Clinic Laboratories are managed by professional physicians and scientists with expert knowledge regarding the clinical implications of each test result and how those results can impact patient care. The type of interaction and collaboration between a clinical laboratory and the care team that I describe above is completely at odds with the idea of a laboratory as a device manufacturer. Medical device manufacturers do not perform patient care using their medical devices; medical device manufacturers do not collaborate with treating physicians; medical device manufacturers are not considered part of the care team for a patient.

51. By subjecting laboratories to regulation as device manufacturers, FDA’s final rule will significantly undermine the relationship between laboratory professionals and the care teams that rely on the diagnostic services of clinical laboratories.

52. Performing a test in a CLIA-certified laboratory involves (1) collecting and appropriately preparing a tissue, blood, or other patient specimen that is sent to the laboratory for testing and analysis; (2) conducting an analysis of the sample by using reagents, instrumentation and other equipment to evaluate the specimen quantitatively and/or qualitatively; and (3) preparing a report that reflects the results of the analysis, the expert medical judgment used in conducting the analysis, and patient-specific information that is shared with a physician or team of physicians to assist in patient diagnosis and treatment.

53. FDA-approved medical devices, including reagents, instrumentation, microscopes, spectrometers, and other sophisticated types of medical equipment, are routinely used by laboratory technicians when providing diagnostic testing services. Medical devices are appropriately subject to FDA’s clearance and pre-approval oversight because they are instruments, machines, or other physical objects that are packaged, labeled, and produced by manufacturers to be used and sold to third parties.

54. Unlike medical devices, laboratory testing services are not a type of equipment, machine, or physical object. They instead reflect medical protocols, information, procedures, and techniques that are used by healthcare professionals to evaluate and diagnose patients. Whereas medical devices are typically sold to healthcare providers or patients for their own use, laboratory-developed tests are not sold or shipped to third-party purchasers. Instead, these testing services are performed by the laboratory itself, and the material to be tested — human blood, tissue, or other specimen samples — is sent to the laboratory for its expert medical analysis. Professional laboratory-developed testing services are performed by expert laboratory technicians in consultation with medical experts and are integral to the broader practice of medicine.

55. To provide a simple analogy: When a patient visits his doctor, the doctor may use certain devices to perform medical procedures. The doctor might, for example, use a biopsy needle to obtain a tissue sample for subsequent analysis. The biopsy needle is an FDA-regulated medical device and must be cleared or approved by FDA before it can be obtained and used by the doctor. In contrast, the technique the doctor follows in performing the biopsy procedure and the information she deems relevant to evaluating her patient — for example, deciding where to place the needle to avoid sensitive 

tissue and whether or how to use anesthetics or ultrasound guidance — is a healthcare service that is provided to the patient by a trained professional in connection with the traditional practice of medicine. The fact that the biopsy procedure uses a device does not transform the procedure itself into a medical device.

56. The same is true for professional laboratory-developed testing services. The equipment, machines, and other instruments used to analyze a blood, tissue, or other sample from a patient are medical devices — like an ultrasound or biopsy needle — that must be approved or cleared by FDA and qualified and validated for use by the laboratory pursuant to the requirements of CLIA. But the sophisticated protocols that laboratories develop and use to test and evaluate the samples received from patients are not themselves instruments, machines, or pieces of equipment.

57. Laboratory-developed testing services have long been subject to regulation under CLIA, a specific statutory and regulatory framework tailored for clinical laboratories. In addition, several states have their own requirements relating to laboratory testing services and the practice of medicine. Most significantly, New York has developed an extensive set of diagnostic-specific, risk-based requirements that apply to laboratory-developed tests offered to New York state residents.

58. One key aspect of the existing regulatory requirements is the flexibility provided to the professionals at clinical laboratories to develop new testing protocols and modify existing ones for the purpose of diagnosing and guiding innovative treatment for patients. When individual patients present unexpected or unusual challenges, laboratory-developed tests are used in combination and may be modified to address the patient’s specific circumstances, which can lead to discovering new diagnostic approaches and new testing protocols.

59. I am deeply concerned about the consequences of FDA’s new rule for patients and for the ability of Mayo Clinic physician specialists to treat patients effectively. The most significant consequence of FDA’s rule will be to make it more difficult for patients to continue to obtain the essential testing services they need. Seeking FDA clearance and approval for medical devices is very expensive and time-consuming. According to FDA, laboratories will face enforcement action if they fail to obtain such clearance or approval for most new testing services and most significant modifications of existing testing services. And even where FDA says it will exercise enforcement discretion (such as for some existing services), the agency insists it can change that policy at any time.

60. If FDA’s final rule is allowed to take effect, clinical laboratories that want to protect themselves — and not rely on FDA’s open-ended exercise of enforcement discretion — will need to seek FDA clearance or approval for the tens of thousands of tests that are already on the market and for every new testing protocol they might develop in the future. That is unrealistic. Many laboratory-developed tests are low-volume tests that are used infrequently. These types of tests do not generate sufficient revenue to justify going through the very expensive FDA clearance or approval process.

61. Moreover, FDA lacks the personnel and financial resources necessary to efficiently review all of the testing services provided by professional laboratories and grant the necessary clearances and approvals. On average, it takes FDA months — and sometimes years — to clear or approve a new medical device. There are tens of thousands of different types of tests provided by laboratories across the United States, and there is no practical way for FDA to review, approve, and clear all of them.

62. It is also unclear how FDA intends to apply its new rule to modifications that are made to existing tests, which are critical to improving laboratory-developed tests and ensuring that they reflect the latest scientific advances. Under existing law, when FDA approves a medical device, the device must be manufactured to the approved specifications, and modifications to the device may need to be separately approved by FDA. As noted above, making modifications to existing testing protocols for particular patients is an essential part of the diagnostic testing process. Clinical laboratories, such as

Mayo Clinic Laboratories frequently make adjustments to tests to respond to the specific circumstances of individual patients or to incorporate new biomarkers to reflect the latest scientific advances. Preventing laboratories from making these modifications without first going through the time-consuming and expensive FDA clearance or approval process would harm both patient care and innovation.

63. The consequences of not having laboratory-developed testing services available would be particularly devastating to the quality of care for patients suffering from rare diseases, many of whom come to Mayo Clinic. One of the particular advantages of laboratory-developed tests is the ability to determine — through a process of elimination — whether a patient has biomarkers for a particular disease.

64. To the uninitiated, it may seem to be only a minor consequence if certain tests are not developed, as only a small number of patients may suffer from any particular rare disease. But that misunderstands how tests are used. Tests are often most useful not for determining that a patient has or may have a particular disease (“ruling in” that disease), but for determining what diseases the patient does not have (“ruling out” diseases). Furthermore, tests developed within the laboratory in response to clinical need often provide new insights into previously identified diseases. Through the implementation of laboratory-developed tests, patients with more or less aggressive forms of a disease are often identified, significantly impacting both their care and the psycho-social response to their diagnosis.

65. Eliminating diseases is often the best way to design an appropriate treatment regimen for a patient, even if no existing test is available to determine precisely what disease the patient does have. As a result, even tests that are used to diagnose especially rare diseases are an extremely valuable tool when diagnosing patients with unknown conditions.

66. Mayo Clinic has invested substantial resources into research and medical care — and into Mayo Clinic Laboratories — in reliance on the existing regulatory regime and on the understanding that Congress has never granted FDA authority to regulate professional laboratory testing services. These settled expectations would be undermined if FDA is permitted, for the first time ever, to take control over all the services provided by clinical laboratories by treating those services as equivalent to medical devices.

67. On several occasions, Congress has evaluated legislation that would change the regulatory requirements for laboratory testing services. But Congress has not enacted legislation and has instead declined to empower FDA with authority to regulate laboratory-developed tests. For many of us who are aware of those legislative debates, FDA’s new rule appears to be an attempt to circumvent the political process and exercise sweeping authority that has never been granted to the agency by Congress.

68. In any event, as a practical matter, Mayo Clinic Laboratories will have less money to invest in developing new and innovative testing services if its resources have to be dedicated to obtaining FDA clearance or approval. Moving resources away from helping patients and in the direction of seeking FDA clearance or approval of testing services that FDA lacks the resources and expertise to evaluate would not benefit patients or the public interest.

In accordance with 28 U.S.C. § 1746, I declare under penalty of perjury that the foregoing is true and correct.

Executed on this 28th day of May, 2024.

By:  
William Morice II, M.D., Ph.D.

## EXHIBIT F

**Privileged & Confidential**  
**Attorney Client Communication**  
**Pre-Decisional**

### MEMORANDUM

**TO:**  
Stephen Hahn, M.D., Commissioner of Food and Drugs

**CC:**  
Eric D. Hargan, Deputy Secretary  
Brian Harrison, Chief of Staff  
Stacy Amin, Deputy General Counsel & Chief Counsel  
Anand Shah, M.D., Deputy Commissioner of Food and Drugs  
Keagan Lenihan, Chief of Staff FDA  
Danielle Steele, Counselor to the Secretary

**FROM:**  
Robert Charrow, General Counsel

**SUBJECT:**  
Federal Authority to Regulate Laboratory Developed Tests

**DATE:**  
June 22, 2020

#### Introduction

We have been asked by departmental leadership to review the legal bases—both substantive and procedural—for FDA's regulation of laboratory developed tests ("LDT").[^1] This memorandum summarizes the results of our analyses.

Since 1992, FDA has taken the position in draft guidances, manuals, and web postings that LDTs are devices within the meaning of the Food, Drug, and Cosmetic Act ("FDCA" or "Act") § 201(h) and subject to the Agency’s jurisdiction. Most recently, FDA announced on its website that

- FDA generally has not enforced premarket review and other legal requirements [with respect to LDTs]. However, LDTs for which an HHS [Emergency Use Authorization] declaration justifies a need (and that potentially meet the EUA criteria) present a higher risk. This is because they are developed to diagnose serious or life-threatening diseases or conditions that not only have serious implications for individual patient care, but also for analyses of disease progression and public health decision-making. Thus, FDA requests that developers of such LDTs submit information about their tests to help FDA better understand their design, validation, and performance characteristics.

https://www.fda.gov/emergency-preparedness-and-response/mcm-legal-regulatory-and-policy-framework/information-laboratories-implementing-ivd-tests-under-eua (March 1, 2020) (last viewed June 15, 2020). We understand that some stakeholders, including many state university laboratories, have complained that this policy hindered their ability to develop and use LDTs to detect the virus that causes COVID-19.

We have undertaken a review of the relevant legal authorities and regulatory processes so as to advise departmental leadership, FDA, and other policymakers of FDA's authority in this area, especially in light of COVID-19. Specifically this memorandum addresses: (i) whether an LDT is a medical device; (ii) if so, under what circumstances, if any, does FDA have the jurisdiction to regulate LDTs; and (iii) whether FDA can properly regulate in this area without notice and comment rulemaking. We also discuss a potential re-assessment of relevant delegations in light of the foregoing analysis.

#### Summary of Conclusions

We believe that the Medical Device Amendments of 1976 ("MDA"), Pub. L. No. 94-295, may be broad enough, in certain settings, to accommodate FDA’s view that LDTs, as opposed to the procedures used to run those tests, are "devices," within the meaning of section 201(h) of the FDCA.[^2] However, the Agency’s jurisdiction to regulate these devices is not uniform and not as plenary as it is for a traditional device; this lack of jurisdictional uniformity is dictated by the FDCA itself. FDA relies on FDCA section 301(k) and the premarket review regime in sections 510(k) and 515 as the primary means of exercising authority over LDTs. This theory has several potential weaknesses. First, it appears likely that LDTs, even if they satisfy the "interstate commerce" requirement of the FDCA, would likely not satisfy the separate "commercial distribution" requirement of the premarket review provisions at sections 510(k) and 515.  Section 301(k), the primary provision dealing with prohibited acts, turns on whether the device is "held for sale." While courts in the past have given that term a broad reading to include devices that never leave a physician’s office, a plain meaning assessment may not be as agency-friendly. Second, many first-line sophisticated laboratories are operated by state public health departments or academic medical centers at large state universities. These laboratories, by definition, are not "persons," within the meaning of the Act, and not subject to many of the Act’s requirements, including registration (§ 510(o)), premarket review (§§ 510(k), 515), and adverse event reporting (21 C.F.R. pt. 803).

Third, the process that FDA used to ordain that LDTs are devices subject to the usual breadth and depth of FDA regulation is, in my view, inconsistent with the rulemaking provisions of the Administrative Procedure Act ("APA"), 5 U.S.C. § 553. Although the FDCA does not mention laboratory tests, FDA’s various issuances have sought to fill this gap. However, where that gap-filling binds the Agency and has significant legal, regulatory and financial implications for those outside of the Agency, it is a legislative rule. This is especially the case here, where, as recently as last year, FDA has taken quasi-enforcement action on the basis of its determination that an LDT is a device, and where the FDA determination is inconsistent with the Secretary’s existing regulations. The APA requires that legislative rules be issued through notice and comment rulemaking coupled with a Regulatory Flexibility Act ("RFA") analysis. See 5 U.S.C. § 601 et seq. Here, FDA did neither.³

All of this is not to say that during a national public health emergency, FDA would lack authority to seize or take other appropriate action against fraudulent or dangerous LDTs. Its authority, though, would not derive from the FDCA. Under the Public Health Service Act § 361(a), 42 U.S.C. § 264(a), the Public Health Service agencies, including FDA and CDC, have authority "to prevent the introduction, transmission, or spread of communicable diseases from foreign countries into the States or possessions, or from one State or possession into any other State or possession." FDA has used this authority to enjoin facilities that transplant stem cells and it could be used in those rare cases when an LDT poses undue risk. There may also be other federal agencies with authority to take action in such circumstances.

#### Background

When the MDA amended the FDCA, Congress authorized the then-Secretary of Health, Education, and Welfare to regulate medical devices through, among other things, premarket review—notification and approval—and the imposition of sanctions on those that failed to heed FDA's regulations or orders. The MDA broadly defined "device" to include, among other things, an "in vitro reagent . . . intended for use in the diagnosis of disease." MDA § 3(a)(1)(A). Laboratory developed tests—tests developed in a single clinical laboratory and used exclusively in that laboratory—were never mentioned in the MDA, in the House Report accompanying it, or during the floor debates.

In 1988, though, Congress addressed clinical laboratory testing when it enacted CLIA, codified at 42 U.S.C. § 263a, which among other things, instructed the Secretary of Health and Human Services to "issue standards to assure consistent performance by laboratories issued a certificate under this section of valid and reliable laboratory examinations and other procedures." 42 U.S.C. § 263a(f)(1). CLIA also required the Secretary to conduct inspections of laboratories to ensure compliance with established standards. See 42 U.S.C. § 263a(g). Since CLIA certification is a prerequisite to receiving Medicare payment, it was viewed primarily as Spending Clause legislation and delegated to CMS for enforcement. Thereafter, the Secretary issued comprehensive rules governing clinical laboratories. See 42 C.F.R. pt. 493. With respect to laboratory developed tests, those regulations provide as follows:

Each laboratory that modifies an FDA-cleared or approved test system, or introduces a test system not subject to FDA clearance or approval (including methods developed in-house and standardized methods such as textbook procedures), or uses a test system in which performance specifications are not provided by the manufacturer must, before reporting patient test results, establish for each test system the performance specifications for the following performance characteristics, as applicable:

(i) Accuracy.  
(ii) Precision.  
(iii) Analytical sensitivity.  

(iv) Analytical specificity to include interfering substances.  
(v) Reportable range of test results for the test system.  
(vi) Reference intervals (normal values).  
(vii) Any other performance characteristic  

42 C.F.R. § 493.1253(b)(2) (2019) (emphasis added). To perform LDTs, laboratories have to be certified under CLIA to perform highly complex tests. See id. § 493.17(c)(4). These tests can usually only be performed under the supervision of a board certified pathologist. See id. § 493.1443(b)(3) (noting that some with Ph.Ds may be grandfathered and medical doctors may satisfy the certification requirement in other ways). CLIA appeared to have occupied the field for regulating LDTs.⁴

FDA seems to have first suggested that LDTs are subject to its jurisdiction in a 1992 draft compliance policy guide aimed at regulating products sold to laboratories for research use only. The draft compliance guide stated that "laboratories have been manufacturing 'home brew' products, either from products already on the market, or from components, and utilizing these unapproved products for diagnostic purposes" and asserted that "[t]hese products are subject to the same regulatory requirements as any unapproved medical device." FDA, *Draft Compliance Policy Guide: Commercialization of Unapproved In Vitro Diagnostic Devices Labeled for Research and Investigation* at 4 (Aug. 1992).

In 1997, FDA issued a final rule regulating analyte specific reagents ("ASR"), the type frequently sold to commercial laboratories. In the preamble to the final ASR rule, FDA expressly stated that LDTs were devices subject to FDA jurisdiction: "FDA believes that clinical laboratories develop [LDTs] are acting as manufacturers of medical devices and are subject to FDA jurisdiction under the act."⁵ 62 Fed. Reg. 62,243, 62,249 (col. b) (Nov. 21, 1997).

During the intervening seventeen years, FDA did little to regulate LDTs, although it did issue a draft guidance for certain high-risk LDTs known as "in vitro diagnostic multivariate index assays" in 2007.⁶ Things changed, though, in 2014, when FDA issued two draft guidances.⁷ Not only did the Agency continue to assert its authority under the FDCA to regulate LDTs, but it noted that, as part of that authority, it was going to require registration for all LDTs, to classify LDTs under section 513, and to require premarket notification or approval under sections 510(k) or 515, respectively, for certain LDTs.

The draft guidances' legal justification for treating LDTs as devices subject to FDA jurisdiction relied on the FDCA definitions of "device" and "manufacturer." A "device" under FDCA § 201(h) is defined as:

- an instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article, including any component, part, or accessory, which is—(1) recognized in the official National Formulary, or the United States Pharmacopeia, or any supplement to them, (2) intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals, or (3) intended to affect the structure or any function of the body of man or other animals, and which does not achieve its primary intended purposes through chemical action within or on the body of man or other animals and which is not dependent upon being metabolized for the achievement of its primary intended purposes. (emphasis added)

A "manufacturer" is defined as any person who owns or operates any establishment engaged in the "manufacture, preparation, propagation, compounding, assembly, or processing of a device" is required to register that establishment with FDA. FDCA § 510(b), (c).⁸ Under FDA's logic, because an LDT is system using one or more in vitro reagents, and hence a "device," that is assembled or prepared in the clinical laboratory, and hence "manufactured," it is, in the Agency's view, subject to its regulatory jurisdiction. Most of the draft guidance is a lengthy justification as to why regulation is warranted and how that regulation would be implemented.

FDA received more than 50 comments in response to the draft guidances. Some were supportive, but many questioned the Agency's legal authority, questioned the absence of any documentation to support its claim that LDTs posed a risk, and questioned whether agency action of this magnitude could be undertaken without going through notice and comment rulemaking. On November 18, 2016, following the Presidential election, FDA announced that it would not finalize the two guidance documents. Notwithstanding this decision, FDA's position—announced initially in a compliance policy guide, later in a preamble to a regulation, and most recently in the web posting—that LDTs are medical devices within its jurisdiction remains in place. At issue is whether that decision is legally viable.

#### Analysis

##### I. Many LDTs Are Likely Medical Devices, But Even Those That Are May Fall Outside of FDA's Full Regulatory Regime

FDA maintains that LDTs are systems that include *in vitro* reagents intended to diagnose disease in humans and are therefore medical devices subject to FDA jurisdiction. Those opposing this position argue that LDTs are not physical embodiments, e.g., "contraptions," but rather are processes or services, and therefore not devices. FDA is correct that, by definition, *in vitro* reagents are devices, but that does not necessarily lead to the conclusion that LDTs fall within FDA’s jurisdiction. For purposes of this memorandum, we assume that had FDA made that determination through notice and comment rulemaking, it would be entitled to *Chevron* deference and would likely withstand scrutiny under that standard. However, even assuming that LDTs are medical devices, three additional requirements must be satisfied before FDA can implement its most significant regulatory authorities: (i) the “device” must satisfy the constitutional and statutory “interstate commerce” requirements; (ii) the device itself must be in commercial distribution or held for sale; and (iii) the laboratory must be a “person.” We believe that the first requirement may be easy to establish with respect to certain authorities; the second more difficult to establish; and the third one cannot be established, as a matter of law, in many significant instances.

###### A. Statutory Interstate Commerce Requirement

The Constitution grants Congress power “[t]o regulate commerce with foreign nations, and among the several States, and with the Indian tribes.” U.S. Const., art. I, § 8, cl. 3. Since the 1940s, the Supreme Court has construed the Commerce Clause broadly. *See, e.g., Wickard v. Filburn*, 317 U.S. 111 (1942). In addition to regulating the channels of interstate commerce, and persons and things therein, Congress has authority to regulate activities that “substantially affect” interstate commerce. *Nat’l Fed’n of Indep. Bus. v. Sebelius*, 132 S. Ct. 2566, 2578 (2012). The Court’s consistent rulings that Congress intended to regulate local, noncommercial activities that have only a nominal or indirect connection to interstate commerce. *See, e.g., Gonzales v. Raich*, 545 U.S. 1, 22 (2005); *Katzenbach v. McClung*, 379 U.S. 294, 300-01 (1964); *United States v. Wrightwood Dairy Co.*, 315 U.S. 110, 121 (1942); *Wickard*, 317 U.S. at 128-29. Given the breadth of the Court’s interpretation, the Commerce Clause poses no barrier to FDA’s theory of jurisdiction.

In the years since *Wickard*, Congress also expanded FDA’s statutory jurisdiction to cover some intrastate activities. Congress revised the FDCA in 1948 to clarify that its prohibitions against adulteration and misbranding apply to articles that are held for sale within a state after being shipped in interstate commerce. Amendments in 1976 authorized FDA to seize misbranded or adulterated medical devices without proof that they have traveled in interstate commerce. *See FDCA § 304(a)(2)*. FDA also has authority under the Public Health Service Act (“PHSA”) to prohibit false labeling of biological products whether or not they move in interstate commerce, and section 361 of the PHSA authorizes FDA regulation to prevent the spread of communicable disease without any interstate commerce limitations. However, despite these extensions of FDA jurisdiction, the Agency still lacks statutory authority to regulate a large segment of wholly intrastate conduct. Nonetheless, as those regulated by FDA have generally engaged in interstate commerce in some fashion, courts have tended to make statements regarding FDA’s statutory jurisdiction as broad as the Commerce Clause, and the Agency prevails in the overwhelming majority of cases where its jurisdiction is challenged because a component of the drug or device was transmitted in interstate commerce. We offer below, in case a litigant were to assert that their conduct is wholly intrastate, advice regarding how FDA can successfully assert that its regulation of LDTs satisfies the statutory interstate commerce requirement in section 301(k) and otherwise consider its litigation position or reviews its regulations in light thereof.

1. Section 301(k)

Section 301(k) of the FDCA (21 U.S.C. § 331(k)) prohibits:

- The alteration, mutilation, destruction, obliteration, or removal of the whole or any part of the labeling of, or the doing of any other act with respect to, a food, drug, device, tobacco product, or cosmetic, if such act is done while such article is held for sale (whether or not the first sale) *after shipment in interstate commerce* and results in such article being adulterated or misbranded. (emphasis added).

The ability of the Agency to satisfy the statutory interstate commerce requirement in section 301(k) hinges on whether a laboratory can show that everything used in its tests came from within the state. In *United States v. Regenerative Sciences, LLC*, 878 F. Supp. 2d 248 (D.D.C. 2012), *aff’d*, 741 F.3d 1314, 1326 (D.C. Cir. 2014), two Colorado physicians developed a cellular therapy for orthopedic patients that involved harvesting stem cells from a patient’s bone marrow or synovial fluid, culturing those cells for several weeks in a laboratory with growth factors from the patient’s blood, placing the cultured cells into a syringe along with the antibiotic doxycycline and other additives, and injecting the contents of the syringe into the patient’s injured area. The doctors formed Regenerative Sciences LLC (“Regenerative”) to commercialize this practice. FDA officials inspected Regenerative’s facilities in 2009 and 2010 and found that its laboratory operations did not conform to FDA manufacturing regulations. When FDA charged Regenerative with manufacturing and distributing adulterated and misbranded biological drug products in violation of section 301(k) of the FDCA and section 351(k) of the PHSA, the defendant physicians responded that they were lawfully practicing medicine within the state of Colorado and that their procedure fell outside FDA’s regulatory purview.

The district court focused on whether defendants’ actions were directly connected to interstate commerce. FDCA section 301(k) prohibits any act “with respect to, a . . . drug . . . if such act is done while such article is held for sale (whether or not the first sale) after shipment in interstate commerce and results in such article being adulterated or misbranded.” The court held that the manipulated cellular brew was “held for sale,” a fact not contested by defendants. The court further went on to hold that

[d]efendants do not dispute that the doxycycline is shipped from out of state to their facilities in Colorado. Id. Therefore, because a component of the drug in this case is shipped through interstate commerce prior to its administration to the patient, the “interstate commerce” requirement [of section 301(k)] is also met.

878 F. Supp. 2d at 259.

The defendants’ conduct satisfied the statutory interstate commerce requirement only because the doxycycline was shipped into Colorado from out of state and added to the stem cells prior to the mixture’s administration to patients. Had the doxycycline been manufactured in Colorado and shipped intrastate, FDA would have lacked a jurisdictional hook. Alternatively, suppose the defendants had administered the doxycycline separately rather than mixing it with the stem cells in a single syringe. If the procedure were modified to comprise two separate injections—a first syringe of stem cells and a second syringe of doxycycline—FDA presumably would lose its regulatory authority under the FDCA, even if the doxycycline were shipped from out of state. In this case, the defendants would be prescribing doxycycline for off-label use, an activity that FDA lacks power to regulate regardless of its connection to interstate commerce. It was the mixing of the doxycycline, an approved drug, with other material that created FDA’s jurisdictional hook.

Thus, under the terms of *U.S. v. Regenerative Sciences, LLC*, in order for FDA to satisfy the statutory interstate commerce requirement in section 301(k), at least one element of an LDT must come from outside the state. With the exception of some academic medical centers, most laboratories use LDTs that have at least one component that came from out-of-state, so we believe FDA can usually successfully defend its jurisdiction in this regard.

2. Section 510(k)

Premarket review, which is set out in section 510(k), was central to the 2014 guidances and is the primary difference between regulation under CLIA and regulation under the FDCA. FDCA section 510(k) (21 U.S.C. § 360(k)) provides:

- Each person who is required to register under this section and who proposes to begin the *introduction or delivery for introduction into interstate commerce for commercial distribution of a device intended for human use* shall, at least ninety days before making such introduction or delivery, report to the Secretary . . . action taken by such person to comply with requirements under section 514 [related to performance standards] or 515 [related to premarket approval] which are applicable to the device. (emphasis added).

Section 510(k) has a meaningful grammatical difference from section 301(k). Section 301(k) prohibits the sale in interstate commerce of a component or other aspect of a device that is then altered or manipulated. Hence, if a laboratory were to purchase in interstate commerce any of its reagents, which it then modified or used to modify other reagents, it would satisfy the jurisdictional prerequisite in section 301(k). However, under section 510(k), the premarket review requirements are only triggered when one proposes to introduce or deliver the device into interstate commerce, even if the reagents were purchased from another state. The typical LDT, though, never physically leaves the laboratory. There is no “introduction” and no “delivery.”

Thus, while the actions of the laboratory operator may have been sufficient to support regulation under the Commerce Clause, as having a substantial effect on interstate commerce, those actions may not be sufficient, if challenged by a savvy litigant, to satisfy the statutory requirements of section 510(k) or section 515 (premarket approval), which uses identical language.

###### B. “Held for Sale” or “Commercial Distribution” Requirement

1. Section 301(k)

Section 301(k) only applies if an article is “held for sale . . . after shipment.” Courts have tended to interpret section 301(k)’s “held for sale” requirement very broadly—far broader than the plain meaning of the statutory text. But it is unclear whether the current Supreme Court would ignore the plain meaning of the text and affirm these expansive readings.

Case law supports the assertion that the “held for sale” standard of section 301(k) has long been afforded a liberal reading, encompassing “[a]ll articles, compound or single, not intended for consumption by the producer.” *United States v. Cassaro, Inc.*, 443 F.2d 153, 156 (1st Cir. 1971) (citing *Hipolite Egg Company v. United States*, 220 U.S. 45, 54 (1911)). The Supreme Court has explained that that 301(k)’s “held for sale” requirement is “designed . . . to extend the [FDCA’s] coverage to every article that had gone through interstate commerce until it finally reached the ultimate consumer.” *United States v. Sullivan*, 332 U.S. 689, 697 (1948). The United States government has repeatedly advocated for this expansive reading of “held for sale,” and stated that the requirement is satisfied if the product can be shown to have been used for any purpose other than personal consumption. See, e.g., *United States v. Rhody Dairy, L.L.C.*, 812 F. Supp. 2d 1239 (E.D. Wash. 2011); *United States v. Scenic Dairy, L.L.C.*, 2011 WL 3879490 at *14 (W.D. Mich. Sep. 1, 2011); *United States v. Torigian Labs., Inc.*, 577 F. Supp. 1514, 1521 (E.D.N.Y. 1984); *Articles of Animal Drug Containing Diethylstilbestrol*, 528 F. Supp. 202, 205 (D. Neb. 1981); *United States v. Articles of Device (Acuflex; Pro-Med)*, 426 F. Supp. 366, 368 n.3 (W.D. Penn. 1977).

Several courts have held that the phrase “held for sale” extends to physicians using devices in the treatment of patients. See, e.g., *United States v. Kaplan*, 836 F.3d 1199, 1208 (9th Cir. 2016). In *Kaplan*, the Ninth Circuit held that a doctor’s use of a device (in that case, single-use plastic needle guides used during prostate biopsy exams) in the course of treating a patient is considered a “sale” within the meaning of “held for sale” in section 301(k). 836 F.3d at 1208. This makes sense—the single-use medical device is being consumed (i.e., sold) when used by the doctor, because the doctor is using the item with the patient during the course of a service. In contrast, LDTs usually involve the development of technologies and processes to conduct testing activities. For example, Medicare does not pay for the physical embodiment of any LDT or any other laboratory test. LDTs are more analogous to a doctor who creates and develops a replicable procedure, rather than a doctor who uses a medical device during an ordinary course of treatment. Just as the doctor’s development and use of a medical procedure would not be considered to be “held for sale,” the development and use of LDTs would also not be considered “held for sale” under the common meaning of that term.

In short, even though courts have given a liberal reading to the “held for sale” requirement, it is unclear whether that reading is sufficient to support liability under section 301(k) with respect to LDTs. Even in light of this uncertainty, we assume for purposes of our analysis that courts would adopt a liberal reading and apply that section to LDTs such that FDA could defend its current position.

2. Sections 510(k), 513(f), and 515

Similar to section 301(k)’s “held for sale” requirement, section 510(k) requires that persons subject to it “begin the introduction or delivery for introduction into interstate commerce for commercial distribution of a device . . . .” Similar language is used in section 513(f) and 515(b). There do not appear to be any judicial interpretations of “commercial distribution” as used in these classification and premarket review sections. But this phrase is used in the grandfathering provision of the FDCA, and with respect to that provision, FDA has interpreted “commercial distribution” to mean “on the market” or “actively promoted” for a specific purpose. See, e.g., *United States v. An Article of Device Consisting of 1,217 Cardboard Boxes*, 607 F. Supp. 990, 994 (W.D. Mich. 1985); see also *Northwest Tissue Center v. Shalala*, 1 F.3d 522, 535 (7th Cir. 1993).

The plain meaning of this phrase makes it much narrower than section 301(k)’s “held for sale” requirement. First, the term “commercial” relates to “commerce” which means the “buying or selling of commodities on a large scale involving transportation from place to place.” *Webster’s New Collegiate Dictionary* 223-24 (1980). Second, the term “distribution” means to “supply.” *Id.* at 330. This means that if LDTs are to satisfy the “commercial distribution” standard, they must be viewed as goods or commodities that are sold and dispersed beyond the laboratory. This, of course, does not occur. Each LDT remains in situ, and is not treated as merchandise by the Secretary for payment purposes, but rather as a service. For example, under Medicare part B, the Secretary only pays for clinical laboratory services. See Social Security Act § 1861(s) (defining “medical and other health services” as including diagnostic laboratory tests); § 1834A (referring to the information generated by laboratory tests). Thus, the Secretary does not purchase the physical embodiment of any LDT or any other laboratory test, for that matter.

The FDA definition of “commercial distribution” appears to be to be in keeping with the phrase’s plain meaning, namely, “any distribution of a device intended for human use which is held or offered for sale but does not include . . . ‘[i]ntracompany or interplant transfer of a device between establishments within the same parent, subsidiary, and/or affiliate company.’” 21 C.F.R. § 807.3. The development and use of LDTs involves purely internal transfers of LDTs because payors or clinicians are not paying for the LDT itself. Thus, even if LDTs could be viewed as being “held for sale,” they certainly almost always involve only internal transfers, or no transfers at all, and thus would not, if challenged by a savvy litigant, satisfy the plain meaning of FDA’s regulation.

To marginalize the importance of movement outside the walls of the laboratory is to equate “held for sale” and “commercial distribution,” which would be inconsistent with both its plain meaning and the well-established canon of statutory interpretation that the use of different words or terms within a statute demonstrates that Congress intended to convey a different meaning for those words. See *Russello v. United States*, 464 U.S. 16, 23 (1983); *Persinger v. Islamic Republic of Iran*, 729 F.2d 835, 843 (D.C. Cir. 1984) (“When Congress uses explicit language in one part of a statute to cover a particular situation and then uses different language in another part of the same statute, a strong inference arises that the two provisions do not mean the same thing.”); *Nat’l Insulation Transp. Comm. v. ICC*, 683 F.2d 533, 537 (D.C. Cir. 1982); *Russell v. Law Enforcement Assistance Admin.*, 637 F.2d 354, 356 (5th Cir. 1981) (stating the “well settled rule of statutory construction that where different language is used in the same connection in different parts of a statute it is presumed that the Legislature intended a different meaning and effect”) (internal quotation marks omitted); *NLRB v. Food Fair Stores, Inc.*, 307 F.2d 3, 10 (3rd Cir. 1962) (stating the rule of statutory construction which holds that different words appearing in the same statute are presumed to have different meanings). Even words with remarkably similar definitions can still convey a unique or distinct meaning or flavor from words that are similar or even synonymous in nature because of their differing tone or usage within a sentence.

###### C. The “Person” Requirement—Sections 510(e), 510(k), 515(e), and 21 C.F.R. pt. 803

In addition to the statutory coverage issue, the “held for sale,” and the “commercial distribution” requirements of the FDCA, portions of the Act only apply to a “person.” Thus, the registration and premarket review requirements of section 510(c) and 510(k), premarket application requirement of section 515(c), and the adverse event reporting requirements of 21 C.F.R. pt. 803 apply only to “persons.” See 21 C.F.R. § 803.3(1) (stating that a “Manufacturer means any person”). Under the Act, though, a “state” is not “person.”

1. State is Not a “Person”

Many of the more sophisticated laboratories that are the ones most likely to first develop LDTs in response to an infectious disease are located in state public health departments and academic medical centers at state universities. A state, including its departments and state-owned universities, is presumed by definition not to be a “person.” See *Vt. Agency for Nat. Res. v. U.S. ex rel. Stevens*, 529 U.S. 765 (2000) (applying its “longstanding interpretive presumption that ‘person’ does not include the sovereign” to find that a state is not a “person” within the meaning of the False Claims Act). That presumption is not necessary here because the Act defines “person” to “include[] [an] individual, partnership, corporation, and association.”

FDCA § 201(e). The Act separately defines “State,” “except as used in the last sentence of section 372(a) of this title, [to mean] any State or Territory of the United States, the District of Columbia, and the Commonwealth of Puerto Rico.” Id. § 201(a). The two definitions are not linked or cross-referenced. Thus, a “person” is not a “State.”

2. Premarket Review, Registration, and Related Provisions Only Apply to “Persons” and Not States

Section 510(c) requires “[e]very person upon first engaging in manufacture ... of ... a device ... shall register with the Secretary. . . .” The premarket notification provision of section 510(k) applies to “[e]ach person who is required to register under this section . . . .” Inasmuch as a state would not be required to register, it is also not required to file a premarket notification under section 510(k). If a device manufacturer does not take advantage of this pathway, it is generally subject to review under other more rigorous premarket review pathways in the Act, such as the premarket approval in section 515 or the de novo review in section 513(f)(2). For most manufacturers, this is true by operation of two provisions in the statute. First, under section 513(f)(1) of the FDCA, a non-grandfathered device is automatically classified in class III unless the device “is substantially equivalent to another device” or has been classified pursuant to a petition or request, such as a de novo request.

If such a device cannot be found to be substantially equivalent under the 510(k) pathway and if it has not been classified through another process, such as the de novo process, it is automatically classified in class III. Second, under section 515(a)(2), a device that is in class III by virtue of section 513(f) is “required to have ... an approval under this section of an application for premarket approval.” Devices that are subject to these provisions and that lack premarket approval are adulterated. See FDCA § 501(f)(1)(B).

In the case of a state actor, though, the normal interplay between sections 501, 510, 513, and 515 breaks down. First, a state is not required to file a premarket notification under section 510(k) or to register under section 510(c). Second, to avoid being automatically treated as a class III device, section 513(f) merely requires that the device is “substantially[ly] equivalent[]” to another lawfully marketed device. The section does not require that the Secretary make this finding or receive a report under section 510(k). Even assuming that section 513(f) applied and the device were by default classified into class III, that would still not impose any burdens on a state because it is likely that the premarket approval provision of section 515 does not apply to states and may not apply to LDTs, at all, and the adulteration provision also does not apply to states.

A device is adulterated if it were classified under section 513(f) into class III; a class III device, under section 515(a) “is required to have an approval under this section of an application for premarket approval . . . .” This presupposes that the state actor is required to file such an application. Section 515(c), though, limits those who may file applications to “persons:” “[a]ny person may file with the Secretary an application for premarket approval for a class III device.” FDCA § 515(c)(1) (emphasis added). It would be anomalous to require an approved application from an entity not required to file an application. One could argue that the phrase “any person may file” does not foreclose a state from filing a PMA. Courts, though, have viewed similar or identical phrases in other statutes as restricting the class that can file. Thus, under the False Claims Act, “[a] person may bring a civil action for a violation of section 3729 for the person and for the United States Government.” 31 U.S.C. § 3730(b). No one has suggested, following the Court’s decision in ex rel. Stevens, that a state could act as a relator and file a qui tam suit under section 3730(b). We believe the most natural way to read these provisions is recognize that when the FDCA and MDA were enacted and amended no one contemplated that they would apply to states. This is especially so given that the penalty provisions of the FDCA only apply to “persons.” See FDCA § 303. The fact that state public health and academic medical center laboratories appear to fall between the regulatory cracks in the case of LDTs, strongly suggest that the Act was never intended to reach these services.

As a result, FDA’s registration, premarket review, and adverse event reporting requirements would not, if challenged by a sophisticated litigant, likely apply, as a matter of law, to any state-owned laboratory, whether in a state department of public health or university.

In sum, although it appears that FDA was acting within its discretion by treating LDTs as medical devices and that section 301(k) could be applied, its premarket review authority would not apply to LDTs, because the provisions require the device itself (i.e., the reagent) to be placed into commercial distribution. Further, the registration, premarket review and adverse event reporting requirements only apply to “persons.” State laboratories are unlikely to be considered persons within the meaning of the FDCA.

##### II. FDA’s Policy that LDTs Are Devices Was Adopted Without Notice and Comment Rulemaking or a Regulatory Flexibility Analysis and Is Therefore Void

The APA establishes the procedures federal administrative agencies must use for “rule making,” defined as the process of “formulating, amending, or repealing a rule.” 5 U.S.C. § 551(5). “Rule,” in turn, is defined broadly to include “statement[s] of general or particular applicability and future effect” that are designed to “implement, interpret, or prescribe law or policy.” Id. § 551(4); see Perez v. Mortgage Bankers Ass’n, 575 U.S. 92, 95–96 (2015). Rules fall into two broad categories—legislative rules and interpretive rules. Legislative rules can only be issued through notice and comment rulemaking, formal rulemaking or negotiated rulemaking; interpretive rules can be issued unilaterally. See 5 U.S.C. § 553(b). Therefore, whether an agency’s issuance is a legislative rule or interpretive rule can have major consequences. Differentiating between the two, though, is complicated.

We believe that FDA’s determination that an LDT is a device is a legislative rule for at least three independent reasons: (1) it fills a gap in the Act; (2) the Agency has treated its determination as legally binding forming the basis of enforcement and the exercise of enforcement discretion; and (3) the Agency’s determination is inconsistent with an extant legislative rule of the Department.

###### A. FDA’s Guidances and Determinations Are “Gap Filling” and Therefore Legislative Rules

First, a legislative rule “performs a legislative function when it makes ‘reasonable but arbitrary (not in the “arbitrary and capricious” sense) rules that are consistent with the statute or regulation under which the rules are promulgated but not derived from it, because they represent an arbitrary choice among methods of implementation.’” Catholic Health Initiatives v. Sebelius, 617 F.3d 490, 495 (D.C. Cir. 2010) (citing Hoctor v. USDA, 82 F.3d 165, 170 (7th Cir. 1996)). Gap-filling is a quintessential characteristic of a legislative rule.

FDA’s determination starting in 1992, confirmed in 1997, reconfirmed in the 2014 draft guidances, and reconfirmed in its January 13, 2017 White Paper that LDTs are devices were all gap-filling policy determinations with significant economic and regulatory implications. The proposed LDT framework in the 2014 draft guidances would have required FDA to review virtually all LDTs. Under the White Paper, FDA review would be limited to “new and significantly modified high and moderate risk LDTs.” White Paper at 4–5. As a result, FDA envisions that the process could be completed in four years, rather than the rather prolonged nine-year period originally envisioned in the draft guidances. In the White Paper, the Agency also stated that “[t]o protect patients from tests that could lead to harm, the Agency would retain its ability to enforce premarket review, quality systems, and other applicable requirements for any LDT, including those listed above, if the agency identified one or more of the following, taking into account all available evidence.” Id. at 4. It also represents a clear choice, opting to define an LDT as a device; it thereby filled a gap in the statute, a quintessential characteristic of a legislative rule.

Nor is this a case where one can argue that the organic legislation unmistakably leads to the conclusion that LDTs are devices and an enforcement action can be based solely on the statute. Laboratory developed tests are not mentioned in the FDCA nor are they defined in it. Rather, they are only defined in FDA guidances and similar issuances. See, e.g., Framework for Regulatory Oversight of Laboratory Developed Tests (LDTs) at 5 (Oct. 3, 2014) (defining LDT in guidance Document”). It would be difficult to argue that FDA regulation of LDTs is so inherent in the FDCA that no regulation is necessary. This is especially so where the Secretary has issued rules implementing Medicare and CLIA that strongly suggest that LDTs are not devices and not within FDA’s jurisdiction.

The argument that LDTs could be regulated without a regulation is also belied by the public comments submitted in response to the 2014 draft guidances which challenged FDA’s contention that an LDT is a device. One commenter argued that

- [i]t is far-fetched to suppose that laboratory-developed testing services become medical devices in their own right merely because they sometimes utilize other medical devices. FDA’s own regulations recognize the distinction between a service that uses devices and a device itself. For example, the FDA regulation excluding laboratories from device registration requirements specifically recognizes that laboratories’ “primary responsibility to the ultimate consumer is to ... provide a service through the use of a previously manufactured device.” 21 C.F.R. §807.65(i) (emphasis added). Laboratories may well draw on both reagents and laboratory equipment of many kinds in executing their clinical testing services, but that plainly does not render the services these laboratories perform themselves “medical devices.”

Comment submitted by Paul D. Clement & Laurence H. Tribe on Behalf of the American Clinical Laboratory Association at 9.

FDA’s ability to regulate LDTs is not inherent in the language of the Act, which is silent

on the point. When, as here, the statute is silent, an agency’s attempt to describe the contours of its authority is “gap-filling.” Gap-filling, though, can only occur through notice and comment rulemaking. *See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc.*, 467 U.S. 837, 843–44 (1984).

###### B. FDA’s Determinations Were Intended to Be Binding and Have the Force and Effect of Law

Second, an agency action that purports to impose legally binding obligations or prohibitions on regulated parties—and that would be the basis for an enforcement action for violations of those obligations or requirements—is a legislative rule. An agency action that sets forth legally binding requirements for a private party to obtain a permit or license is a legislative rule. *National Min. Ass’n v. McCarthy*, 758 F.3d 243, 251–52 (D.C. Cir. 2014); *see also Appalachian Power Co. v. E.P.A.*, 208 F.3d 1015, 1021 (D.C. Cir. 2000). Correspondingly, legislative rules, as opposed to interpretive rules, “‘grant rights, impose obligations, or produce other significant effects on private interests;’ ‘narrowly constrict the discretion of agency officials by largely determining the issue addressed;’ and ‘[has] substantive legal effect.’” *U.S. Telecom Ass’n v. FCC*, 400 F.3d 29, 35 (D.C. Cir. 2005) (quoting *Batterton v. Marshall*, 648 F.2d 694, 701–02 (D.C. Cir. 1980)).

The finding that an LDT is a device is a sentinel determination enabling the Agency, at any time, to take enforcement action, to require registration, listing, compliance with quality systems, and premarket review and clearance or approval, at the Agency’s discretion. Because this finding has significant economic effects on private interests, it raises the specter that the Agency, exercising discretion that is arguably not reviewable, could require laboratories to comply with any or all of these requisites that govern ordinary devices, the determination fits the profile of a legislative rule. The Agency’s determination that an LDT is a device “would be the basis for an enforcement action for violations of those obligations.” *National Min. Ass’n, supra.* Consistent with that description, the Agency recently took action in the form of a Warning Letter against a laboratory based on its determination that an LDT is a device. *See* Warning Letter to Inova Genomics Laboratory (April 4, 2019). That Warning Letter is, to a reviewing court, convincing evidence that the decision to treat an LDT as a device was intended to have the force and effect of law and has been treated as such by the Agency. Since the determination was issued without the benefit of notice and comment rulemaking, it is likely to be considered void if challenged.

FDA has acknowledged that its issuances have the force and effect of law, even though its guidance documents trumpet the opposite in boilerplate. The Agency noted that it had exercised enforcement discretion with respect to LDTs, but that the 2014 draft guidances, when finalized, would have ended that. *See* Guidance Document at 6. Two years later, when abandoning the 2014 draft guidances, the Agency indicated its intent to continue extending enforcement discretion to LDTs, except in the event of a pandemic. *See* https://www.fda.gov/emergency-preparedness-and-response/mcm-legal-regulatory-and-policy-framework/information-laboratories-implementing-ivd-tests-under-eua (March 1, 2020) (last reviewed June 15, 2020) (“FDA generally has not enforced premarket review and other legal requirements [with respect to LDTs].”). Obviously, an agency can only exercise enforcement discretion if it believes that it has the authority to enforce its determination, in this case that an LDT is a device.

###### C. FDA’s Determination Is Inconsistent with Extant Departmental Regulations and Is Therefore Void Absent Notice and Comment Rulemaking

Third, if an agency adopts a position that is “inconsistent with an existing regulation, or effects ‘a substantive change in the regulation,’ notice and comment are required.” *U.S. Telecom Ass’n v. F.C.C.*, 400 F.3d 29, 35 (D.C. Cir. 2005) (quoting *Shalala v. Guernsey Mem’l Hosp.*, 514 U.S. 87, 100 (1995)). Here, before FDA’s initial determination in 1992 that LDTs are devices, the Secretary issued a comprehensive regulation implementing CLIA that expressly recognized three classes of laboratory tests—(i) those purchased by a laboratory that were FDA cleared or approved, (ii) those that are modifications of FDA cleared or approved tests, or (iii) “test system[s] not subject to FDA clearance or approval (including methods developed in-house [i.e., LDTs] and standardized methods such as text book procedures . . .).” 42 C.F.R. § 493.1253(b) (2019) (adopted 68 Fed. Reg. 3640, 3707 (Jan. 24, 2003), previously codified at 42 C.F.R. 493.1213(b), 57 Fed. Reg. 7163 (Feb. 28, 1992)). Those laboratories that use FDA cleared or approved tests are required to employ fewer quality controls than those laboratories that use modified tests or LDTs.

At bottom, the 1992 CLIA regulation recognizes LDTs as a separate class of tests not subject to FDA clearance or review, and by implication, FDA jurisdiction. That regulation is consistent with the legislative history underlying the 1988 amendments to the original Clinical Laboratory Improvement Act of 1967, which was limited to specimens traveling in interstate commerce. In supporting the 1988 amendments, Chairman Waxman noted the complete absence of federal regulation in certain instances: “many laboratories, particularly physicians' offices and smaller laboratories not accepting specimens in interstate commerce, are not subject to such Federal regulations.” FDA’s determinations that LDTs are devices, none of which was published for notice and comment rulemaking, are inconsistent with this CLIA rule and the legislative history surrounding the 1988 amendments.

This inconsistency leads to one outcome for two alternative reasons. An interpretive rule or other policy issued without the benefit of notice and comment is void *ab initio* if it is inconsistent with an existing legislative rule. The legislative rule takes precedence over the interpretive one. *See F.C.C. v. Fox Television Stations, Inc.*, 556 U.S. 502, 515 (2009) (“An agency may not, for example, depart from a prior policy *sub silentio* or simply disregard rules that are still on the books.”); *Berkovitz v. United States*, 486 U.S. 531 (1988) (HHS is not free to ignore its own legislative rules); *Gunderson v. Hood*, 268 F.3d 1149, 1154 (9th Cir. 2001) (If a rule is inconsistent with or amends an existing legislative rule, then it cannot be interpretive.”). Alternatively, the Agency determinations are void because they seek to modify or repeal an existing legislative rule which can only be accomplished through notice and comment rulemaking. *See Perez*, 575 U.S. at 105 (“APA rulemaking would still be required if [an agency] adopted a new position inconsistent with . . . existing regulations.”); *Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Ins. Co.*, 463 U.S. 29 (1983).

Thus, even if none of the statutory impediments were ever existed, the Agency’s determination that an LDT is a “device” would still falter for want of notice and comment rulemaking.

##### III. CLIA and Other Provisions in the PHS Act Can Provide Appropriate Safeguards

Regardless of the advice outlined above, we note, in light of concerns raised by some regarding dual regulation by FDA and CMS in this space, that the Secretary retains authority to ensure administrative efficiency by channeling regulation of LDTs through one agency and to determine which agency should exercise that authority. Policymakers may wish to consider whether CMS, which regulates through the Spending Clause and already regulates the actual use of tests in the laboratory, is better suited legally and logistically to regulate LDTs than is FDA, which is tethered by the Commerce Clause and by statutory commerce clause requirements.

One might take the position that Congress has addressed the federal regulation of laboratory testing in CLIA, and the Secretary has determined or can determine that CMS is the agency within HHS to regulate clinical laboratories. This is not to say that by enacting CLIA Congress expressed an intent that no further regulation was necessary. Rather, CLIA’s comprehensive scheme arguably makes regulation by another HHS agency less essential. That is especially the case where the second agency’s authority is relatively limited or vulnerable to legal challenge. In such situations, the other agency often provides technical assistance and that could be a model in this instance as well.

## EXHIBIT G
**same as contents of 2023-12-04_Letter to FDA - ACLAs Comments on FDAs Proposed Regulation of LDTs as Medical Devices**

## EXHIBIT H

### Laboratory Developed Tests Final Rule

Docket No. FDA-2023-N-2177

Final Regulatory Impact Analysis  
Final Regulatory Flexibility Analysis  
Unfunded Mandates Reform Act Analysis  

Economics Staff  
Office of Economics and Analysis  
Office of Policy, Legislation, and International Affairs  
Office of the Commissioner  

#### Executive Summary

This final rule amends FDA’s regulations in part 809 (21 CFR part 809) to make explicit that “in vitro diagnostic products” (IVDs) are devices as defined in section 201(h)(1) of the Federal Food, Drug, and Cosmetic Act (FD&C Act) (21 U.S.C. 321(h)(1)) including when the manufacturer of the IVD is a laboratory. In conjunction with this amendment, FDA is phasing out its general enforcement discretion approach for laboratory developed tests (LDTs) so that IVDs manufactured by a laboratory will generally fall under the same enforcement approach as other IVDs, as discussed further in section V of the preamble to the rule.

We quantify benefits to patients from averted health losses due to problematic IVDs offered as LDTs.¹,² We focus mainly on certain broad disease categories associated with the majority of misdiagnosis-related harms in the U.S. Additional benefits include averted non-health losses from reduced spending on problematic IVDs offered as LDTs and unquantified reduction in costs from lawsuits. We quantify costs to affected laboratories for complying with statutory and regulatory requirements, as described in the phaseout policy. Additional costs include costs to FDA, which we include in our estimates. We estimate that the annualized benefits over 20 years range from $0.99 billion to $11.1 billion at a seven percent discount rate, with a primary estimate of $3.51 billion, and from $1.24 billion to $13.62 billion at a three percent discount rate, with a primary estimate of $4.34 billion. The annualized costs range from $566 million to $3.56 billion at a seven percent discount rate, with a primary estimate of $1.29 billion, and from $603 million to $3.79 billion at a three percent discount rate, with a primary estimate of $1.37 billion.

¹ See discussion of “problematic IVDs” in section I.B below.  
² See discussion of “IVDs offered as LDTs” in section V.A.1 of the preamble to the final rule and section II.D below.

#### Table of Contents

I. Introduction and Summary ................................................................................................. 6  
   A. Introduction ..................................................................................................................... 6  
   B. Overview of Benefits, Costs, and Transfers .................................................................... 7  
   C. Comments on the Preliminary Economic Analysis of Impacts and Our Responses ...... 10  
      1. Comments (Number of Laboratories) ......................................................................... 10  
      2. Comments (Number of IVDs Offered as LDTs) ....................................................... 11  
      3. Comments (Percent of IVDs That Are Offered as LDTs) ......................................... 13  
      4. Comments (Attribution of Diagnostic Error to Analytic Phase of Laboratory Tests) 14  
      5. Comments (Percent of IVDs Offered as LDTs that are Problematic) ...................... 15  
      6. Comment (IVDs Offered as LDTs Perform Better Than FDA-Authorized Tests) .... 15  
      7. Comments (Effectiveness of FDA Review in Assuring Reliability of Diagnostic Tests) 16  
      8. Comment (Diagnostic Tests Only a Part of Diagnosis) ............................................ 17  
      9. Comments (StatinCheck Problem) ............................................................................ 18  
      10. Comments (Non-Invasive Prenatal Screening Tests Require Follow-Up) .............. 18  
      11. Comments (Innovation) ............................................................................................ 19  
      12. Comments (Impact on Prices) .................................................................................. 21  
      13. Comments (Increased Labor Cost/Strain) ................................................................. 24  
      14. Comments (Underestimation of Costs) ..................................................................... 25  
      15. Comments (Outsourcing Costs and Costs of Switching to FDA-Authorized Tests) . 26  
      16. Comment (FDA Would Not Have Sufficient Resources To Review IVDs Offered as LDTs) .............................................................................................................. 27  
      17. Comment (Small Entities) ......................................................................................... 30  
      18. Comment (Firm Exit and Market Concentration) ...................................................... 33  
   D. Summary of Changes from the Proposed Rule .............................................................. 34  
      1. Changes to the Phaseout Policy ................................................................................... 34  
      2. Baseline Conditions ..................................................................................................... 35  
      3. Benefits ........................................................................................................................ 36  
      4. Costs ............................................................................................................................ 37  
      5. Regulatory Alternatives ............................................................................................... 38  
II. Final Economic Analysis of Impacts ................................................................................ 38  
    A. Background .................................................................................................................. 38  
    B. Need for the Rule ......................................................................................................... 40  
    C. Purpose of the Rule ...................................................................................................... 46  
    D. Baseline Conditions ..................................................................................................... 46  
       1. Number of Affected Entities ..................................................................................... 50  
       2. Baseline Market Revenue ......................................................................................... 55  
       3. Baseline FDA Premarket Reviews of Submissions/Applications .............................. 56  
       4. Baseline Risk of Problematic IVDs .......................................................................... 57  
    E. Benefits ........................................................................................................................ 59  
       1. Reduction in Harms from Diagnostic Errors ............................................................. 59  
       2. Non-Health Benefits .................................................................................................. 85  
       3. Summary of Benefits ................................................................................................. 88  
    F. Costs ............................................................................................................................ 91  
       1. Costs Under Stage 1 .................................................................................................. 92  
       2. Costs Under Stage 2 ................................................................................................. 100  
       3. Costs Under Stage 3 ................................................................................................. 106  
       4. Costs Under Stages 4 and 5 ...................................................................................... 112  
       5. Summary of Costs ..................................................................................................... 123  
       6. Other Unquantified Costs .......................................................................................... 125  
    G. Budgetary Impacts ....................................................................................................... 128  
    H. Transfers ...................................................................................................................... 132  
    I. Stream of Benefits, Costs, and Transfers ..................................................................... 135  
    J. Analysis of Regulatory Alternatives to the Final Phaseout Policy ............................... 137  
    K. Distributional Effects ................................................................................................... 142  
    L. International Effects .................................................................................................... 145  
III. Final Small Entity Analysis ............................................................................................ 148  
     A. Description and Number of Affected Small Entities ................................................... 148  
     B. Description of the Potential Impacts of the Phaseout Policy on Small Entities ........... 149  
     C. Alternatives to Minimize the Burden on Small Entities .............................................. 153  
IV. References ....................................................................................................................... 155  
Appendix A. Estimation of the Number of Affected Labs and Tests ..................................... 163  
Appendix B. Final Small Entity Analysis Estimates ............................................................... 173  

#### I. Introduction and Summary

##### A. Introduction

We have examined the impacts of the final rule under Executive Order 12866, Executive Order 13563, Executive Order 14094, the Regulatory Flexibility Act (5 U.S.C. 601-612), and the Unfunded Mandates Reform Act of 1995 (Pub. L. 104-4).

Executive Orders 12866, 13563, and 14094 direct us to assess all benefits, costs, and transfers of available regulatory alternatives and to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety, and other advantages; distributive impacts; and equity). Rules are “significant” under Executive Order 12866 Section 3(f)(1) (as amended by Executive Order 14094) if they “have an annual effect on the economy of $200 million or more (adjusted every 3 years by the Administrator of [the Office of Information and Regulatory Affairs (OIRA)] for changes in gross domestic product); or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, territorial, or tribal governments or communities.” OIRA has determined that this final rule is a significant regulatory action under Executive Order 12866 Section 3(f)(1).

Because this rule is likely to result in an annual effect on the economy of $100 million or more or meets other criteria specified in the Congressional Review Act/Small Business Regulatory Enforcement Fairness Act, OIRA has determined that this rule falls within the scope of 5 U.S.C. 804(2).

The Regulatory Flexibility Act requires agencies to analyze regulatory options that would minimize any significant impact of a rule on small entities. Because most facilities that will be affected by this rule are defined as small businesses and the final rule is likely to impose a substantial burden on the affected small entities, we find that the rule will have a significant economic impact on a substantial number of small entities.

We prepared an analysis consistent with the Unfunded Mandates Reform Act of 1995 (section 202(a)), which requires the preparation of a written statement that includes estimates of anticipated impacts before issuing “any rule that includes any Federal mandate that may result in the expenditure by State, local, and tribal governments, in the aggregate, or by the private sector, of $100,000,000 or more (adjusted annually for inflation) in any one year.” The current threshold after adjustment for inflation is $177 million, using the most current (2022) Implicit Price Deflator for the Gross Domestic Product. This final rule will result in an expenditure in at least one year that meets or exceeds this amount.

##### B. Overview of Benefits, Costs, and Transfers

This final rule amends FDA’s regulations to make explicit that in vitro diagnostic products (IVDs) are devices under the Federal Food, Drug, and Cosmetic Act (the FD&C Act) including when the manufacturer of the IVD is a laboratory. As discussed in section V of the preamble to the final rule, FDA is phasing out its general enforcement discretion approach for LDTs so that IVDs manufactured by a laboratory will generally fall under the same enforcement approach as other IVDs.

We anticipate that the benefits of phasing out FDA’s general enforcement discretion approach for LDTs includes a reduction in healthcare costs associated with unsafe or ineffective IVDs offered as LDTs (generally referred to in this document as “problematic IVDs”), including IVDs offered as LDTs that are promoted with false or misleading claims, and from therapeutic decisions based on unreliable results of those tests. Quantified benefits are the annualized sum of both health and non-health benefits. Unquantified benefits include, among others, possible reduction in costs from lawsuits. We discuss the benefits of phasing out of FDA’s general enforcement discretion approach for IVDs offered as LDTs in section II.E.

This phaseout policy will result in compliance costs for laboratories that are ensuring their IVDs offered as LDTs are compliant with statutory and regulatory requirements, as described in section V of the preamble. We discuss the costs of the phaseout policy in section II.F. These costs overlap somewhat with effects associated with this phaseout policy in the form of user fees, including annual registration fees, fees for premarket applications/submissions, and annual fees for periodic reporting concerning PMA-approved devices, which are paid from laboratories to FDA. These fees are paid by laboratories but are revenue for FDA; the approach to estimating fee effects is distinct from the approaches for either benefits or costs, so they will be presented as transfers. We discuss transfers in section II.H.

Table 1 summarizes the annualized benefits, costs, and transfers of the phaseout policy. At a seven percent discount rate, 20-year annualized benefits range from about $0.99 billion to $11.1 billion, with a primary estimate of $3.51 billion per year. At a three percent discount rate, 20-year annualized benefits range from $1.24 billion to $13.62 billion, with a primary estimate of $4.34 billion per year. At a seven percent discount rate, 20-year annualized costs range from about $566 million to $3.56 billion, with a primary estimate of $1.29 billion per year. At a three percent discount rate, annualized costs range from about $603 million to $3.79 billion, with a primary estimate of $1.37 billion per year. At a seven percent discount rate, 20-year annualized transfers range from $20 million to $81 million, with a primary estimate of $41 million per year. At a three percent discount rate, 20-year annualized transfers range from $29 million to $115 million, with a primary estimate of $58 million per year. These estimates do not include anticipated offsets from user fees. At a seven percent discount rate, 20-year annualized costs to FDA range from $61 million to $243 million, with a primary estimate of $121 million per year. At a three percent discount rate, 20-year annualized costs to FDA range from $65 million to $259 million, with a primary estimate of $129 million per year. Factoring in offsets from user fees at current levels, estimated costs to FDA are reduced to $40 million to $162 million at a seven percent discount rate, with a primary estimate of $81 million, and to $36 million to $144 million at a three percent discount rate, with a primary estimate of $72 million, covering approximately 30 to 40 percent of the estimated costs to FDA.

Table 1. Summary of Benefits, Costs and Transfers (millions of 2022 U.S. dollars)

| Category      | Primary Estimate | Low Estimate | High Estimate | Year Dollars | Discount Rate | Period Covered | Notes |
|---------------|------------------|--------------|---------------|--------------|---------------|----------------|-------|
| **Benefits**  |                  |              |               |              |               |                |       |
| Annualized Monetized ($m/year) | $3,509         | $988          | $11,096      | 2022          | 7%            | 20 years       | Major sources of benefits will be the avoidance of harms to patients from use of problematic IVDs offered as LDTs and the avoidance of spending on such IVDs. |
|               | $4,341          | $1,244        | $13,619       | 2022          | 3%            | 20 years       |       |
| Annualized Quantified |                  |              |               |              | 7%            |                |       |
|               |                  |              |               |              | 3%            |                |       |
| Qualitative   |                  |              |               |              |               |                |       |
| **Costs**     |                  |              |               |              |               |                |       |
| Annualized Monetized ($m/year) | $1,287         | $566          | $3,559       | 2022          | 7%            | 20 years       | A portion of foreign costs will be passed on to domestic consumers. We estimate that up to $147 million in annualized costs (7%, 20 years) to foreign facilities could be passed on to domestic consumers. |
|               | $1,372          | $603          | $3,789       | 2022          | 3%            | 20 years       |       |
| Annualized Quantified |                  |              |               |              | 7%            |                |       |
|               |                  |              |               |              | 3%            |                |       |
| Qualitative   |                  |              |               |              |               |                |       |
| **Transfers** |                  |              |               |              |               |                |       |
| Federal Annualized Monetized ($m/year) | $41            | $20           | $81          | 2022          | 7%            | 20 years       | The main portion of transfers will be user fees for premarket submissions. |
|               | $58            | $29           | $115         | 2022          | 3%            | 20 years       |       |
| Other Annualized Monetized ($m/year) |                  |              |               |              | 7%            |                |       |
|               |                  |              |               |              | 3%            |                |       |
| **Effects**   |                  |              |               |              |               |                |       |
| State, Local, or Tribal Government: No significant effects | | | | | | | |
| Small Business: The phaseout policy will have a significant economic impact on a substantial number of small laboratories that manufacture IVDs offered as LDTs. | | | | | | | |
| Wages: N/A    |                  |              |               |              |               |                |       |
| Growth: N/A   |                  |              |               |              |               |                |       |

##### C. Comments on the Preliminary Economic Analysis of Impacts and Our Responses

On October 3, 2023, FDA published the proposed rule Medical Devices: Laboratory Developed Tests (88 FR 68006). Accompanying the proposed rule was a comprehensive preliminary regulatory impact analysis (hereinafter referred to as the preliminary analysis or PRIA) on which we requested public comments (Ref. [1]). We received many comments and have organized these comments and our responses by topic in the paragraphs below. The number assigned to each comment is purely for organizational purposes and does not signify the comment’s value, importance, or the order in which it was received.

###### 1. Comments (Number of Laboratories)

Comments suggested using data from the CMS Laboratory Registry which provides information on the number of laboratories in the United States and their accreditation status to estimate the number of laboratories affected by the rule.

**Response:** As mentioned in the PRIA, we acknowledge that we do not know the exact number of laboratories that will be affected by this rule. After reviewing comments, FDA revised the number of affected laboratories from 12,000 to 11,808 using data from the CMS Laboratory Registry. We still use information about laboratories in New York State (NYS) to estimate the percent of CLIA-certified laboratories that both comply with high complexity requirements and make IVDs offered as LDTs, assuming that NYS is representative of the U.S. laboratory community. We explain our revised estimate in greater detail in section II.D.1 and appendix A of this analysis.

###### 2. Comments (Number of IVDs Offered as LDTs)

Some comments claimed FDA overestimated the number of IVDs offered as LDTs on the market while others claimed FDA underestimated this number. One comment stated that there are 160,000 IVDs offered as LDTs in the United States from 12,000 laboratories (or 13 IVDs offered as LDTs per laboratory). Other comments provided estimates of the number of IVDs offered as LDTs ranging from 92 to 310 IVDs offered as LDTs per laboratory.

**Response:** As mentioned in the preliminary analysis (section II.D.1), we acknowledge that some large reference laboratories may make a large number of new IVDs offered as LDTs per year, whereas smaller laboratories may focus on fewer IVDs overall and may not introduce many or any new IVDs each year. Using the additional estimates on the number of IVDs offered as LDTs received from comments, the weighted average estimate of the affected tests is calculated to be approximately 69 IVDs offered as LDTs per affected entity, which is close to
However, we adjusted our estimate to reflect the enforcement discretion policies in the final phaseout policy as well as the Agency’s intention to initiate the reclassification process for most IVDs that are currently class III (high risk) into class II (moderate risk).[^4] FDA will also continue taking a risk-based approach in the initial classification of individual IVDs to determine the appropriate level of regulatory controls and whether a new test may be classified into class II through De Novo classification (and special controls established), rather than being class III and subject to the PMA pathway. Based on its experience, the Agency believes that special controls could be developed, along with general controls, that could provide a reasonable assurance of safety and effectiveness for most future companion diagnostic and infectious disease IVDs. As such they would be regulated as class II devices. As a result of this adjustment, and prior to additional adjustments to address enforcement policies, the estimated numbers of PMAs and PMA supplements submissions are lower while the estimated numbers of 510(k)s and De Novo submissions are higher after potential reclassification (see Table A.5). As the final phaseout policy includes an enforcement discretion policy with regards to QS requirements (except for requirements under 21 CFR part 820, subpart M (Records)) and premarket review requirements for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified in certain limited ways as described in the preamble, we expect that fewer IVDs offered as LDTs will be affected by stages 3 through 5 of the phaseout policy than we estimated in our preliminary analysis. In addition, FDA has revised the phaseout policy to include several other enforcement discretion policies for certain other types of IVDs (see section V.B of the preamble) and we have revised our estimates accordingly. Our updated estimates are addressed in sections II.F.3 and II.F.4 of this document.

###### 3. Comments (Percent of IVDs That Are Offered as LDTs)

As discussed in section VI.C of the preamble, some comments claimed FDA overestimated the number of IVDs offered as LDTs on the market. Relatedly, comments suggested that the percent of test order volume for IVDs offered as LDTs is lower than 50%. One comment claimed that FDA’s estimate of the number of IVDs offered as LDTs was more than “10 times what researchers found in a peer-reviewed study published in the American Journal of Clinical Pathology of actual clinical test orders at University of Utah Health: 3.9%” (see Ref. [3]). Another comment stated that only 6% of tests performed in their laboratory are offered as LDTs.

Other comments suggested that FDA underestimated the number of IVDs offered as LDTs on the market. Among these, a comment noted that their laboratory, itself part of an academic medical center, offers 123 LDTs out of 124 tests, a percentage higher than 99%. Another comment stated that more than 99% of their tests are LDTs, and further clarified that these tests comprised an even higher percent of their test volume.

**Response:** Based on public comments, the percent of IVDs offered as LDTs and the percent of test order volume comprised by such appear to vary widely across settings. Using test orders from a U.S. academic hospital system, Rychert et al. (2023) estimate that IVDs offered as LDTs are 3.9% of test order volume and 45% of distinct tests (Ref. [2]). Specifically for estimating the percent of patients who are tested with IVDs offered as LDTs, we thus consider a range from 3.9% to 45%, with a primary estimate of 10%. This reflects the assumption that, while we take 45% as a reasonable estimate of the percent of IVDs that are offered as LDTs, we consider the percent of patients tested with those IVDs offered as LDTs to be closer to the referenced 3.9%, and, using professional judgement, selected 10% as our primary estimate, rather than using 3.9% directly, to reflect uncertainty given that the 3.9% was based on information for one single laboratory. Compared to the estimate of 50% used in the preliminary analysis, the revised estimate of 10%, if holding all else equal, reduces estimated benefits by a factor of five. With respect to the analysis in section II.E.2, this reflects that order volume likely better represents distinct patients.

###### 4. Comments (Attribution of Diagnostic Error to Analytic Phase of Laboratory Tests)

Comments suggested that the percent of diagnostic errors attributable to faulty diagnostic test results is likely lower than 50%, the estimate we used in the preliminary analysis. A comment suggested instead, based on published literature, a range of 1-4%, with a central estimate of 2.5% (Refs. [3], [4], [5], [6], [7], [8], [9], [10]).

**Response:** Our final analysis uses an updated methodology which no longer directly estimates this parameter. However, our range of estimates of the number of diagnostic errors attributable to faulty tests and resulting in harm is consistent with much rarer attribution of diagnostic errors to tests. This analysis estimates in total about 53,000 annual preventable harms attributable to diagnostic tests (the sums of the primary estimates of avoidable harms across Table 5, Table 11, and Table 14). Singh et al. (2014) estimate that approximately 12 million U.S. adult patients experience diagnostic errors in outpatient care every year (Ref. [11]). Our primary estimate thus represents about half a percent of this total. As our estimate of preventable harms

###### 5. Comments (Percent of IVDs Offered as LDTs that are Problematic)

Comments stated that the percent of IVDs offered as LDTs that are problematic is likely much lower than 47%, the estimate we used in the preliminary analysis. Comments also suggested that it is inappropriate to extrapolate this parameter from as narrow a sample as cited in the preliminary analysis.

**Response:** We agree that a broader basis for estimating this parameter is appropriate and have revised the relevant analysis accordingly, as described in detail in section II.E.1.a “Cancer: Mortality Risk.” Using statistics from the NYS Department of Health Clinical Laboratory Evaluation Program (CLEP) (Ref. [12]) and FDA’s 2020 assessment of EUA requests from laboratories for molecular diagnostic COVID tests, we consider a range of scenarios in which 22%, 38%, and 54% of IVDs offered as LDTs without FDA oversight would be a problematic IVD. This was extrapolated to estimate that 22%, 38%, and 54% of patients tested with IVDs offered as LDTs would be tested with problematic IVDs. Compared to the estimate of 47% used in the preliminary analysis, the revised primary estimate of 38%, if holding all else equal, reduces estimated benefits by approximately one fifth.

###### 6. Comment (IVDs Offered as LDTs Perform Better Than FDA-Authorized Tests)

As further detailed in section VI.C.4 of the preamble, some comments pointed out omission of multiple publications claiming comparable or better performance of IVDs offered as LDTs compared to “FDA IVDs.” Some comments suggested that patients would lose access to IVDs offered as LDTs that perform well, even some IVDs offered as LDTs that may perform better than FDA-authorized IVDs. Comments also suggested adjusting downwards our estimates of benefits from avoiding preventable misdiagnosis-related harms by subtracting harms from baseline problems of FDA-authorized tests.

**Response:** We do not agree that IVDs offered as LDTs generally perform comparably to or better than FDA-authorized tests. Thus, we do not agree that this analysis should reflect such a situation. Concerning the scientific merits of the claims in these comments, please refer to FDA’s responses in section VI.C.4 of the preamble, “Evidence of the Need for Greater FDA Oversight.” In particular, we discuss publications purported to compare the performance of IVDs offered as LDTs and FDA-authorized tests in our response to Comment 34.

With respect to estimating the difference in reliability between problematic IVDs offered as LDTs, specifically, and FDA-authorized competitor tests, we lack systematic data on the exact issues with all problematic IVDs offered as LDTs and their particular uses in the process of diagnosis. As described in section II.E.1.a “Cancer: Mortality Risk,” we consider a range of rates at which avoidable diagnostic error might result from usage of problematic IVDs offered as LDTs that would not occur using an FDA-authorized test.

###### 7. Comments (Effectiveness of FDA Review in Assuring Reliability of Diagnostic Tests)

Citing examples of FDA-authorized tests with alleged issues affecting reliability, comments suggested that our analysis of the proposed phaseout policy overestimated the effectiveness of FDA review in assuring the reliability of diagnostic tests and reducing the use of problematic IVDs offered as LDTs.

**Response:** Unlike in our preliminary analysis, as explained in section II.E.1.a, we now use statistics from the NYS CLEP to inform our estimates of avoidable problematic IVDs offered as LDTs. We believe this is a relevant extrapolation to expected detection of reliability issues through FDA oversight.

###### 8. Comment (Diagnostic Tests Only a Part of Diagnosis)

One comment expressed concern that FDA does not consider that the risks of IVDs offered as LDTs, including erroneous results, "is mitigated by the fact that they are part of a multi-faceted medical assessment and are rarely used in isolation for clinical decision-making.”

**Response:** Due to uncertainty about the rate at which erroneous test results lead to erroneous treatment decisions, this analysis considers that inaccurate results might be identified during follow-up or other parts of the process of diagnosis before leading to harm from diagnostic error. As described in section II.E.1.a “Cancer: Mortality Risk,” we consider a range of rates at which avoidable errors might result from usage of problematic IVDs offered as LDTs that would not occur using an FDA-authorized test. However, although for the purpose of this analysis we consider a wide range for these rates, FDA expects that erroneous test results often result in erroneous treatment decisions. As discussed in the response to Comment 6 in the preamble, FDA does not consider all clinicians to be aware of the limitations of tests. FDA routinely consults with healthcare providers and has encountered many who do not understand the limitations of tests and do not consider that a test result provided by a test may be incorrect. For additional discussion of this comment please see our response to Comment 6 in the preamble.

###### 9. Comments (StatinCheck Problem)

Comments requested an explanation of why we included, in the preliminary analysis, the StatinCheck test for KIF6 genotype as an example of a problematic test.

**Response:** This test was marketed as a way to predict the risk of heart disease and determine a patient’s response to statin drugs, based on the belief that patients with the Trp719Arg polymorphism of the KIF6 protein had an elevated risk of cardiovascular disease (CVD) events and would have a greater reduction in CVD events when on statin therapy than patients without this polymorphism. However, the results from studies of the association between the polymorphism, CVD risk, and statin response were conflicting, and multiple scientific publications reported no association between the polymorphism and elevated CVD risk or statin response (Refs. [13] [14]). Accordingly, the totality of scientific evidence does not support that there is a clinically valid relationship between the polymorphism, elevated CVD risk, and statin response. In 2011, FDA informed the manufacturer that its submission for premarket approval of this test was not approvable stating that the evidence submitted was insufficient to support the test’s safety and effectiveness in determining risk of heart disease or in predicting statin response. As described in FDA’s 2015 report, additional problems included that the test was incorrectly validated and marketed for unproven uses. Inaccurate assessment of patient risk or likelihood of responding to statin therapy could lead to overtreatment, with an associated risk of adverse events, as well as undertreatment, with the risk of failing to prevent CVD events and death.

###### 10. Comments (Non-Invasive Prenatal Screening Tests Require Follow-Up)

As detailed in section VI.C of the preamble, FDA received comments regarding FDA’s use of a New York Times article on NIPS as evidence of a problem (Ref. [15]). Specifically, comments stated that the article conflated screening with diagnostic testing. They asserted that the article mischaracterized false positive results as test failures and that the “problem” with this category of tests is with “the lack of understanding of its purpose and limitations by the providers and patients who were interviewed by the reporters.”

**Response:** FDA agrees that NIPS tests, which may tell people the risk of their fetus having certain genetic abnormalities, are different from diagnostic tests used to more definitively confirm or rule out a suspected genetic abnormality. FDA agrees with comments that NIPS tests should not be used to confirm or rule out a suspected abnormality. While higher false positive rates are often more acceptable for screening tests than tests used for making a diagnosis, appropriate false positive rates for any particular test needs to be considered in the context of a full benefit-risk evaluation for that particular test. After publication of the New York Times article, FDA issued a safety communication to explain the limitations of NIPS tests and provide information to educate both patients and health care providers to help reduce the inappropriate use of NIPS tests.7 Increased oversight of NIPS tests, including labeling requirements, can help ensure such tests are appropriately labeled with transparent information regarding performance, clear instructions, and appropriate limitations. Including the example of NIPS in this final analysis reflects our expectation that phasing out the enforcement discretion approach will ensure tests are appropriately safe and effective for their intended use. We also expect truthful, accurate, and clear statements about test use and performance to prevent patient and provider misunderstanding.

###### 11. Comments (Innovation)

FDA received comments stating that the phaseout policy will have a negative impact on innovation in the testing space, as laboratories working to come into compliance would be either unable or unwilling to engage in innovative test development. Some comments stated that the regulatory constraints associated with the phaseout policy would cause laboratory manufacturers to develop fewer tests, hindering the timely development and deployment of cutting-edge therapies and diagnostic tools and ultimately harming patients. A comment from the Association of Pathology Chairs (APC) stated that it had conducted a survey of its members and found that 92% (36/39) of APC survey respondents reported that there will be less innovation to create and offer new tests to improve patient care due to the FDA’s proposed phaseout policy. Comments claimed that the high cost of premarket review may lead to less investment in innovation, fewer new tests developed, and longer timelines for new innovation to reach the market, and that some tests may not have market viability, given the premarket review costs.

Several comments noted that laboratories must be able to modify existing tests quickly to diagnose new conditions and monitor the impact of new therapies. Some comments stated that stifling modifications of currently marketed IVDs offered as LDTs would force pathologists and other healthcare providers to use older, less optimal tests, and noted that many patients do not have the time to wait for diagnostic development and rely on laboratories to be nimble and adapt to changing diagnostic criteria. One comment asserted that predetermined change control plans (PCCPs) would not help alleviate delays in modifications because only manufacturers can submit PCCPs, and thus laboratories seeking to modify an IVD for local conditions would need to undertake premarket review to do so. One other comment expressed concern that the phaseout policy would lead to slowed growth in the number of LDTs manufactured by laboratories because the phaseout policy would “prohibit” labs from sharing their discoveries about such

**Response:**

As explained in the preamble to the final rule, the phaseout policy is intended to better protect public health by helping to assure the safety and effectiveness of IVDs offered as LDTs—while also accounting for other important public health considerations such as patient access and reliance.

Premarket review is not required for all IVDs offered as LDTs. FDA premarket review is required only for certain tests (generally those in class II or class III). FDA estimates that approximately 50 percent of IVDs offered as LDTs will not require premarket review. A manufacturer's modifications to tests that have already been cleared, approved, or granted marketing authorization by FDA only require FDA review in certain circumstances (see 21 CFR 814.39; 21 CFR 807.81(a)(3)). Even when premarket review is required for an IVD offered as an LDT, FDA does not agree that such review necessarily impairs innovation. In fact, sponsors have sought and obtained FDA authorization for innovative IVDs offered as LDTs. FDA also has several programs that may facilitate the development and premarket authorization of innovative tests.

Moreover, better assuring the safety and effectiveness of IVDs offered as LDTs will foster innovation. By applying the same general oversight approach to laboratories and non-laboratories that manufacture IVDs, FDA’s phaseout policy will remove a disincentive for non-laboratory manufacturers to develop novel tests. We anticipate that phasing out the general enforcement discretion approach for LDTs will spur innovation for IVDs for which there is a reasonable assurance of safety and effectiveness.

###### 12. Comments (Impact on Prices)

Several comments stated that ending the general enforcement discretion approach for LDTs would lead to higher prices for clinical tests due to the costs of complying with applicable FDA requirements. Some comments further stated that the cost of complying with applicable requirements would result in the closure of many laboratories or the outsourcing of certain laboratory tests, which in turn will increase the costs of tests due to decreased test availability, decreased market competition, increased handling costs (e.g., costs associated with shipping samples to a centralized laboratory), or supply chain contractions. One comment expressed skepticism regarding FDA’s statement that any losses may be offset by the market entry of IVDs from other manufacturers. FDA also received a comment which argued that increased prices for clinical tests will disincentivize people from seeking preventive care until they suffer an emergency, which will increase costs for the overall healthcare system. Collectively, these comments suggested that laboratories will pass increased costs to their customers, which some comments noted could result in higher insurance premiums. However, one comment stated that insurance companies will be more likely to cover tests (because they will have FDA authorization), which may allow for greater access to more affordable testing. One comment noted that it is inaccurate to assume that IVDs offered as LDTs are always cheaper.

**Response:**

FDA recognizes that some laboratories may pass the costs of compliance with applicable requirements, including the specific examples listed in the comments, to their customers by raising prices for IVDs offered as LDTs. We also recognize that if many laboratories reduce operations or exit the market, production may be concentrated in a few large laboratories, which may cause prices for certain IVDs offered as LDTs to increase. However, we note that in the final phaseout policy, which will also affect small laboratories and Academic Medical Centers (AMCs), there may be less laboratories that scale back operations or exit the market relative to the estimates in our preliminary analysis.

FDA intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements (except for requirements under 21 CFR 820, subpart M (Records)) for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified in certain limited ways as described in the preamble. FDA also intends to exercise enforcement discretion and generally not enforce premarket review requirements for LDTs approved by NYS CLEP. FDA also intends to exercise enforcement discretion and generally not enforce premarket review requirements and QS requirements (except for requirements under 21 CFR part 820, subpart M (Records)) for LDTs manufactured and performed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system. These enforcement discretion policies may significantly reduce the costs of compliance under the final phaseout policy, thus reducing the number of laboratories that scale back operations or exit the market. In addition, we anticipate that FDA oversight could help to support coverage and reimbursement determinations for IVDs offered as LDTs, which we anticipate will make certain IVDs offered as LDTs for which there is a reasonable assurance of safety and effectiveness more affordable for patients. As a result, FDA does not agree that patients will necessarily be disincentivized from seeking preventive care resulting in increased costs to the healthcare system as a result of the phaseout policy.

In addition, phasing out the general enforcement discretion approach for LDTs will help to reduce other healthcare costs. Greater oversight by FDA will help to address the hidden costs associated with unsafe or ineffective IVDs (including IVDs promoted with false or misleading claims), such as costs incurred from inappropriate treatments, additional or repeat testing, unnecessary consultations with providers, or additional treatment that became necessary due to the progression or worsening of a disease or condition following misdiagnosis. While certain costs may be passed on to individuals and insurers, we expect some of these costs will be offset by the associated benefits.

###### 13. Comments (Increased Labor Cost/Strain)

FDA received comments expressing concern that phasing out the general enforcement discretion approach for LDTs would require laboratories to have increased resources to afford the necessary staffing and other costs related to test development and regulatory submissions and emphasized the thin financial margins with which small laboratories operate. Some comments stated that the impact on small laboratories will result in a loss of expertise and infrastructure. In addition, comments noted that such centralization of IVDs offered as LDTs at large laboratories may negatively impact medical education and training in pathology, resulting in labor shortages. Some comments also suggested that workforce shortages will make it difficult for FDA to recruit and retain adequate numbers of qualified reviewers trained in laboratory diagnostics needed to review premarket submissions, which could potentially lead to delays in FDA’s premarket review process and patient access to tests.

**Response:** FDA appreciates the concerns regarding financial and administrative challenges for smaller laboratories. FDA anticipates that the enforcement discretion policies discussed in section V.B of the preamble will sufficiently address these concerns and help to avoid undue disruption to the testing market. For example, FDA intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements (except for requirements under 21 CFR part 820, subpart M (Records)) for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified as described in section V.B.3 of the preamble, which means laboratories would generally not need to dedicate staff or resources to handle premarket submissions for their existing IVDs offered as LDTs.

It is possible that some laboratories could need additional staff to handle premarket submissions for new IVDs and we account for this in our analysis. However, we expect that FDA’s enforcement discretion policy for currently marketed IVDs offered as LDTs will greatly reduce the volume of submissions from the estimate in our preliminary analysis, thereby avoiding any sudden or drastic increase in labor costs.

###### 14. Comments (Underestimation of Costs)

Several comments stated that costs are substantially underestimated. Some comments elaborated on specific types of costs, especially costs of premarket review. In support of their arguments some comments provided cost estimates for premarket review per entity while others provided cost ranges per test including analytical and clinical validation costs. Another comment focused on how the cost of modifying an SOP could be more burdensome than estimated as it would have to occur for every IVD offered as an LDT. Some comments also conveyed concern that the cost of a possible increase in LDT outsourcing due to FDA’s phaseout of enforcement discretion were not considered. Several comments stated that the costs of hiring new labor to comply with the phaseout of enforcement discretion was underestimated.

**Response:** FDA has revised the phaseout policy to include several enforcement discretion policies for certain types of IVDs as described in section V.B of the preamble. For example, FDA intends to exercise enforcement discretion for premarket review and QS requirements (except for requirements under 21 CFR 820, subpart M (Records)) for IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule and that are not modified, or that are modified as described in the preamble. As such, we do not expect that premarket submissions will be submitted for most currently marketed IVDs offered as LDTs in the immediate future, thus reducing the costs of the phaseout policy including the costs of premarket submission and review. We have revised our estimates consistent with revisions to the phaseout policy as explained in section II.F of this analysis.

###### 15. Comments (Outsourcing Costs and Costs of Switching to FDA-Authorized Tests)

Some comments stated that if FDA phases out the general enforcement discretion approach for LDTs, the commenters may decide to switch from an IVD offered as an LDT to an FDA-authorized test or to outsource their tests to other laboratories. Some comments provided information about the cost differential between an IVD offered as an LDT and an FDA-authorized test or from outsourcing certain tests. Some comments provided estimates on the number or percentage of tests that they would consider outsourcing or switching to an FDA-authorized test.

**Response:** FDA has revised the phaseout policy to include several enforcement discretion policies for certain types of IVDs as described in section V.B of the preamble. For example, FDA intends to exercise enforcement discretion for premarket review and QS requirements (except for requirements under 21 CFR 820, subpart M (Records)) for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule and that are not modified, or that are modified as described in the preamble.
We generally do not expect laboratories with currently marketed IVDs offered as LDTs to switch from an IVD offered as an LDT to an FDA-authorized test or to outsource their tests to other laboratories. We agree, however, that some laboratories may pursue outsourcing their testing needs or switching to use of an FDA-authorized test rather than introducing a new test, that does not fall within an enforcement discretion policy in the phaseout policy. However, as we explain in section II.F.6. of this FRIA, the cost of switching to an FDA-authorized test when available, would cost less than the cost of submitting a premarket submission. A laboratory would likely switch to FDA-authorized tests or outsource their testing needs only if submitting a premarket submission was more costly than either of these alternatives. Either way, the decision would be a private decision made according to their business plan. To the extent that any number of laboratories switch to any number of FDA-authorized tests, their costs would be less than the costs of submitting a premarket submission.

###### 16. Comment (FDA Would Not Have Sufficient Resources To Review IVDs Offered as LDTs)

Some comments expressed concerns that FDA would not have sufficient resources to conduct timely premarket review of IVDs offered as LDTs to meet the public health needs. Some of these comments questioned whether FDA would have adequate capacity to provide timely review of applications/submissions for IVDs offered as LDTs because many EUA requests were not reviewed due to resource limitations during the COVID-19 pandemic. At least one comment cited FDA’s review of a particular EUA request for an LDT during the COVID-19 pandemic, in which FDA’s review of the request did not conclude until after the subject LDT had been removed from the market, as proof that FDA does not have adequate resources to conduct premarket review of IVDs offered as LDTs. Another comment referenced a supposed FDA delay in recognizing a particular consensus standard, based on FDA’s “prolonged review.”

Other comments referenced FDA’s MDUFA IV performance report from FY2020 to 2022 (during the COVID-19 pandemic) and predicted that the increased volume of submissions from laboratory manufacturers that would result from the phaseout policy would affect FDA’s overall ability to review premarket submission for all IVDs, meet its MDUFA performance goals, and conduct other essential work, including policy and post-market activities. Finally, some comments recommended that FDA modify the phaseout policy to prolong the period of time prior to phasing out the general enforcement discretion approach with respect to premarket review requirements, and/or continue to apply the general enforcement discretion approach with respect to premarket review requirements for certain LDTs, to reduce the FDA resource needs.

**Response:**

FDA disagrees that the Agency will lack sufficient resources to conduct premarket review of IVDs offered as LDTs in a timely manner. First, FDA does not intend to phase out the general enforcement discretion approach with respect to premarket review requirements for high-risk IVDs until 3½ years after publication of the phaseout policy, and for moderate- and low-risk IVDs (that require premarket submissions), until 4 years after publication of the phaseout policy. This timeline aligns with the next reauthorization of MDUFA. This alignment will provide an opportunity for FDA and industry to negotiate regarding user fees and performance goals with the knowledge that laboratory manufacturers will be expected to comply with applicable premarket review requirements.

Second, FDA generally intends to exercise enforcement discretion with respect to certain requirements for certain tests as described in the final phaseout policy. These enforcement discretion policies are discussed further in section V.B of the preamble and collectively will significantly reduce the number of premarket submissions for IVDs offered as LDTs, as compared to our preliminary estimates.

Third, FDA’s device authorities require premarket review only for certain IVDs. FDA estimates that approximately 50 percent of IVDs currently under active oversight are low risk and do not require premarket review, and FDA assumes this estimate also applies to IVDs offered as LDTs (see section II.F.2.c of this analysis). However, there are uncertainties surrounding the estimate of total numbers of IVDs offered as LDTs on the market because FDA generally has not enforced the registration and listing requirements for LDTs under section 510 of the FD&C Act (21 U.S.C. 360), 21 CFR part 607, and 21 CFR part 807 (excluding subpart E). By 2 years after publication of this final rule, at stage 2 of the phaseout policy, FDA will obtain registration and listing information from laboratory manufacturers offering IVDs as LDTs. This information will help FDA assess and plan for the resources needed for premarket review of those IVDs before stages 4 and 5 of the phaseout policy.

Fourth, FDA is currently working to enhance its 510(k) Third Party Review Program to handle the review of low- and moderate-risk devices by recognized Third Party review organizations. This will free up Agency staff time to review more complex, innovative, high-risk devices. FDA estimates that half of the IVDs offered as LDTs for which a 510(k) will be submitted will be reviewed under the Third Party review program. FDA also recognizes that if CLIA accreditation organizations seek accreditation under FDA’s Third Party review program, there may be certain efficiencies or other advantages, because the two programs are complementary, as described in response to Comment 7 of the preamble. FDA also intends to exercise enforcement discretion and generally not enforce premarket review requirements for LDTs approved by NYS CLEP and to generally not enforce requirements for LDTs.
Finally, FDA disagrees that decision timelines on EUA requests, in general, are a good indicator to predict FDA’s timelines for review of premarket applications/submissions for IVDs offered as LDTs, and further disagrees that FDA’s review of any one particular EUA request submitted for an LDT during the COVID-19 pandemic is indicative of how FDA will review premarket applications/submissions for IVDs offered as LDTs generally.

As discussed in response to comment 275 in the preamble, FDA’s authority to issue EUAs for LDTs is under a different statutory provision (section 564 of the FD&C Act (21 U.S.C. 360bbb-3)) than traditional premarket reviews. Moreover, FDA is not required to review individual EUA requests submitted to FDA or review them on a specific timeline, or to authorize the emergency use of a medical product even if it meets the relevant criteria for an EUA, giving FDA flexibility to determine how to prioritize its efforts in emergencies to protect and promote public health.

Second, during the COVID-19 pandemic, FDA received a large influx of submissions that had not been anticipated. In the context of the phaseout policy, FDA has estimated the number and type of premarket submissions we can expect in Stages 4 and 5, and annually thereafter, and can prepare for those submissions. See our responses to comments 37 and 275 in the preamble for additional discussion on this topic.

###### 17. Comment (Small Entities)

As discussed in section VI.G of the preamble, FDA received comments expressing concern that phasing out the general enforcement discretion approach for LDTs will put financial and administrative pressure on small laboratories. These comments state that the phaseout of general enforcement discretion could result in laboratory closures and potential monopolies in the testing space. Several comments stated that large laboratories will be able to monopolize LDT processing as they have the resources to afford the necessary staffing and other costs related to test development and regulatory submission. One comment discussed small laboratories within a medical system closing, stating that the removal of pathologists due to this kind of laboratory exit would decrease the quality of patient care.

**Response:** FDA appreciates the concerns regarding financial and administrative challenges for smaller laboratories. Specifically, FDA recognizes that smaller laboratories may face an increase in total cost such that they will exit the market and potentially cause increased testing market concentration. The extent in which smaller laboratories may be disproportionately impacted by the phaseout of the general enforcement discretion approach for LDTs, is dependent on the number of IVD's offered as LDTs per lab.

FDA anticipates that the enforcement discretion policies discussed in the preamble of the final rule will moderate these concerns and help to avoid complete disruption to the test market. As noted in Appendix B – Table 8, the average costs per LDT are smallest for stages 1 through 3 of the phaseout policy representing 10% of costs and up to 59% of affected tests, whereas average costs per LDT for stages 4 and 5 represent 90% of costs affecting 3% of tests. The percentage of tests that may experience costs under stages 4 and 5 will increase as new laboratories and tests enter the market during and after stages 4 and 5, as they will fall within the enforcement discretion policy for currently marketed tests.

However, they may still fall within the scope of other enforcement discretion policies described in the preamble to the final rule, including those for unmet needs and LDTs approved by NYS CLEP. However, in the event that a new lab does not fall within the scope of other enforcement discretion policies, costs under stages 4 and 5 could present as a potential barrier to entry in the LDT market for new laboratories.

In Table B.7, total costs and transfers for all stages of the phaseout policy are estimated to be on average anywhere between 2.5, 5.8 and 16 percent over receipts for all entities. We do not have the information about labs to determine how the average estimates are distributed among the firms (including new firms) according to their size categories.

Also, costs would be higher for a lab that has several IVDs offered as LDTs but sells fewer unit tests whereas costs would be smaller for labs with only one IVD offered as LDTs selling a large number of unit tests. In the same manner, profit margins could be higher for labs with a smaller number of IVDs offered as LDTs but with high volume unit tests sold, compared to labs with a larger number of IVDs offered as LDTs but with low volume units tests sold.

Depending on profit margins with respect to revenue, the costs of this rule may be prohibitive for some small labs, making it more likely that some small entities in this size category will exit the market, reduce operations, sell the business, be subject to acquisitions by larger firms or not enter the market. If profit margins were too small for many small firms considering the costs, it is possible that this rule will be too burdensome for some small entities.

While we do not have the data on profit margins to properly estimate the number of labs that would be adversely impacted by this rule, we estimate that small laboratories make fewer IVDs offered as LDTs than large firms. We estimate that small labs make up 92 percent of all labs, and that they also hold a 24 percent share of IVDs offered as LDTs. With the low number of IVDs offered as LDTs per small lab, it is more likely that the percent of costs over receipts per lab would be closer to our low average estimate 2.7 percent.

With the above referenced revisions to the phaseout policy, we do not expect significant disruptions to access to IVDs offered as LDTs, significant increases in test prices, or delays in diagnosis and treatment. However, the high cost of pre-market approval also makes innovation less likely to come from smaller labs.

###### 18. Comment (Firm Exit and Market Concentration)

Some comments have claimed that certain laboratories, such as academic, small, public health, and specialty laboratories, will disproportionately exit the market relative to their counterparts. These comments claim that since these labs already have low revenues, any additional cost could be enough to cause market exit.

Other comments stated that by reducing the availability of IVDs offered as LDTs through market exit, the phaseout policy would lead to delays in testing, including by potentially increasing reliance on reference laboratories which may increase the time individuals obtained test results. Some comments expressed concern that the laboratories surviving general enforcement discretion phaseout would receive an influx of test orders and may not be able to handle the test volume, which may have an overall negative impact on the turnaround time for test results. Two comments have addressed the exit of academic labs as particularly concerning given their role in diagnosing and monitoring rare diseases. Further, many comments stated that a potential increase in firms exiting the market could increase unemployment among laboratory technicians and increase market concentration in the healthcare industry.

A few comments also addressed children’s hospital laboratories as particularly likely to be negatively affected by the phaseout policy. These comments stated that because these hospitals rely largely on Medicaid payment, the laboratories in these hospitals may have small revenue levels and may reduce their IVD offerings, exit the market, or send samples to other laboratories for testing if the total cost increase for meeting applicable requirements that result from the phaseout policy is too high.

**Response:** FDA appreciates the comments on potential firm exit and market concentration as a result of the phaseout policy. Given that FDA intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements (except for requirements under 21 CFR 820, subpart M (Records)) for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule and that are not modified, or that are modified as described in the preamble, and for LDTs developed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system (as described in section V.B.3 of the preamble), we do not expect significant market concentration and firm exit to result from the phaseout policy. With the above referenced revisions to the phaseout policy, we also do not expect current disruption to access to IVDs offered as LDTs, significant increases in test prices, or delays in diagnosis and treatment.

##### D. Summary of Changes from the Proposed Rule

Compared to the preliminary economic analysis, this final analysis reflects revisions to the phaseout policy and to our analytical methodology. We include updates and revisions to our discussion of baseline conditions, estimated health and non-health benefits, costs, budgetary impacts, transfers, regulatory alternatives, and impacts to small entities as summarized below.

###### 1. **Changes to the Phaseout Policy**

   The final phaseout policy differs significantly from the proposed policy in that it includes the following additional enforcement discretion policies:

   - FDA intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements, with the exception of requirements under 21 CFR part 820, subpart M (Records), for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified in certain limited ways as described in section V.B.3 of the preamble.

   - FDA intends to exercise enforcement discretion and generally not enforce premarket review requirements for LDTs approved by NYS CLEP;

   - FDA intends to exercise enforcement discretion and generally not enforce premarket review requirements and QS requirements, with the exception of requirements under 21 CFR part 820, subpart M (Records), for LDTs manufactured and performed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system;

   - FDA intends to exercise enforcement discretion and generally not enforce requirements for LDTs manufactured and performed within the Veterans Health Administration (VHA) or the Department of Defense (DoD); and

   - FDA intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements, with the exception of requirements under 21 CFR part 820, subpart M (Records), for non-molecular LDTs for rare red blood cell antigens where such tests are manufactured and run in transfusion services and immunohematology laboratories and where there is no alternative available to meet the patient’s need for a compatible blood transfusion.

   Where relevant, we adjust estimates in this final analysis in accordance with these changes to the phaseout policy. More details of IVDs within the scope of the phaseout policy are described in section V.A of the preamble.

###### 2. Baseline Conditions

   After reviewing comments, FDA revised the number of laboratories affected by the phaseout policy using data from the CMS Laboratory Registry as explained in appendix A.
Using the CMS data, the revised primary estimate of affected laboratories is 1,181, which is close to the estimate of 1,200 in the preliminary analysis.

###### 3. Benefits

We have made several changes to our analysis of health benefits. While there are individual changes that increase as well as decrease estimated benefits, overall total estimated benefits have decreased due to incorporating new information and assumptions. However, while including data and information from public comments as well as additional research has lowered estimated annualized benefits from expected reductions in cancer mortality to about one fiftieth of the preliminary estimate, we now also use this information to quantify benefits that we previously discussed qualitatively or only addressed incompletely. In particular, we now include general, yearly estimates of mortality avoidable in cardiovascular disease and morbidity avoidable in infectious diseases.

In response to comments, and as described in section II.E.1.a of this analysis, we have revised our methods for estimating avoidable harms related to cancers. We have also applied revised methods in newly estimating avoidable harms related to cardiovascular and infectious diseases. Revisions concern both reference information and analytical assumptions.

We now source certain parameters addressed in public comments, such as the rate of usage of IVDs in cancer, cardiovascular diseases, and infectious diseases, from published literature and other data. Additionally, we now estimate the rate of usage of problematic IVDs offered as LDTs based on application review statistics from the NYS Department of Health Clinical Laboratory Evaluation Program (CLEP) (Ref. [12]) and FDA’s 2020 assessment of EUA requests from laboratories for molecular diagnostic COVID tests.^9

We have also refined analytical assumptions. For example, although we did not assume in the preliminary analysis that 100% of uses of problematic IVDs offered as LDTs result in a harm from diagnostic error, we have refined our assumptions to avoid any implication that this is the case.

Due to our high degree of uncertainty about several of the parameters used to estimate health benefits, we now use Monte Carlo simulations to determine a plausible range for benefits pertaining to each disease category by allowing parameters to vary independently of each other. Not using such an approach would implicitly convey a strong and unrealistic assumption that all uncertain parameters share a joint probability distribution and are perfectly dependent (i.e., aligning all best- and worst-case scenarios across parameters).

With respect to non-health benefits, we have removed discussion and estimates made in section II.E.3 of the Preliminary Regulatory Impact Analysis (PRIA) (Ref. [1]) where we had previously requested supporting information in public comment and did not receive any. Additionally, we have edited discussion for clarity, in part to address certain public comments.

Finally, as described in sections II.E.1 and II.E.3, we adjust estimates of benefits to account for existing review by NYS CLEP and the enforcement discretion policy for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule.

###### 4. Costs

We have made several changes to our cost analysis. We use updated data for wages, FDA costs, and MDUFA fees. We made the following revisions as a result of changes to the final phaseout policy: exclude currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule from premarket review and QS (except for records) compliance costs; exclude LDTs manufactured and performed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system from premarket review and QS compliance costs (except for records); and exclude LDTs expected to be reviewed by NYS CLEP from premarket review compliance costs. We also consider that some of the IVDs currently classified in class III, requiring PMAs/PMA supplements, may be reclassified to Class II or Class I (described in Appendix A and section II.F.4).

We have revised our estimates of FDA review costs to adjust the FTE weight of PMA review costs, to include review costs of MDRs, IDEs, and Q-submissions, and to consider premarket submissions that will be reviewed by third party or NYS CLEP (see section II.G). We have also refined some estimates in the analysis based on data received in public comments.

###### 5. Regulatory Alternatives

The section on regulatory alternatives in the final analysis retains only the first two regulatory options from the preliminary analysis. For the final analysis, we include as an additional alternative the phaseout policy as initially proposed. See section II.J.

#### II. Final Economic Analysis of Impacts

##### A. Background

In 1976, the Medical Device Amendments (MDA) amending the FD&C Act created a comprehensive system for the regulation of devices intended for human use, including IVDs. Since 1976, FDA has considered IVDs to be devices within the meaning of the device definition in the FD&C Act (see section 201(h)(1) of the FD&C Act (21 U.S.C. 321(h)(1)); 21 CFR
However, in implementing the MDA since 1976, FDA has exercised enforcement discretion such that it generally has not enforced applicable legal requirements with respect to most LDTs because they mostly:

- were manufactured in small volumes by local laboratories that served their local communities;
- were typically intended for use in diagnosing rare diseases or other uses to meet the needs of a local patient population or were generally similar to well-characterized, standard IVDs;
- tended to employ manual techniques (and did not use automation) and were performed by laboratory personnel with specialized expertise;
- were to be used and interpreted by physicians or pathologists in a single institution responsible for the patient (and who were actively involved in patient care); and
- tended to be manufactured using components legally marketed for clinical use, such as general purpose reagents or immunohistochemical stains marketed in compliance with FDA requirements.

This enforcement discretion approach for LDTs developed as a matter of general practice.

However, since 1976, the development and usage of LDTs have evolved considerably. LDTs are now more complex, sometimes including proprietary algorithms. Today’s LDTs are also used more widely, by a more diverse population, with an increasing reliance on high-tech instrumentation and software, and more frequently for the purpose of guiding critical healthcare decisions. They are often performed in large volumes in reference laboratories for patients from different institutions around the world and are sometimes assembled using components intended for research use only. Some LDTs are manufactured by corporations that market the IVDs nationwide as they accept specimens from patients across the country and run their tests in very large volumes in a single laboratory. In this regard, most LDTs today are similar to other IVDs that have not been under FDA’s general enforcement discretion approach.

Clinical laboratory tests are foundational to healthcare. The Centers for Disease Control and Prevention (CDC) estimates that 70 percent of medical decisions are based on laboratory test results (Ref. [16]). IVDs offered as LDTs are a growing sector of that market (Ref. [17]). Given the role these tests play in modern healthcare, their safety and effectiveness significantly impact public health. Although many of the IVDs offered as LDTs today are similar to other IVDs and may often serve the same role in clinical practice, FDA has generally not enforced applicable device requirements for LDTs. As a result, there is generally less assurance of the safety and effectiveness of IVDs offered as LDTs compared to other IVDs.

##### B. Need for the Rule

As the growing number of IVDs offered as LDTs entering and currently on the market (some of which may be problematic IVDs) typically are not reviewed by FDA, patients might be at risk when their providers rely on certain IVDs offered as LDTs to guide their care. Results from problematic IVDs can lead to delayed diagnosis or treatment of the true disease or condition, unwarranted interventions (some of which may carry risk of serious side effects), needless distress, progression of disease (in some cases costing the opportunity for life-saving treatment), and the spread of infectious diseases.

While laboratories that offer IVDs are regulated by the Centers for Medicare & Medicaid Services (CMS) under the Clinical Laboratory Improvement Amendments of 1988 (CLIA), CLIA addresses laboratory operations and personnel qualifications and not the development of individual tests in a laboratory (see 42 U.S.C. 263a and 42 CFR part 493) (Ref. [18]). In particular, under CLIA, CMS does not:

- regulate laboratory test development;
- evaluate the performance of an IVD before the test is offered to patients and healthcare providers;
- assess clinical validity (i.e., the accuracy with which a test identifies, measures, or predicts the presence or absence of a clinical condition or predisposition in a patient);
- regulate certain manufacturing activities, such as design controls and acceptance activities;
- provide human subject protections for patients who participate in IVD clinical research trials; or
- require adverse event reporting.

By contrast, the device provisions of the FD&C Act and FDA’s regulations focus on the safety and effectiveness of the IVDs themselves. Given this distinction, CMS has described the FDA and CMS “regulatory schemes” as “different in focus, scope and purpose, but […] intended to be complementary (Ref. [19]).”

FDA’s experience with emergency use authorization (EUA) requests from laboratories for COVID-19 tests during the COVID-19 pandemic increased FDA’s concerns about the safety and effectiveness of IVDs offered as LDTs.[^12] While FDA had received requests for EUAs for tests from laboratories in prior emergencies, the scope of the COVID-19 pandemic resulted in an unusually high number of EUA requests from laboratories, revealing the approach that many laboratories might take to test validation. In an analysis of the first 125 EUA requests received from laboratories during the COVID-19 pandemic for molecular diagnostic tests, FDA found that 82 tests had design or validation problems, or both. The tests involved relatively well-understood techniques and the laboratories represented these tests as appropriately validated.[^13] To the extent that this sample represents larger trends in the performance of IVDs offered as LDTs, it indicates the need for greater oversight.

Problems with IVDs offered as LDTs have also come to light in the scientific literature, news articles, and anecdotal reports submitted to the Agency, among other sources. Multiple publications in the scientific literature have described a high degree of variability among IVDs offered as LDTs (Ref. [20]). For example, in one study, analytical accuracy was significantly lower than that of the parallel test approved by FDA for almost half of the tests studied (Ref. [21]).

News and other outlets have also reported on problems with IVDs offered as LDTs, including the New York Times (Ref. [15]), and lawsuits have been filed relating to pharmacogenomic and non-invasive prenatal screening IVDs offered as LDTs.[^14]

FDA has received complaints, allegations, and reports regarding IVDs offered as LDTs for oncology, non-invasive prenatal screening, and infectious diseases, among others. Some laboratories have submitted data to FDA in premarket submissions for their IVDs offered as LDTs, and we have observed that many failed to perform the appropriate studies to show that their IVDs work. Some have submitted data from appropriate studies, but the data show that their IVDs do not work. In both cases, laboratories have continued to offer such IVDs for clinical use.

While it is theoretically possible that, over time, patients and providers might learn the differences between competing tests and eventually stop purchasing ineffective tests regardless of regulation, that has not universally happened to date, even though the disparity between IVDs offered as LDTs and IVDs meeting applicable FDA requirements has been ongoing for decades. Further, we know from experience that providers and patients often do not even know what test was performed by a laboratory and, without widespread awareness of the different types of tests and regulatory disparities, we expect that learning of this kind would be rare, if it ever occurred, and would be complicated by the rapidly changing market, with new tests introduced regularly. Moreover, during the time that it would take for any such learning to occur, providers and patients may be using inaccurate or unreliable tests, with all the associated risks to patients. As for patients, ability to internalize the relevant risks may be precluded by not knowing the difference between IVDs offered as LDTs and FDA-authorized IVDs or having meaningful informed choice in the purchase decision.

Furthermore, FDA is aware that some entities have adopted business practices that claim a connection to laboratories in order to offer IVDs as LDTs, even when they are not LDTs, because they are not actually designed, manufactured, and used within a single laboratory (See for example Refs. [22] and [23]). For example, FDA notes:

- Manufacturers offering unauthorized home specimen collection kits manufactured outside of the laboratory for use with LDTs;
- Software developers offering software for high-risk clinical use with LDTs through laboratory partnerships;
- Laboratories offering test kits previously alleged to be “research use only” test kits;
- Manufacturers of home specimen collection kits with consumer facing platforms providing the ordering and resulting interface while outsourcing testing to unspecified laboratories; and
- Contract manufacturers claiming to be consulting firms that design and validate tests for customer laboratories to perform.

This puts non-laboratory, conventional test manufacturers who develop IVDs, whose IVDs have not been under FDA’s general enforcement discretion approach for LDTs, at a competitive disadvantage compared to laboratory manufacturers of IVDs offered as LDTs. IVD manufacturers who are not laboratories might currently be discouraged from investing time and resources into developing novel tests due to the concern that once the manufacturer receives marketing authorization for its test, laboratories will develop similar tests and market their tests without complying with FDA requirements (Refs. [24] and [25]). We anticipate that applying the same general oversight approach to laboratories and non-laboratories that manufacture IVDs, and phasing out the general enforcement discretion approach for LDTs, will better assure the safety and effectiveness of IVDs offered as LDTs, and remove a disincentive for non-laboratory manufacturers to develop novel tests, thereby spurring innovation and access to IVDs for which there is a reasonable assurance of safety and effectiveness. Without the phaseout policy, and without better assurance of the safety and effectiveness of IVDs offered as LDTs, limited investment and healthcare funding may be expended on improving problematic IVDs.

The enforcement discretion approach for LDTs has created distortions in the diagnostics market.15 These distortions not only complicate understanding the IVDs used in clinical practice, impeding FDA’s ability to ensure the safety and effectiveness of IVDs, but might also disincentivize high standards of quality control and accuracy and thus entail social costs.16

In order to curtail offering of problematic tests, FDA is phasing out the general enforcement discretion for LDTs so that IVDs manufactured by a laboratory will generally fall under the same enforcement approach as other IVDs.

In addition, to ensure clarity and understanding by industry and the public, FDA is amending its regulations to make explicit that IVDs are devices under the FD&C Act including when the manufacturer of the IVD is a laboratory.

##### C. Purpose of the Rule

The purpose of the rule, which amends 21 CFR part 809, is to make explicit that IVDs are devices under section 201(h)(1) of the FD&C Act (21 U.S.C. 321(h)(1)) including when the manufacturer of the IVD is a laboratory. This amendment will reflect the fact that the device definition in the FD&C Act does not differentiate between entities manufacturing the device, and will provide further clarity to stakeholders affected by the accompanying changes to FDA’s general enforcement discretion approach for LDTs.

In addition, as discussed in section V of the preamble to the rule, FDA is also phasing out its general enforcement discretion approach for LDTs so that IVDs manufactured by a laboratory will generally fall under the same enforcement approach as other IVDs (i.e., FDA’s expectations for compliance will generally be the same). This phaseout policy includes limited enforcement discretion policies for specific categories of IVDs manufactured by a laboratory, including currently marketed IVDs offered as LDTs and LDTs for unmet needs. This phaseout policy is intended to better protect the public health by helping to assure the safety and effectiveness of IVDs offered as LDTs, while also accounting for other important public health considerations such as patient access and reliance. In addition, by applying the same general oversight approach to laboratories and non-laboratories that manufacture IVDs, FDA will give stakeholders more stability, clarity, and confidence, and facilitate investment in the development of innovative IVDs.

For additional discussion, see section III.B of the preamble.

##### D. Baseline Conditions

We consider the current environment, including the general enforcement discretion approach, as a reasonable approximation of the baseline (the projected future without phasing out FDA’s general enforcement discretion approach for LDTs) against which to measure the costs and benefits of the phaseout policy and the regulatory alternatives discussed in section II.J.

FDA has generally described LDTs as IVDs that are designed, manufactured, and used in a single laboratory that is certified under CLIA and that meets the regulatory requirements under CLIA to perform high complexity testing (Ref. [26]).17 However, as discussed in the preamble and section II.F “Costs,” the phaseout policy will affect not only LDTs, but IVDs manufactured and offered as LDTs, even if those IVDs do not fall within FDA’s traditional understanding of an LDT because they are not designed, manufactured, and used within a single laboratory.18, 19 Throughout this document, we refer to these IVDs as “IVDs offered as LDTs.”

As described in section V of the preamble, FDA is including various enforcement discretion policies with regard to all applicable requirements for certain categories of tests manufactured by laboratories. One such category of tests is referred to in this document as “1976-Type LDTs.” Such tests have the following characteristics common among LDTs offered in 1976 (discussed in section III of the preamble):

- use of manual techniques (without automation) performed by laboratory personnel with specialized expertise;
- use of components legally marketed for clinical use; and
- design, manufacture, and use within a single CLIA-certified laboratory that meets the requirements under CLIA for high complexity testing.

FDA will also continue the general enforcement discretion approach for Human Leukocyte Antigen (HLA) tests that are designed, manufactured and used in a single CLIA-certified, high-complexity histocompatibility laboratory that meets the requirements to perform high-complexity histocompatibility testing when used:

- in connection with organ, stem cell, and tissue transplantation to perform HLA allele typing,
- for HLA antibody screening and monitoring, or
- for conducting real and “virtual” HLA crossmatch tests.

FDA will also continue the general enforcement discretion approach for tests intended solely for forensic (law enforcement) purposes. This approach has been in place for over 20 years and applies to such tests regardless of whether they are offered as an LDT.

FDA also intends to continue the general enforcement discretion approach for LDTs manufactured and performed within DoD or VHA. To meet the needs of their patient populations (i.e., military personnel, veterans, and their families) and fulfill their mandates, DoD and VHA often manufacture unique LDTs, such as tests for diseases or chemicals to which their patients may be exposed while serving abroad but which do not exist at home.

We lack information to quantify the number of tests that fall in the above categories and thus the exclusions are not assessed in this regulatory impact analysis.

FDA also intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements, with the exception of requirements under 21 CFR 820, subpart M (Records), for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule that are not modified, or that are modified in certain limited ways. FDA generally expects compliance with premarket review and QS requirements when a laboratory changes the indications for use of the IVD, alters the operating principle of the IVD (e.g., changes in critical reaction components), includes significantly different technology (e.g., addition of artificial intelligence / machine learning to the test algorithm, a change from targeted sequencing to whole genome sequencing, a change from immunoassay to mass spectrometry, or a change from manual to automated procedures), or adversely changes the performance or safety specifications of the IVD.

In addition, FDA intends to exercise enforcement discretion and generally not enforce premarket review and QS requirements, with the exception of requirements under 21 CFR 820, subpart M (Records), for LDTs manufactured and performed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system. For the purpose of this phaseout policy, FDA considers an LDT to be for an unmet need where there is no available FDA-authorized test that meets the patient’s needs. This may be because – (1) there is no FDA-authorized IVD for the disease/condition; (2) there is an FDA-authorized IVD for the disease/condition but it is not indicated for use on the patient, or a unique attribute needs to be added to the IVD to meet the patient’s needs; or (3) there is an FDA-authorized IVD but it is not available to the patient. This is discussed further in section V.B.3 of the preamble.

FDA also intends to exercise enforcement discretion and generally not enforce premarket review requirements for LDTs approved by NYS CLEP.

We expect that a number of laboratories offering IVDs as LDTs, and those IVDs, do not currently meet applicable requirements—including premarket review, quality system, registration and listing, and adverse event reporting requirements—given FDA’s general enforcement discretion approach for LDTs.

We do not have complete information about IVD performance or patient harm. As discussed in detail in section III.B of the preamble, FDA has increasingly seen problems with IVDs offered as LDTs that have caused or might be causing harm. However, the tests involved likely do not represent all problematic tests that might be affected by the phaseout of the general enforcement discretion approach for LDTs, as laboratories do not typically submit premarket submissions for IVDs offered as LDTs to FDA or report adverse events associated with those tests given the general enforcement discretion approach.

Without registration and listing information, it is difficult to estimate the exact baseline number of manufacturers of IVDs offered as LDTs that will be affected by the phaseout policy. It is also difficult to estimate the number of IVDs offered as LDTs currently on the market, when or why many of them are used, or exactly how they each perform compared to other IVDs.

Without adverse event reporting or other information that FDA will obtain upon the phaseout of the general enforcement discretion approach, it is difficult to estimate the exact baseline number of patients that can benefit from the phaseout of the general enforcement discretion approach given current information. In order to account for potential uncertainty and variability, we present all expected costs and benefits in ranges of low, central, and high estimates. We address baseline risks (and costs due to risks) in the benefits section of this analysis.

###### 1. Number of Affected Entities
Since laboratories that offer IVDs as LDTs have not generally registered and listed, we do not know the exact number of laboratories or IVDs offered as LDTs that will be affected by the phaseout of the general enforcement discretion approach for LDTs.

Comments suggested we use CMS data to estimate the number of affected labs. Using the CMS data, we estimate that there are 11,808 high complexity CLIA laboratories that have IVDs offered as LDTs that will be affected by the phaseout of the general enforcement discretion approach. We therefore revise our original estimate of 12,000 to 11,808. The steps in developing this estimate are explained in appendix A. Laboratories that meet the requirements to perform high complexity testing are the only laboratories that can perform LDTs under CLIA regulations, because LDTs are considered high complexity tests (Ref. [27]). Additionally, while CLIA regulations contemplate that such laboratories may deploy IVDs offered as LDTs, we do not expect that every such laboratory does so. We are not aware of suitable sources for the exact number of such laboratories that are currently offering IVDs as LDTs.

We rely on information about laboratories and IVDs in NYS to estimate the percent of high complexity labs that make IVDs offered as LDTs (Ref. [28]). NYS requires laboratories offering tests to NYS residents, whether or not the laboratory is located in NYS, to obtain a permit through the NYS CLEP, as well as “explicit test-specific approval” for certain IVDs that are not “designated as FDA-cleared, approved or exempt.” (Ref. [29]) To FDA’s knowledge, NYS is the only state that requires approval for LDTs that are not FDA-cleared, approved, or exempt. Further, NYS is a relatively large space with a variety of demographics, including urban to rural areas, and a variety of laboratories such as academic medical centers, reference laboratories, public health laboratories, and local hospital laboratories, similar to the variety found throughout the U.S. Therefore, FDA determined that the information about laboratories and IVDs in NYS could be extrapolated to estimate the number of laboratories throughout the U.S. that might be offering IVDs as LDTs.

NYSDOH provided information indicating that there are approximately 500 laboratories located in NYS with a NYS CLEP permit that are certified under CLIA and that meet or exceed the regulatory requirements under CLIA to perform high complexity testing, and that approximately 50 of such laboratories offers at least one IVD as an LDT approved by NYS CLEP (Ref. [28]). From these data, we calculate that approximately 10% of laboratories located in NYS that are certified under CLIA and that meet the regulatory requirements under CLIA to perform high complexity testing are manufacturing IVDs offered as LDTs.

For our primary estimate, we assume that NYS is representative of the U.S. laboratory community, as discussed above. Based on the information from NYS and the assumption that NYS is representative of the entire U.S., we estimate that approximately 10% of 11,808 (or 1,181) laboratories in the U.S. that are certified under CLIA and that meet the regulatory requirements under CLIA to perform high complexity testing currently manufacture IVDs offered as LDTs. To account for potential variability across the country, we estimate the proportion of high complexity laboratories making IVDs offered as LDTs to vary from 5% of 11,808 (or 590) laboratories to a high estimate of 20% of 11,808 (or 2,362) affected laboratories by reducing the primary estimate by 50% and doubling the primary estimate, respectively.

Based on these two sources and methods, for purposes of this analysis, we use 590, 1,181 and 2,362 as low, central (primary), and high estimates of the number of laboratories affected by the phaseout of the general enforcement discretion approach for LDTs. We also expect that there will be new laboratories entering the market every year. To calculate the number of new laboratories per year, we use an average of firms’ entry and exit rates from 2010 to 2018 in the United States (approximately 8 percent) (Ref. [30]). Multiplying this by the number of affected entities, we estimate the number of new laboratories per year to range from 47 to 189, with a primary estimate of 94.^20^

Because there is no single source containing information on the number of IVDs offered as LDTs currently on the market, FDA also used information about laboratories and IVDs reviewed in NYS to extrapolate estimates for affected IVDs across the country. According to NYSDOH’s website, there are currently approximately 2,200 IVDs with approval from NYSDOH offered by laboratories located in NYS (Ref. [29]). NYSDOH provided the number of distinct laboratories within NYS that are certified under CLIA, that meet the regulatory requirements under CLIA to perform high complexity testing, and that are manufacturing and offering at least one IVD offered as an LDT (Ref. [28]), as well as the breakdown of risk categories for submissions to NYS, as determined by NYS CLEP risk criteria. From these data, FDA calculated that each laboratory in NYS that manufactures IVDs offers an average of 67 IVDs as LDTs. Extrapolating to the rest of the country, FDA estimates that 39,557, 79,114, or 158,227 IVDs may be currently offered as LDTs and therefore affected by the phaseout of the general enforcement discretion approach, based on the low, central, and high estimates of affected entities discussed above (see Table 2). These estimates assume that NYS is representative of the U.S. laboratory community.

We took a similar approach to estimating the number of new IVDs offered as LDTs that are expected to be introduced per laboratory per year. NYSDOH provided information indicating that laboratories within NYS that manufacture IVDs offered as LDTs introduce an average of 6 new IVDs offered as LDTs per year. For purposes of this analysis, we assume that laboratories in NYS are representative of the U.S. laboratory community, and estimate that 3,542, 7,085, or 14,170 new IVDs offered as LDTs may be affected per year. We also expect that there would be new IVDs offered as LDTs from new laboratories entering the market every year. In addition, we expect 50 percent of currently marketed IVDs offered as LDTs (34 IVDs offered as LDTs per laboratory = 67 * 0.5) will be modified in such a way as to require premarket review over the next twenty years. We thus estimate 2 IVDs offered as LDTs per year per laboratory will be modified in such a way as to require premarket review (2 = 34 / 20 years). The total number of new IVDs offered as LDTs per year is estimated to range from 5,007 to 20,026, with a primary estimate of 10,013. We understand anecdotally that some large reference laboratories may make as many as 100 new IVDs per year, whereas smaller or more specialized laboratories may focus on one or a few IVDs overall and may not introduce many or any new IVDs every year. Throughout this analysis, we define the terms “affected labs” and/or “affected entities” as laboratories offering IVDs as LDTs and therefore affected by the phaseout of the general enforcement discretion approach for LDTs. In a similar manner, we also define the terms “affected IVDs” as IVDs offered as LDTs associated with costs as incurred during their relevant policy stages 1 through 5 (discussed in detail in section II.F).

Table 2 shows the estimated number of laboratories and IVDs offered as LDTs affected by the phaseout of the general enforcement discretion approach for LDTs.

**Table 2. Estimated Number of Laboratories and IVDs Offered as LDTs Affected by this Rule**
|   | Primary Estimate | Low Estimate | High Estimate |
|---|------------------|--------------|---------------|
| Affected Labs | 1,181 | 590 | 2,362 |
| New Affected Labs Entering the Market Per Year | 94 | 47 | 189 |
| Affected Tests Currently on the Market | 79,114 | 39,557 | 158,227 |
| New Affected Tests Per Year | 10,013 | 5,007 | 20,026 |
||

*Notes: Product across table may not be exact due to rounding. The number of new affected IVDs per year include currently marketed IVDs that would be modified and new affected IVDs from both affected labs and new labs entering the market per year. These numbers reflect the baseline numbers for affected laboratories and affected tests and are further adjusted in later sections of this analysis to estimate costs under Stages 4 and 5, where only a subset of these laboratories and tests may incur costs, as described in sections II.F.4 and II.G.*

###### 2. Baseline Market Revenue

Data from the 2017 U.S. Census for the entire industry under NAICS code 621511 reported 3,365 firms with $36 billion in annual revenues. From Table 2 above, we estimate 1,181 affected laboratories or firms which represent 35% of the 3,365 firms in the Census data. If we assume the same average annual receipts for all firms, then the corresponding annual receipts for affected laboratories would represent 35% of total annual receipts or $15 billion (in 2022 dollars).

In Table 3, we estimate annual industry revenue in 2023 between $19 and $21 billion based on a projection from 2017 Census data of the $15 billion using compounded annual growth rates (CAGRs) of 4.2% and 6% (Refs. [17] [31]).

**Table 3. Estimated Market Revenue for IVDs Offered as LDTs ($1,000, 2022 U.S. dollars)**
| Year | Primary (Average between low and high projection) | Low Projection ($1,000) (4.2% CAGR) | High Projection ($1,000) (6% CAGR) |
|------|--------------------------------------------------|------------------------------------|----------------------------------|
| 2023 | $20,093,935 | $19,062,398 | $21,125,471 |
| 2030 | $28,594,674 | $25,424,450 | $31,764,898 |
||

###### 3. Baseline FDA Premarket Reviews of Submissions/Applications

To better understand the magnitude of anticipated premarket submissions/applications for IVDs offered as LDTs that FDA would receive on an annual basis, Table 4 below shows the 5 year average number of submissions/applications for all devices (2017-2021) along with the estimated annual number of submissions/applications expected for IVDs offered as LDTs after this rule become effective (Ref. [32]). The estimated annual reviews for premarket submissions/applications are adjusted to account for the enforcement discretion policies discussed above, potential reclassifications of class III IVDs to class II, and the 510(k) Third Party Review Program.

**Table 4. FDA Review Workload by Submission Type**
| Submission/Application Type                  | 5-Year Average (FY 2017 to 2021) | Expected Annual Reviews for New IVDs Offered as LDTs |
|----------------------------------------------|----------------------------------|------------------------------------------------------|
|                                              |                                  | Primary | Low | High |
| Original PMAs, PDPs, Panel-Track PMA Supplements | 73                               | 103*   | 52*  | 206* |
| 510(k) Premarket Notifications               | 3,877                            | 1,090  | 545  | 2,179 |
| De Novo Classification Requests              | 66                               | 267    | 134  | 534   |
||

Note: FDA annual reviews of Q-submissions, MDRs and IDE applications are not included in Table 4 but are estimated in section II.G Budgetary Impacts and described in Table 38.  
\* The estimated reviews include original PMAs and panel-track PMA supplements. See Table 29 and Table 31 for estimated numbers of original PMAs and panel-track PMA supplements, respectively. Totals may not add due to rounding.

###### 4. Baseline Risk of Problematic IVDs

We measure benefits of phasing out the general enforcement discretion approach for LDTs against a baseline scenario in which FDA continues the general enforcement discretion approach with respect to all applicable requirements for all IVDs offered as LDTs. Due to the current general enforcement discretion approach for LDTs, we lack systematic data on the exact incidence of harms specifically resulting from usage of problematic IVDs offered as LDTs that can be avoided by phasing out the general enforcement discretion approach for LDTs.

To estimate the baseline incidence of harms that can be avoided by phasing out the general enforcement discretion approach for LDTs, we focus mainly on three broad disease categories identified by Newman-Toker et al. (2021) as accounting for about 75% of serious misdiagnosis-related harms in the U.S.: cancers, cardiovascular disease, and infections (Ref. [33]). As described in the following section, we multiply the numbers of U.S. patients in each disease category—estimated from sources described below—by rates from literature of the proportions of patients in each category that are tested using IVDs. Based on Rychert et al. (2023) (Ref. [2]), as described in section II.E.1, we estimate that 3.9% to 45% of patients tested with IVDs are tested with IVDs offered as LDTs. This yields our baseline estimates of the numbers of patients tested using IVDs offered as LDTs. To estimate usage specifically of problematic IVDs that can be curtailed by phasing out the general enforcement discretion approach for LDTs, we refer to statistics from NYS CLEP on application review outcomes and FDA’s 2020 assessment of EUA requests from laboratories for molecular diagnostic COVID tests. Since we do not expect that LDTs approved by NYS CLEP would undergo FDA premarket review, when estimating numbers of U.S. patients in each disease category, we exclude from this analysis the number that we attribute, proportionally by population, to New York state. We note however for LDTs approved by NYS CLEP, FDA intends to exercise enforcement discretion with respect to premarket review requirements but not other applicable requirements under the FD&C Act and FDA regulations.

Lacking systematic data on the exact issues with applications for LDTs initially rejected by NYS CLEP and those LDTs’ roles in the process of diagnosis, we consider a range of rates at which avoidable errors might result from usage of problematic IVDs offered as LDTs.

Finally, in section II.E.3 “Summary of Benefits,” we adjust estimated total benefits to account for the enforcement discretion policy for IVDs offered as LDTs that are already currently on the market. As a result of exercising enforcement discretion with respect to premarket review and QS requirements, with the exception of requirements under 21 CFR 820 subpart M (Records), for currently-marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule, IVDs offered as LDTs will generally undergo premarket review only in certain circumstances. However, FDA expects these IVDs to be in compliance with other applicable requirements under the FD&C Act and FDA regulations, including post market requirements, as discussed in the phaseout policy. In order to account for this, we adjust estimated total benefits in each year—beginning with enforcement of QS and premarket review requirements—based on the proportions of IVDs falling within an enforcement discretion policy versus IVDs not falling within an enforcement discretion policy. We note, however, as described in section V.B.3 of the preamble, FDA intends to take targeted steps to address currently marketed IVDs offered as LDTs that are problematic. In particular, we intend to use available tools to identify and act against currently marketed IVDs offered as LDTs that specifically raise concerns, such as IVDs that are potentially inaccurate or poorly validated.

##### E. Benefits

Expected benefits of phasing out the general enforcement discretion approach for LDTs consist mainly of avoided harms to patients, including unnecessary costs, monetary and otherwise, that would result from usage of problematic IVDs offered as LDTs. Harms can vary in severity according to the particular problematic aspects of an IVD and their consequences for patient care.

We expect the phaseout to produce both health and pecuniary benefits via averting diagnostic error and its consequences, such as incorrect or unnecessary treatment, treatment delays, and disease progression or transmission. Pecuniary benefits also include reduced spending on problematic IVDs offered as LDTs, including non-invasive prenatal screening (NIPS) tests, as well as reduced spending on litigation over alleged harms caused by problematic IVDs.

###### 1. Reduction in Harms from Diagnostic Errors

We expect public health benefits from phasing out the general enforcement discretion approach for LDTs due to improved safety and effectiveness of IVDs offered as LDTs. To...
estimate the baseline incidence of harms that can be avoided by the phaseout, we focus mainly on three broad disease categories identified by Newman-Toker et al. (2021) as accounting for about 75% of serious misdiagnosis-related harms in the U.S.: cancers, cardiovascular disease, and infections (Ref. [33]).

a. Cancer: Mortality Risk

We quantify health benefits in the form of reduced baseline mortality risk based on expected reduction of cancer related misdiagnosis with problematic IVDs offered as LDTs. Based on the data available to us, this analysis focuses specifically on benefits in the form of reduced mortality risk (i.e., benefits associated with reducing false negative diagnoses). However, we anticipate that the phaseout policy will lead to other benefits as well, such as reduced risk of undergoing unnecessary, potentially harmful treatments based on false positive diagnoses.

We also present these estimates with the caveat that the incidence of misdiagnosis-related mortality depends on the manner of attribution of harm to diagnostic delays, and therefore our estimates might imply a number of cases bearing mortality risk consequences that differs from certain available estimates of the number of deaths attributable to misdiagnosis (Ref. [34]).[^27]

With a correct diagnosis, death can be delayed to a later date than one following an incorrect diagnosis. However, depending on when a misdiagnosis occurs, death might still be delayed to a degree depending on how soon a patient seeks follow-up and receives a correct diagnosis at a later time. Life expectancy in this case would still be shortened compared to if the initial diagnosis had been correct, but this would not necessarily be counted as a death due to misdiagnosis.[^28] It is also possible that the differences in risk of death from a delayed diagnosis could be attributable to treatment differences such as fewer effective therapies for later-stage lung cancers contributing to the adverse impact of diagnostic delays (Ref. [33]).

Although we do not estimate the benefits from avoiding false positives, accurate testing for patients can help maximize the benefits of certain therapies that patients need to treat or manage their condition. False test results may result in some treatments being denied to eligible patients, which may worsen their health outcomes.

Expected Reduction in Cancer Misdiagnosis

To estimate the reduction in cancer mortality risk from the phaseout policy, we start from an estimate of annual deaths attributable to diagnostic error and apply four probabilities: the probability of a patient having been tested with an IVD; the probability that the IVD had been offered as an LDT; the probability that the IVD offered as an LDT was problematic; and, finally, the probability that the problematic IVD offered as an LDT resulted in preventable diagnostic error. With respect to this last probability, we note that an IVD that yields a false result in an individual case is not necessarily a problematic IVD (indeed, no test is perfect 100% of the time).

Newman-Toker et al. 2019 (Ref. [35]) state that about 5-10% of the 2.7 million deaths annually in the United States are attributable to diagnostic error—or between 0.135 million and 0.27 million fatalities across all misdiagnosed conditions. Based on the annual US incidence of serious misdiagnosis-related harms across vascular events, infections, and cancers per Newman-Toker et al. 2023 (Ref. [34]), we estimate that about 11% (= 1.5M / 13.7M) of misdiagnosis-related fatalities are associated with cancer.[^29] This results in a range of 0.015 million, 0.022 million and 0.03 million misdiagnosis-related deaths from cancer.

Excluding the state of New York proportionally by population,[^30] we assume that about 20,900 U.S. misdiagnosis-related cancer deaths occur outside of NY and could thus potentially involve tests not approved by NYS CLEP. It is likely that some patients outside of the state of New York are also tested with LDTs approved by NYS CLEP and hence would not necessarily benefit from the phaseout of enforcement discretion for premarket review requirements but will still benefit from the phaseout of enforcement discretion for other requirements, as would patients inside the state of New York. According to Rohr et al. (2016), 91% of U.S. oncology patients undergo IVD testing (Ref. [36]). Using test orders from a U.S. academic hospital system, Rychert et al. (2023) estimate that IVDs offered as LDTs are 3.9% of test order volume and 45% of distinct assays (Ref. [2]). In estimating the percent of patients tested with IVDs who are tested with IVDs offered as LDTs, we thus consider a range from 3.9% to 45%, with a primary estimate of 10%. Using the primary rate, we estimate that about 13,700 U.S. (non-NY) cancer patients would rely on IVDs offered as LDTs (= 150,554 x 0.91 x 0.10).

To estimate the number of these patients tested with IVDs offered as LDTs that would not be authorized by FDA following a premarket submission to the Agency (i.e., following the phaseout of the general enforcement discussion approach), we consult FDA’s 2020 assessment of EUA requests from laboratories for molecular diagnostic COVID tests and statistics from NYS CLEP on application outcomes and assume that similar rates of initial denial would apply under FDA oversight.

A public comment from NYSDOH informed us that, since September 30, 2021, among applications subject to technical review:

- 46% were approved based on the original application;
- another 33% were approved in a second round of review after the applicant provided additional information; and
- 20% could not be approved after the second round of review (though they might have been approved later) (Ref. [12]).

Regarding applications not initially approved, NYSDOH stated in its comment:

- “Tests that are not approved based on the original application have a range of issues. Analytical validity issues include design flaws, inadequate validation data, and process problems that call into question the reliability of the results. For example, the test may not be capable of detecting the target analytes. One application claimed to detect cytomegalovirus (CMV) in a transplant recipient population, but a primer/probe design flaw resulted in the detection of only two CMV subtypes. This error would have endangered patient safety and was only identified during NYS CLEP review. The laboratory redesigned the assay with input from NYS CLEP subject matter experts so that all four subtypes could be detected.” (Ref. [12])

We thus consider a range of scenarios reflecting different possible rates of problematic IVDs offered as LDTs. As a low estimate, we consider that 22% of EUA requests in FDA’s 2020 assessment of EUA requests from laboratories for molecular diagnostic COVID tests had significant design issues and indications for use issues. As a high estimate, we use the 54% of submissions that are not initially approved by New York State. We consider initial approval rates because information contained in New York State’s public comment suggests that “additional information” included in response to original review may include design changes, and thus tests not initially approved might, without review, have gone on to yield unreliable results. For our primary estimate, we average the above two, resulting in 38%.

In the central scenario with an expected initial rejection rate of 38%, about 720 misdiagnosis-related cancer deaths involve patients tested using IVDs offered as LDTs that will not be authorized by FDA following a premarket submission—at least without changes. We assume that some of the time when these IVDs are used, they yield inaccurate results that would not occur using an IVD that could be authorized by FDA. Of these instances of inaccurate results, some might be caught during follow-up or other parts of the process of diagnosis before leading to harm from diagnostic error.

Lacking systematic data on the exact issues with applications for IVDs offered as LDTs initially rejected by NYS CLEP and these LDTs’ roles in the process of diagnosis, we consider a range of rates at which avoidable error might result from usage of problematic IVDs offered as LDTs: from a low of 25% to a high of 75%, with a central estimate of 50%. However, some of these errors might not have consequences for patient care if a patient would in any case be unable or unwilling to obtain treatment. A patient diagnosed with cancer may go untreated for various reasons—including, in some cases, because no effective treatment exists. Ward et al. (2013) analyze data on nontreatment of cancer from the National Cancer Data Base and the Iowa Cancer Registry, which show that between roughly 8 and 12 percent of newly diagnosed cancer patients in Iowa did not receive a first course of treatment (Ref. [37]). Assuming that Iowa cancer patients are representative of the rest of the U.S., we thus estimate that diagnostic error has treatment implications for 88% to 92% of patients, with a central estimate of 90%. Thus, using our central estimates, we expect the phaseout policy to avoid about 325 harms from diagnostic error among cancer patients.

**Table 5. Avoidable Harms Related to Diagnostic Error Among Cancer Patients**
|  | Primary | Low | High |
|---|---|---|---|
| a) Deaths from Misdiagnosis (Non-NYS) | 20,863 | 13,908 | 27,817 |
| b) Percent Tested with IVDs | 91% | 91% | 91% |
| c) Probability of IVD Being Offered as an LDT | 10.0% | 3.9% | 45.0% |
| d) Patients Tested with IVDs Offered as LDTs (= a * b * c) | 1,898 | 494 | 11,391 |
| e) Percent of IVDs offered as LDTs Not Authorized by FDA Following a Premarket Submission | 38% | 22% | 54% |
| f) Tests Using Problematic IVDs Offered as LDTs (= d * e) | 721 | 109 | 6,151 |
| g) Percent Leading to Diagnostic Error | 50% | 25% | 75% |
| h) Treatment-to-Diagnosis Ratio | 0.9 | 0.88 | 0.92 |
| i) Harms Avoidable by the Phaseout Policy (= f * g * h) | 325 | 24 | 4,244 |
||

Value of Reduced Mortality Risk

As a first step in valuing reduced mortality risk from the phaseout policy, we estimate the gain in life expectancy associated with a correct diagnosis for someone who has cancer.

First, we consult 2023 data on estimated new cancer cases along with the five-year relative survival rate covering 2012-2018 (Ref. [38]). The five-year relative survival rate (RSR) in column B of Table 6, represents the percentage of individuals surviving their cancer diagnosis 5 years after diagnosis compared to individuals who are cancer free.
estimate the absolute survival rate (of individuals with cancer who are diagnosed) further down below.

At the bottom of column D, we obtain the average five-year RSR across cancer sites, weighting by percent of total new cancer cases. For example, the weight on the RSR for breast cancer is the number of breast cancers divided by the sum of all new cancer cases (290,560 / 1,818,030 = 16%). The estimated five-year weighted average RSR for all new cancer cases is the sum of column D, 68.6%.

**Table 6. Calculating the Weighted Average Relative Survival Rate (RSR) for New Cancer Cases**
| Site                              | Estimated New Cases (2022) A | Relative Survival (%) (2012–2018) B | % New Cases C | RSR x Percent weight D (=B*C/100) |
|-----------------------------------|------------------------------|------------------------------------|---------------|----------------------------------|
| Breast                            | 290,560                      | 90.5                               | 16%           | 14.46                            |
| Prostate                          | 268,490                      | 96.8                               | 15%           | 14.30                            |
| Lung and Bronchus                 | 236,740                      | 22.9                               | 13%           | 2.98                             |
| Colon and Rectum                  | 151,030                      | 65.1                               | 8%            | 5.41                             |
| Melanoma of the Skin              | 99,780                       | 93.7                               | 5%            | 5.14                             |
| Bladder                           | 81,180                       | 77.1                               | 4%            | 3.44                             |
| Non-Hodgkin Lymphoma              | 80,470                       | 73.8                               | 4%            | 3.27                             |
| Kidney and Renal Pelvis           | 79,000                       | 76.5                               | 4%            | 3.32                             |
| Uterus                            | 65,950                       | 81.3                               | 4%            | 2.95                             |
| Pancreas                          | 62,210                       | 11.5                               | 3%            | 0.39                             |
| Leukemia                          | 60,650                       | 65.7                               | 3%            | 2.19                             |
| Oral Cavity and Pharynx           | 54,000                       | 68                                 | 3%            | 2.02                             |
| Thyroid                           | 43,800                       | 98.4                               | 2%            | 2.37                             |
| Liver and Intrahepatic Bile Duct  | 41,260                       | 20.8                               | 2%            | 0.47                             |
| Myeloma                           | 34,470                       | 57.9                               | 2%            | 1.10                             |
| Other                             | 168,440                      | 60.35                              | 9%            | 4.78                             |
| **Sum**                           | **1,818,030**                |                                    | **100%**      | **68.60%**                       |
||

Note: Product across table may not be exact due to rounding.

Thus, on average, a person with cancer who is diagnosed has 68.6% of the chance of living another five years than a person who is cancer free. According to the National Cancer Institute, the median age of a cancer diagnosis is 66 years (Ref. [39]). Per CDC life tables, the 5-year survival rate for all age-66 individuals is approximately 91.13%.^33 To estimate the absolute 5-year survival of persons with cancer who receive a correct diagnosis from diagnostic testing, we multiply the RSR of 68.60% by 91.13%, thereby obtaining 62.52%. This estimate is likely lower than the true 5-year survival of persons with cancer who receive a correct diagnosis from diagnostic testing for the following reasons: 1) 91.13% does not in fact represent the 5-year survival rate of cancer-free individuals aged 66, but instead the 5-year survival rate of all age-66 individuals, including those with cancer, and 2) the SEER data attempts to represent all cases, which would thus include some that are missed upon initial diagnostic testing and only detected later. Based on the above, and the fact that the 5-year survival rate of 62.52% is more than half, or 50%, the median remaining life expectancy of someone with cancer who is correctly diagnosed by diagnostic testing is at least 5 years.

Next, based on survival of untreated individuals, we estimate the median remaining life expectancy of someone with cancer who is not diagnosed as such. The median survival time for untreated individuals is 2.3 years in cases of breast cancer (Ref. [40]) and 11.94 months, or 0.995 years, in cases of lung cancer (Ref. [41]). We average these two survival times, weighting by the numbers of new cases of breast and lung cancer, respectively, from Table 6 above, and thus obtain a survival time for untreated cancer patients of about 1.71 years. While we acknowledge that uncertainty is introduced by assuming that lung and breast cancers are representative of cancer in general, we received no comment indicating that this assumption is unsuitable.

We therefore estimate the gain in life expectancy from appropriate treatment upon diagnostic testing to be about 3.29 years (= 5 – 1.71). Thus, for an age-66 person with cancer who has just been tested, treating the cancer is worth about 3.29 more years of life starting about 1.71 years from the time of testing. Table 7 shows these life years discounted to the time of the diagnostic test at rates of three and seven percent.

**Table 7. Life Years Due to Treatment of Cancer**
| Time from treatment (years) | Treatment | Discounted to time of treatment (3%) | Discounted to time of treatment (7%) |
|-----------------------------|-----------|-------------------------------------|-------------------------------------|
| 1.714                       | 1         | 0.951                               | 0.890                               |
| 2.714                       | 1         | 0.923                               | 0.832                               |
| 3.714                       | 1         | 0.896                               | 0.778                               |
| 4.714                       | 0.286     | 0.249                               | 0.208                               |
| **Total**                   | **3.286** | **3.018**                           | **2.708**                           |
||

We note that untreated and undiagnosed cancers may not have the same average prognosis. A patient diagnosed with cancer may go untreated for various reasons—including, in some cases, because no effective treatment exists.

Finally, we value these mortality risk reductions (at the time of the diagnostic test) using estimates of the value per statistical life year (VSLY), which is the rate at which a consumer or patient substitutes money for reductions in mortality risk, measured by the willingness to pay for an increase in life expectancy by one year. We use VSLYs derived from the value of a statistical life (VSL) under assumptions of three and seven percent discounting, paired with estimates of the statistical life years gained per case.^34,35 VSLYs are those projected for 2024 but using 2022 base year dollars for consistency with the rest of this analysis. Table 8 below represents our estimates of the value (at the time of the diagnostic test), of the additional expected life years from an accurate diagnosis.

**Table 8. Estimated Value Per Case of Accurate Diagnosis (2022$)**
|       | (a) VSLY (3% discounting) | (b) Value Per Case (VSLY 3%) (= a * 3.018) | (c) VSLY (7% discounting) | (d) Value Per Case (VSLY 7%) (c = c * 2.708) |
|-------|---------------------------|--------------------------------------------|---------------------------|--------------------------------------------|
| Primary | $546,735                  | $1,650,181                                 | $915,435                  | $2,479,329                                 |
| Low   | $255,143                  | $770,084                                   | $427,203                  | $1,157,020                                 |
| High  | $832,253                  | $2,511,942                                 | $1,393,495                | $3,774,090                                 |
||

Note: Product across table may not be exact due to rounding.

To estimate total benefit in Table 9 below, we multiply the estimated reduction in harms by the benefit per avoided harm and by the portion of relevant risk not already internalized in decision-making by medical providers and patients. Because providers who frequently order tests might note quality trends across different labs, we assume that only 95% of the risk of problematic IVDs offered as LDTs is not already internalized at baseline, with a range from 90-100%. Total internalization is unlikely, because without deliberate study of records aided by statistical tools, internalization of the risks of different tests would depend on provider recall and coincident identification of an association from the noise of a provider’s experiences.

**Table 9. Widest Range of Recurring (Annual) Benefit from Reduced Mortality from Cancer-Related Diagnostic Error**
|                                     | Primary   | Low      | High     |
|-------------------------------------|-----------|----------|----------|
| a) Harms Avoidable by the Phaseout Policy | 325       | 24       | 4,244    |
| b) Value Per Harm (VSLY using 3% discounting) | $1,650,181 | $770,084  | $2,511,942 |
| c) Value Per Harm (VSLY using 7% discounting) | $2,479,329 | $1,157,020 | $3,774,090 |
| d) Percent Not Internalized at Baseline | 95%       | 90%      | 100%     |
| e) Total Benefit (VSLY using 3% discounting) | $508,931,983 | $16,558,006 | $10,661,349,657 |
| f) Total Benefit (VSLY using 7% discounting) | $764,649,367 | $24,877,722 | $16,018,239,255 |
||

However, due to our high degree of uncertainty about several of the parameters used to estimate the reduction in mortality risk from misdiagnosis related to cancer, we use a Monte Carlo simulation to determine a plausible range for total benefits by allowing each parameter (the rows in Table 5, the values per case in Table 8, and the internalization percentage in Table 9) to vary independently of the others. Whereas Table 9 implicitly assumes that all uncertain parameters share a joint probability distribution and are perfectly dependent (i.e., aligning all best- and worst-case scenarios across parameters), Table 10 below assumes certain parameters to be independent random variables as follows:

- the yearly number of premature deaths from misdiagnosis related to cancer follows a PERT distribution with a minimum, mean, and maximum taken from row (a) of Table 5;
- the probability that an IVD is offered as an LDT follows a PERT distribution with a minimum, mode, and maximum taken from row (c) of Table 5;
- the percent of such tests that would not be authorized by FDA follows a uniform distribution defined by the low and high estimates of row (e) of Table 5;
- the percent of such tests leading to a preventable misdiagnosis follows a uniform distribution defined by the low and high estimates of row (g) of Table 5;
- the treatment-to-diagnosis ratio follows a uniform distribution defined by the low and high of row (h) of Table 5;
- the values per harm using VSLYs that assume three and seven percent discounting follow triangular distributions with minimums, means, and maximums taken from rows (b) and (c), respectively, of Table 9; and
- the percent of risk of problematic IVDs not already internalized at baseline follows a uniform distribution defined by the low and high estimates from row (d) of Table 9.

Per HHS guidance (Ref. [42]), the low, primary, and high benefit estimates in Table 10 represent the 5th, 50th, and 95th percentiles from running the above simulation 100,000 times.

**Table 10. Simulated Plausible Range of Recurring (Annual) Benefit from Reduced Mortality from Cancer-Related Diagnostic Error**
|                                     | Primary     | Low         | High       |
|-------------------------------------|-------------|-------------|------------|
| e) Total Benefit (VSLY using 3% discounting) | $616,052,630 | $191,389,437 | $1,800,143,089 |
| f) Total Benefit (VSLY using 7% discounting) | $926,834,309 | $286,994,300 | $2,704,265,559 |
||

Given the uncertainty in this analysis and the implausibility of all best- and worst-case scenarios perfectly aligning across the uncertain parameters, we use the results from Table 10 to inform our total benefits estimates in II.E.4 “Summary of Benefits” and in Table 1, the main summary of benefits, costs, and transfers. As we explain in II.E.4 “Summary of Benefits” below, estimates in Table 1 are further adjusted to account for the enforcement discretion policy and the timing of the phase-in.

b. Cardiovascular Disease

Cardiovascular disease is prevalent in the U.S. According to the American Heart Association, in 2020, 48.6% of U.S. adults aged 20 and older had some form of cardiovascular disease, including coronary heart disease, heart failure, stroke, and hypertension (Ref. [43]). Additionally, the CDC notes that “heart disease is the leading cause of death for men, women, and people of most racial and ethnic groups in the United States” (Ref. [44]).

Table 11 below shows our estimated range for the number of harms from diagnostic error among patients with cardiovascular disease that we expect phasing out the general enforcement discretion approach for LDTs to avoid.

Using an estimate by Raisi-Estabragh et al. (2022) of about 20.6 million cardiovascular emergency department encounters in adults in the U.S. between 2016-2018 (Ref. [45]), we estimate about 6.9 million (= 20.6 million / 3) annual cardiovascular emergency department visits. It is likely that not all of these healthcare encounters involve initial diagnoses. However, IVDs are still used in visits concerning known conditions for monitoring of disease, prognosis, predicting treatment response, assessing the risk of developing a disease or disorder, and guiding patient management (Ref. [36]).

Excluding the state of New York proportionally by population, we assume that about 6.5 million such cases are outside of NY and could thus potentially be managed using tests not approved by NYS CLEP. It is likely that some patients outside of the state of New York are also tested with LDTs approved by NYS CLEP and hence will not necessarily benefit from FDA’s phaseout of enforcement discretion for premarket review requirements, but will still benefit from the phaseout of enforcement discretion for other requirements. According to Rohr et al. (2016), 62% of U.S. cardiology patients undergo IVD testing (Ref. [36]). Using test orders from a U.S. academic hospital system, Rychert et al. (2023) estimate that IVDs offered as LDTs are 3.9% of test order volume and 45% of distinct assays (Ref. [2]). In estimating the percent of patients tested with IVDs who are tested with IVDs offered as LDTs, we thus consider a range from 3.9% to 45%, with a primary estimate of 10%. Using the primary estimate, we estimate that about 0.4 million U.S. (non-NY) patients with cardiovascular disease would rely on IVDs offered as LDTs (= 6,461,000 x 0.62 x 0.10).

As described in the previous section on harms related to diagnostic error among cancer patients, we consider a range of estimates of the number of these patients tested using IVDs offered as LDTs that will not be initially authorized by FDA following a premarket submission. As explained above, based on statistics from NYS CLEP on approved and denied applications, we consider initial rejection rates of 22%, 38%, and 54%.

In the central scenario with an expected rejection rate of 38% percent, about 152,000 patients with cardiovascular disease are managed using IVDs offered as LDTs that will not be authorized by FDA following a premarket submission—at least without changes. We assume that some of the time when these tests are used, they yield inaccurate results that would not occur using a test that could be authorized by FDA. Of these instances of inaccurate results, some might be caught during follow-up or other parts of the process of diagnosis before leading to harm from diagnostic error.

Lacking systematic data on the quality of applications for IVDs offered as LDTs rejected by NYS CLEP and these tests’ roles in the process of diagnosis, we consider a range of rates at which avoidable errors might result from usage of problematic IVDs offered as LDTs: from a low of 25% to a high of 75%, with a central estimate of 50%. However, some of these errors might not have consequences for patient care if a patient would in any case be unable or unwilling to obtain treatment. Cardiovascular disease includes several different conditions, each of which might go untreated for various reasons. The American Heart Association reports treatment rates of high cholesterol (44.9%) and hypertension (52%), as well as the rate of diabetes patients with established atherosclerotic cardiovascular disease who are treated with a statin (58.6%) (Ref. [43]). Based on these, we estimate that diagnostic error has treatment implications for 44.9% to 58.6% of patients, with a central estimate of 52%. Thus, using our central estimates, we expect the phaseout policy to avoid about 39,600 harms among patients with cardiovascular disease.

**Table 11. Avoidable Harms Related to Diagnostic Error Among Cardiovascular Disease Patients**
|  | Primary | Low | High |
|---|---|---|---|
| a) Annual US Cardiovascular Emergency Department Encounters (Non-NYS) | 6,461,262 | 6,461,262 | 6,461,262 |
| b) Percent Tested with IVDs | 62% | 62% | 62% |
| c) Probability of IVD Being Offered as an LDT | 10.0% | 3.9% | 45.0% |
| d) Patients Tested with IVDs Offered as LDTs (= a * b * c) | 400,598 | 156,233 | 1,802,692 |
| e) Percent of IVDs Offered as LDTs Not Authorized by FDA Following a Premarket Submission | 38% | 22% | 54% |
| f) Tests Using Problematic IVDs Offered as LDTs | 152,227 | 34,371 | 973,454 |
| g) Percent Leading to Diagnostic Error | 50% | 25% | 75% |
| h) Treatment-to-Diagnosis Ratio | 0.52 | 0.449 | 0.586 |
| i) Harms Avoidable by the Phaseout Policy (= f * g * h) | 39,579 | 3,858 | 427,833 |
||

Harms from diagnostic error are diverse and can vary widely in severity, from avoidable inconvenience and expense to unnecessary treatments, disability, and premature mortality. As one example, between 2008 and early 2011, one laboratory sold over 160,000 StatinCheck tests designed to determine an individual’s KIF6 genotype. This test was marketed as a way to determine a patient’s response to statin drugs, based on the idea that patients with the Trp719Arg polymorphism of the KIF6 protein would have a greater reduction in cardiovascular disease (CVD) events when on statin therapy than patients without this polymorphism. However, research showed no association between the polymorphism and statin response (Refs. [13] [14]).
Additionally, in April 2011, FDA denied premarket approval of this test, citing lack of sufficient evidence of the safety and effectiveness of the test based in particular on clinical validity concerns.

Approximately 35% of patients in studies on CVD have the Trp719Arg polymorphism (Refs. [13] [14]). If 35% of the StatinCheck test recipients were identified as having the Trp719Arg polymorphism, then 56,000 patients may have been informed that they would respond better to statin therapy than other patients. If these patients received lower-potency statin treatment than is standard, a loss of health likely occurred, though medical expenditures were likely reduced. According to Conly et al. (2011), the use of high-potency statins results in an increase of 0.13 QALYs relative to the use of low-potency statins (Ref. [46]). Since this consists almost entirely of mortality effects, we value the health gains from high-potency statins in terms of life years. Conly et al. report life expectancy averages of 21.0 years for patients taking low-potency statins and 21.4 years for patients taking high-potency statins. Discounting at three and seven percent, this represents a gain of about 0.22 and 0.10 discounted life years, respectively, at the time of initiation of statin use. However, the use of high-potency statins in Canada costs CAD $1,200 more than low-potency statins (Ref. [46]). Converting to USD at the current rate of CAD $1.36 to USD $1.00, this is about $882. Based on a report prepared for the Department of Health and Human Services comparing international prescription drug prices, prices across all prescription drugs in Canada are about 44% of U.S. prices (Ref. [47]). We thus divide again by 0.44, resulting in a cost difference between high and low potency statins of about $2,005. Using a Value of a Statistical Life Year (VSLY) of $546,735 (the central VSLY that assumes three percent discounting), the value of lost health from using low-potency statins instead of high-potency statins is $121,086 (= $546,735 x 0.2215). The net lost benefit for each person using low-potency statins is $119,080 (= $121,086 - $2,005), and the estimated total welfare losses are thus about $6.7 billion (= $119,080 x 56,000). Using a VSLY of $915,435 (the central VSLY that assumes seven percent discounting), the value of lost health from using low-potency statins instead of high-potency statins is $94,626 (= $915,435 x 0.1034). VSLYs are those projected for 2024 but using 2022 base year dollar values for consistency with the rest of this analysis. The net lost benefit for each person using low-potency statins is $92,621 (= $94,626 - $2,005), and the estimated total welfare losses over the period 2008-2011 are thus about $5.2 billion (= $92,621 x 56,000).

Thus, as a proxy for the value of an average harm from diagnostic error to a patient with a cardiovascular disease, we use $119,080 as our primary estimate given three percent discounting and $92,621 as our primary estimate given seven percent discounting. Further below, we also include estimates based on the low and high VSLY estimates. 

To estimate total benefit in Table 12 below, we multiply the estimated reduction in harms by the benefit per avoided harm and by the portion of relevant risk not already internalized in decision-making by medical providers and patients. Because providers who frequently order tests might note quality trends across different labs, we assume that only 95% of the risk of problematic IVDs offered as LDTs is not already internalized at baseline, with a range from 90-100%. Total internalization is unlikely, because without deliberate study of records aided by statistical tools, internalization of the risks of different tests would depend on provider recall and coincident identification of an association from the noise of a provider’s experiences.

**Table 12. Widest Range of Recurring (Annual) Benefit from Avoiding Harms from Diagnostic Error Related to Cardiovascular Disease**
|                             | Primary    | Low     | High        |
|-----------------------------|------------|---------|-------------|
| a) Harms Avoidable by the Phaseout Policy | 39,579    | 3,858   | 427,833    |
| b) Value Per Harm (VQALY using 3% discounting) | $119,080  | $54,501 | $182,314   |
| c) Value Per Harm (VQALY using 7% discounting) | $92,621   | $42,154 | $142,037   |
| d) Percent Not Internalized at Baseline  | 95%        | 90%     | 100%       |
| e) Total Benefit (VQALY using 3% discounting) | $4,477,436,064 | $189,248,233 | $77,999,873,099 |
| f) Total Benefit (VQALY using 7% discounting) | $3,482,562,142 | $146,372,612 | $60,768,070,609 |
||

Due to our high degree of uncertainty about several of the parameters used to estimate the reduction in morbidity risk from diagnostic error related to cardiovascular disease, we use a Monte Carlo simulation to determine a plausible range for total benefits by allowing each parameter (the rows in Table 11 and the values per case and internalization percentage in Table 12) to vary independently of the others. Whereas Table 12 implicitly assumes that all uncertain parameters share a joint probability distribution and are perfectly dependent (i.e., aligning all best- and worst-case scenarios across parameters), Table 13 below assumes certain parameters to be independent random variables as follows:

- the percent of patients tested with IVDs who are tested with IVDs offered as LDTs follows a PERT distribution with a minimum, mode, and maximum taken from row (c) of Table 11;
- the percent of such tests that would not be authorized by FDA follows a uniform distribution defined by the low and high estimates of row (e) of Table 11;
- the percent of such tests leading to a preventable misdiagnosis follows a uniform distribution defined by the low and high estimates of row (g) of Table 11;
- the treatment-to-diagnosis ratio follows a PERT distribution with a minimum, mode, and maximum taken from row (h) of Table 11; 
- the values per harm using VSLYs that assume three and seven percent discounting follow triangular distributions with minimums, means, and maximums taken from the low, primary, and high estimates in rows (b) and (c), respectively, of Table 12; and
- the percent of risk of problematic IVDs not already internalized at baseline follows a uniform distribution defined by the low and high estimates from row (d) of Table 12.

Per HHS guidance (Ref. [42]), the low, primary, and high benefit estimates in Table 13 represent the 5th, 50th, and 95th percentiles from running the above simulation 100,000 times.

**Table 13. Simulated Plausible Range of Recurring (Annual) Benefit from Avoiding Harms from Diagnostic Error Related to Cardiovascular Disease**
|                                       | Primary       | Low           | High           |
|---------------------------------------|---------------|---------------|----------------|
| e) Total Benefit (VQALY using 3% discounting) | $5,430,362,350 | $1,716,966,298 | $15,579,610,547 |
| f) Total Benefit (VQALY using 7% discounting) | $4,233,288,851 | $1,327,584,767 | $12,131,010,945 |
||

Given the uncertainty in this analysis and the implausibility of all best- and worst-case scenarios perfectly aligning across the uncertain parameters, we use the results from Table 13 to inform our total benefits estimates in II.E.4 “Summary of Benefits” and in Table 1, the main summary of benefits, costs, and transfers. As we explain in II.E.4 “Summary of Benefits” below, estimates in Table 1 are further adjusted to account for the enforcement discretion policy and the timing of the phase-in.

c. Morbidity Due to Infections

Table 14 below shows our estimated range for the number of harms from diagnostic errors related to infections that we expect the phaseout policy to avoid. There are many kinds of infectious diseases. Because we cannot comprehensively analyze expected consequences of the phaseout policy with respect to all possible infectious diseases, we base our estimate of the number of relevant cases on the CDC’s statistics on selected national notifiable infectious diseases (Ref. [48]). For the year 2019, CDC reports 2,738,992 cases of those selected infectious diseases, over 93% of which are sexually transmitted infections. 

We believe this to be about half or more of all serious infectious diseases in the U.S., but we use this figure for lack of a comprehensive accounting of all other possible infectious diseases. Excluding the state of New York proportionally by population, we assume that about 2,577,300 such cases occur outside of NY and could thus potentially involve tests not approved by NYS CLEP. It is likely that some patients outside of the state of New York are also tested with LDTs approved by NYS CLEP and hence would not necessarily benefit from the phaseout of enforcement discretion for premarket review requirements, but will still benefit from the phaseout of enforcement discretion for other requirements. 

According to Rohr et al. (2016), 64% of U.S. oncology and cardiology clinical decisions involve IVD testing (Ref. [36]), and due to lack of more directly relevant data, the oncology and cardiology estimate is extrapolated to the infectious disease context. Using test orders from a U.S. academic hospital system, Rychert et al. (2023) estimate that IVDs offered as LDTs are 3.9% of test order volume and 45% of distinct assays (Ref. [2]). In estimating the percent of patients tested with IVDs who are tested with IVDs offered as LDTs, we thus consider a range from 3.9% to 45%, with a primary estimate of 10%. Using the primary rate, we estimate that about 164,900 U.S. (non-NY) patients with infectious disease would rely on IVDs offered as LDTs (= 2,577,283 x 0.64 x 0.10).

As described in the earlier section on harms from diagnostic error related to cancers, we consider a range of estimates of the number of these patients tested using IVDs offered as LDTs that will not be initially authorized by FDA following a premarket submission. As explained above in the section on cancers, based on statistics from NYS CLEP on approved and denied applications, we consider rejection rates of 22%, 38%, and 54%.

In the central scenario with an expected rejection rate of 38% percent, about 62,700 patients with infections are tested using IVDs offered as LDTs that will not be authorized by FDA following a premarket submission—at least without changes. We assume that some of the time when these tests are used, they yield inaccurate results that would not occur using a test that could be authorized by FDA. Of these instances of inaccurate results, some might be caught during follow-up or other parts of the process of diagnosis before leading to harm.

Lacking systematic data on the exact issues with applications for IVDs offered as LDTs initially rejected by NYS CLEP and these tests’ roles in the process of diagnosis, we consider a range of rates at which avoidable error might result from usage of problematic IVDs offered as LDTs: from a low of 25% to a high of 75%, with a central estimate of 50%. However, some of these errors might not have consequences for patient care if a patient would in any case be unable or unwilling to obtain treatment. There are many kinds of infections, each of which might go untreated for various reasons. 

Based on treatment rates from Li et al. (2023) for chlamydia and gonorrhea among symptomatic and asymptomatic men and women in the US (Ref. [49]), and assuming equal numbers of men and women patients, overall treatment rates are about 36.42% for gonorrhea and 42.18% for chlamydia. Averaging between these two rates while weighting by their respective shares of notifiable disease cases (about three cases of chlamydia to one case of gonorrhea), we estimate that diagnostic error has treatment implications for 40.7% of infectious disease patients, with a range from 30-50%. Thus, using our central estimates, we expect the phaseout policy to avoid about 12,800 harms among patients with infections.

**Table 14. Avoidable Harms from Diagnostic Error Related to Infections**
|                                      | Primary   | Low     | High     |
|--------------------------------------|-----------|---------|----------|
| a) Yearly U.S. Infections (Non-NYS)  | 2,577,283 | 2,577,283 | 2,577,283 |
| b) Percent Tested with IVDs          | 64%       | 64%     | 64%      |
| c) Probability of IVD Being Offered as an LDT | 10.0%    | 3.9%    | 45.0%   |
| d) Patients Tested with IVDs Offered as LDTs (= a * b * c) | 164,946  | 64,329  | 742,258  |
| e) Percent of IVDs Offered as LDTs Not Authorized by FDA Following a Premarket Submission | 38%      | 22%     | 54%      |
| f) Tests Using Problematic IVDs Offered as LDTs (= d * e) | 62,680   | 14,152  | 400,819  |
| g) Percent Leading to Diagnostic Error | 50%      | 25%     | 75%      |
| h) Treatment-to-Diagnosis Ratio      | 0.41      | 0.30    | 0.50     |
| i) Harms Avoidable by the Phaseout Policy (= f * g * h) | 12,761   | 1,061   | 150,307  |
||

Harms from diagnostic error are diverse and can vary widely in severity, from avoidable inconvenience and expense to unnecessary treatments, disability, and premature mortality. As over 93% of CDC notifiable infectious disease cases are sexually transmitted diseases, we use average QALY loss from chlamydia and gonorrhea as a proxy for the value of an average harm from diagnostic error to a patient with an infectious disease. Based on discounted lifetime QALY loss estimates from Li et al. (2023) for chlamydia and gonorrhea among men and women in the US (Ref. [49]), and assuming equal numbers of men and women patients, average QALY losses are about 0.008 from gonorrhea and 0.024 from chlamydia. However, part of this health loss occurs during the acute infection, prior to diagnosis. Considering only the portions of health loss attributable to sequelae based on Figure 6 (Ref. [49]), these losses are about 0.007 from gonorrhea and 0.022 from chlamydia.39

Averaging between these two diseases while weighting by their respective shares of notifiable disease cases (about three cases of chlamydia to one case of gonorrhea), we estimate that the average case entails a discounted lifetime QALY loss of about 0.0185. Health loss estimates by Li et al. consider lifetime sequelae and complications from lack of treatment such as pelvic inflammatory disease, but both gonorrhea and chlamydia are curable within a few days with antibiotics.40 According to Malek et al. (2013), “delay in seeking care for STDs can [...] increase the likelihood of consequences such as infertility and chronic pelvic pain” (Ref. [50]).

Using a VQALY of $649,215 (the central VQALY that assumes three percent discounting), we thus assume that the value of lost health from an average infectious disease case without timely identification and treatment is about $12,000 (= $649,215 x 0.0185). Using a VQALY of $1,070,162 (the central VQALY that assumes seven percent discounting), the value of lost health is about $19,800. VQALYs are those projected for 2024 but using 2022 base year dollar values for consistency with the rest of this analysis.

Diagnostic error for infectious disease tests may lead to uncontrolled spread of communicable infectious diseases from contact with patients relying on false results from problematic IVDs. Our estimates do not account for harms from these downstream infections.

To estimate total benefit in Table 15 below, we multiply the estimated reduction in harms from diagnostic error by the benefit per avoided harm and by the portion of relevant risk not already internalized in decision-making by medical providers and patients. Because providers who frequently order tests might note quality trends across different labs, we assume that only 95% of the risk of problematic IVDs offered as LDTs is not already internalized at baseline, with a range from 90-100%. Total internalization is unlikely, because without deliberate study of records aided by statistical tools, internalization of the risks of different tests would depend on provider recall and coincident identification of an association from the noise of a provider’s experiences.

**Table 15. Widest Range of Recurring (Annual) Benefit from Avoiding Harms from Diagnostic Error Related to Infections**
|                                    | Primary  | Low     | High       |
|-----------------------------------|----------|---------|------------|
| a) Harms Avoidable by the Phaseout Policy | 12,761   | 1,061   | 150,307    |
| b) Value Per Harm (VQALY using 3% discounting) | $11,995  | $5,598  | $18,259    |
| c) Value Per Harm (VQALY using 7% discounting) | $19,772  | $9,227  | $30,098    |
| d) Percent Not Internalized at Baseline | 95%      | 90%     | 100%       |
| e) Total Benefit (VQALY using 3% discounting) | $145,414,428 | $5,347,319 | $2,744,440,950 |
| f) Total Benefit (VQALY using 7% discounting) | $239,700,470 | $8,814,497 | $4,523,923,748 |
||

Due to our high degree of uncertainty about several of the parameters used to estimate the reduction in morbidity risk from diagnostic error related to infectious diseases, we use a Monte Carlo simulation to determine a plausible range for total benefits by allowing each parameter (the rows in Table 14 and the values per case and internalization percentage in Table 15) to vary independently of the others. Whereas Table 15 implicitly assumes that all uncertain parameters share a joint probability distribution and are perfectly dependent (i.e., aligning all best- and worst-case scenarios across parameters), Table 16 below assumes certain parameters to be independent random variables as follows:
- the percent of patients tested with IVDs who are tested with IVDs offered as LDTs follows a PERT distribution with a minimum, mode, and maximum taken from row (c) of Table 14;
- the percent of such tests that would not be authorized by FDA follows a uniform distribution defined by the low and high estimates of row (e) of Table 14;
- the percent of such tests leading to a preventable misdiagnosis follows a uniform distribution defined by the low and high estimates of row (g) of Table 14;
- the treatment-to-diagnosis ratio follows a PERT distribution with a minimum, mode, and maximum taken from row (h) of Table 14;
- the values per harm using VQALYs that assume three and seven percent discounting follow triangular distributions with minimums, means, and maximums taken from the low, primary, and high estimates in rows (b) and (c), respectively, of Table 15; and
- the percent of risk of problematic IVDs not already internalized at baseline follows a uniform distribution defined by the low and high estimates from row (d) of Table 15.

Per HHS guidance (Ref. [42]), the low, primary, and high benefit estimates in Table 16 represent the 5th, 50th, and 95th percentiles from running the above simulation 100,000 times.

**Table 16. Simulated Plausible Range of Recurring (Annual) Benefit from Avoiding Harms from Diagnostic Error Related to Infections**
|   | Primary        | Low         | High         |
|---|----------------|-------------|--------------|
| e) Total Benefit (VQALY using 3% discounting) | $189,393,046  | $54,670,583 | $577,904,403 |
| f) Total Benefit (VQALY using 7% discounting) | $311,957,235  | $90,634,571 | $953,299,062 |
||

Given the uncertainty in this analysis and the implausibility of all best- and worst-case scenarios perfectly aligning across the uncertain parameters, we use the results from Table 16 to inform our total benefits estimates in II.E.4 “Summary of Benefits” and in Table 1, the main summary of benefits, costs, and transfers. As we explain in II.E.4 “Summary of Benefits” below, estimates in Table 1 are further adjusted to account for the enforcement discretion policy and the timing of the phase-in.

###### 2. Non-Health Benefits
a. Spending on Inappropriate Use of Non-invasive Prenatal Screening (NIPS)

Non-invasive prenatal screening (NIPS) tests can provide information about the possibility of a fetus having certain genetic abnormalities that could result in a child being born with a serious health condition. Negative results can help pregnant individuals avoid the risks to fetal health of undergoing more invasive tests. However, as screening tests, positive results only indicate risk of a condition and require follow-up with diagnostic tests to confirm or rule out the suspected condition—in turn requiring discussion between patients and healthcare providers. NIPS test results should not be used by themselves to make critical healthcare decisions and should be discussed with a healthcare provider.

Given the increased use and marketing of these tests and recent media reports, FDA has warned the public of the risk of false results, inappropriate use, and inappropriate interpretation of NIPS test results, which might be addressed by phasing out the general enforcement discretion approach for LDTs. FDA is particularly concerned about reports of patients and health care providers that have made critical health care decisions based on results from these screening tests alone and without additional confirmatory testing, possibly related to misleading marketing.

Screening tests for extremely rare conditions caused by genetic microdeletions have generated significant revenue: “adding microdeletions can double what an insurer pays — from an average of $695 for the basic tests to $1,349 for the expanded panel, according to the health data company Concert Genetics” (Ref [15]). However, the five most common microdeletion tests screen for conditions that affect only one in 5,000-20,000 births. According to the NY Times, patients or their providers might lack the requisite understanding of NIPS to make informed purchase decisions: “doctors already order many tests during short prenatal care visits, meaning some probably thought little of tacking on a few more.” Additionally, the NY Times reveals evidence that patients or their providers might not understand that positive results from NIPS for rare conditions can be wrong up to 81 to 93 percent of the time (Ref. [15]). Misunderstanding by patients or their providers might result in avoidable distress and premature medical decisions or false reassurance. We consider that increased regulatory oversight might reduce spending on NIPS IVDs offered as LDTs to screen for a particular condition that have potentially unreliable, inaccurate, or misinterpreted results and require confirmatory diagnostic testing. Increased oversight of NIPS tests, including regarding labeling requirements, can help ensure such tests are appropriately labeled with transparent information regarding performance, clear instructions, and appropriate limitations.

A potentially unreliable, inaccurate, or misinterpreted test result imposes, at minimum, the monetary cost of the test to the patient or health care system payor and the burden of any resulting health consequences. We do not attempt to quantify any expected reduction in spending on these tests, but we note that the number of screening tests for microdeletions sold in 2020 was above 400,000, and patients or payors paid approximately an additional $654 for each expanded test.

b. Reduction in Expenses from Lawsuits.
Compliance with applicable legal requirements for IVDs offered as LDTs might also reduce the incidence of litigation related to problematic IVDs apart from that which stems directly from diagnostic error. We cannot quantify the welfare losses due to tort expenses that might be avoided by phasing out the general enforcement discretion approach for LDTs and do not include the avoidance of any such expenses in our estimate of benefits. However, we provide one case study concerning a COVID-19 test offered without emergency use authorization from FDA as an illustrative example.

On March 1st, 2022, Blue Cross and Blue Shield of Minnesota (Blue Cross or BCBSM) filed a lawsuit against COVID-19 testing laboratory GS Labs, LLC (GS Labs) to recover more than $10 million in overpayments made since the start of the pandemic (Ref. [51]). Blue Cross alleged violations of Minnesota consumer protection law, fraud, and ERISA violations. Among other issues, confusion between quality control processes specified by the test system manufacturer and the lab led GS Labs to issue a correspondence to patients about PCR tests that “inadvertently deviated from applicable laboratory standards for testing facilities” (Ref [52]). The case is currently ongoing (Refs. [51] and [53]).

In a 2022 report from the U.S. Chamber of Commerce Institute for Legal Reform (ILR), high costs in the tort system led to higher prices for other things in the economy. Compensation to claimants (when they win a case) only represents 53 percent of the total size of the tort system, while the remaining litigation and risk transfer costs make up about 47 percent of expenses in the system. In other words, for every $1.00 received by claimants, $0.88 was paid in legal and other costs ($1 / $1.88 =53%) (Ref. [54]). We assume that those total litigation and risk transfer costs are recaptured as savings when problems are prevented by compliance with applicable requirements instead of corrected via litigation. In a $10 million dollar case, for example, litigation and risk transfer costs would be 47 percent of $10 million dollars or $4.7 million.

###### 3. Summary of Benefits

Quantified health benefits include the avoidance of harms from diagnostic errors related to mortality (cancer and cardiovascular disease) and infections-related morbidity. Unquantified benefits include, among others, costs savings from avoiding payment for problematic IVDs, namely NIPS tests, and possible reduction in costs from lawsuits and reduction in costs to healthcare systems. We also note that we do not count individuals who may have contracted communicable infectious diseases from contact with patients relying on false results from problematic IVDs. We are not able to quantify the extent to which the phaseout policy might prevent the spread of communicable infectious diseases. Additionally, the phaseout policy might remove a disincentive for non-laboratory manufacturers, who do not have the benefit of enforcement discretion, to develop novel tests. These manufacturers may otherwise be discouraged from investing in novel tests due to the prospect of laboratory competitors offering IVDs as LDTs that claim to fulfill, equally effectively, the same needs without having to invest in meeting FDA requirements. This benefit would be distinct from avoiding patient harms by improving reliability in existing testing applications, since novel tests might offer new capabilities. We present total benefits and subtotal health and non-health benefits in Table 17.

**Table 17. Total Undiscounted Benefits (Millions 2022$)**
| Type         | Level       | Primary | Low  | High   |
|--------------|-------------|---------|------|--------|
| **Health Benefits (VSLY 3%)**   |             |         |      |        |
| Cancer       | Generalized | $616   | $191 | $1,800 |
| Cardiovascular | Generalized | $5,430 | $1,717 | $15,580 |
| Infections   | Generalized | $176   | $55  | $502   |
| **Health Benefits (VSLY 7%)**   |             |         |      |        |
| Cancer       | Generalized | $927   | $287 | $2,704 |
| Cardiovascular | Generalized | $4,233 | $1,328 | $12,131 |
| Infections   | Generalized | $291   | $91  | $830   |
| **Total Sum of Benefits (VSLY 3%)** |         | $6,223 | $1,963 | $17,881 |
| **Total Sum of Benefits (VSLY 7%)** |         | $5,451 | $1,705 | $15,666 |
||

We expect benefits to begin to accrue two years after publication of the final phaseout policy, though we do not expect all estimated benefits to take place all at once. Instead, we assume that one-time benefits will occur evenly over Stages 1 to 5 of the final phaseout policy (year 3 to year 5). We also expect recurring benefits to begin to accrue at an incremental rate of 0%, 50%, 75%, and 100% for the first four years (Table 18).

**Table 18. Undiscounted Potential Benefits Over Time (Primary Estimate in Millions 2022$, 20 years, 3% and 7%)**
| Stage | Rate       | Year | If VSLY based on 3% discounting | If VSLY based on 7% discounting |
|-------|------------|------|---------------------------------|---------------------------------|
|       | One-time | Recurring | Total | One-time | Recurring | Total |
| 1     | 0        | 0    | $0     | $0     | $0     | $0     | $0     |
| 2     | 1/3      | 1/2  | 3     | $3,111 | $3,111 | $0     | $2,725 | $2,725 |
| 3 & 4 | 1/3      | 3/4  | 4     | $0     | $3,151 | $3,151 | $0     | $2,760 | $2,760 |
| 4 & 5 | 1/3      | 1    | 5     | $0     | $4,420 | $4,420 | $0     | $3,871 | $3,871 |
| 0     | 1        | 6    | $0     | $4,613 | $4,613 | $0     | $4,041 | $4,041 |
| :     | :        | :    | :     | :     | :     | :     | :     |
| 0     | 1        | 20   | $0     | $5,899 | $5,899 | $0     | $5,167 | $5,167 |
| **Sum** |          |      | $0     | $91,718| $91,718| $0     | $80,339| $80,339|
||

However, as a result of exercising enforcement discretion with respect to premarket review and QS requirements, with the exception of requirements under 21 CFR 820, subpart M (Records), for currently-marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule, IVDs offered as LDTs will generally undergo review and comply with most QS requirements only in certain circumstances. However, FDA expects such IVDs offered as LDTs to be in compliance with other applicable requirements under the FD&C Act and FDA regulations, including post-market requirements, as discussed in the phaseout policy. We note that as described in section V.B.3 of the preamble, FDA intends to take targeted steps to address currently marketed IVDs offered as LDTs that are problematic. In particular, we intend to use available tools to identify and act against currently marketed IVDs offered as LDTs that specifically raise concerns, such as IVDs that are potentially inaccurate or poorly validated.

In order to account for this, we adjust estimated total benefits in each year—beginning with enforcement of QS and premarket review requirements—based on the proportions of IVDs falling within an enforcement discretion policy versus those that are not. IVDs falling within an enforcement discretion policy will not generally be expected to comply with premarket review and QS requirements (but will be expected to comply with all other applicable requirements as discussed in the phaseout policy). We thus assume that among patients using these IVDs, only half of the estimated potential benefits will be realized. The low and high benefits estimates reflect alternative assumptions that only 25% and 75% of benefits, respectively, will be realized among patients using IVDs falling within an enforcement discretion policy. In each year beginning with enforcement of QS and premarket review requirements in stage 3 and 4, we apply the following adjustment factor to estimated benefits:

*adjustment = prop. of IVDs NOT falling within ED policy + 0.5*  
*prop. of IVDs falling within an ED policy*

In Table 19, the resulting adjusted annualized benefit estimates using three and seven percent discounting are approximately $4.3 billion and $3.5 billion, respectively.

**Table 19. Expected Benefits Over Time Accounting for Exercising Enforcement Discretion (Primary Estimate in Millions 2022$, 20 years)**
| Stage | Year | Proportions of IVDs by enforcement | c) Effect of ED | d) Adj. factor (= b + c*a) | Benefits (3% VSLY) | Benefits (7% VSLY) |
|-------|------|-----------------------------------|----------------|--------------------------|-------------------|-------------------|
|       |      | a) With ED | b) Without ED | policy (premarket review and most QS requirements) on benefits | e) Without ED | f) With ED (=d*e) |
| 1     | 1    | 0.918      | 0.082        | 0.5 | 0.541 | $- | $- |
| 2     | 2    | 0.818      | 0.182        | 0.5 | 0.591 | $- | $- |
|       | 3    | 0.729      | 0.271        | 0.5 | 0.636 | $3,111 | $2,725 |
| 3 & 4 | 4    | 0.650      | 0.350        | 0.5 | 0.675 | $4,667 | $3,151 |
| 4 & 5 | 5    | 0.580      | 0.420        | 0.5 | 0.710 | $6,223 | $4,420 |
| 6     | 6    | 0.517      | 0.483        | 0.5 | 0.741 | $6,223 | $4,613 |
| 7     | 7    | 0.462      | 0.538        | 0.5 | 0.769 | $6,223 | $4,786 |
| 8     | 8    | 0.413      | 0.587        | 0.5 | 0.794 | $6,223 | $4,939 |
| 9     | 9    | 0.369      | 0.631        | 0.5 | 0.816 | $6,223 | $5,075 |
| 10    | 10   | 0.330      | 0.670        | 0.5 | 0.835 | $6,223 | $5,197 |
| 11    | 11   | 0.295      | 0.705        | 0.5 | 0.853 | $6,223 | $5,305 |
| 12    | 12   | 0.264      | 0.736        | 0.5 | 0.868 | $6,223 | $5,403 |
| 13    | 13   | 0.236      | 0.764        | 0.5 | 0.882 | $6,223 | $5,490 |
| 14    | 14   | 0.211      | 0.789        | 0.5 | 0.895 | $6,223 | $5,568 |
| 15    | 15   | 0.188      | 0.812        | 0.5 | 0.906 | $6,223 | $5,638 |
| 16    | 16   | 0.168      | 0.832        | 0.5 | 0.916 | $6,223 | $5,701 |
| 17    | 17   | 0.149      | 0.851        | 0.5 | 0.925 | $6,223 | $5,758 |
| 18    | 18   | 0.133      | 0.867        | 0.5 | 0.934 | $6,223 | $5,810 |
| 19    | 19   | 0.118      | 0.882        | 0.5 | 0.941 | $6,223 | $5,856 |
| 20    | 20   | 0.104      | 0.896        | 0.5 | 0.948 | $6,223 | $5,899 |
| Sum   |      |            |              |     | $107,344 | $91,718 | $94,026 | $80,339 |
| Present Value (Discount Rate matches VSLY) | | | | | $64,584 | $37,177 |
| Annualized Value (Discount Rate matches VSLY) | | | | | $4,341 | $3,509 |
||

##### F. Costs

FDA is phasing out the general enforcement discretion approach for LDTs so that IVDs manufactured by a laboratory will generally fall under the same enforcement approach as other IVDs and be expected to meet applicable requirements. This phaseout is intended to help assure the safety and effectiveness of IVDs offered as LDTs, while also accounting for other important public health considerations such as patient access and reliance.

FDA intends that the phaseout of enforcement discretion will occur over a four-year period in five key stages as described in section V.C of the preamble. For a few categories of IVDs manufactured by laboratories, FDA is adopting enforcement discretion policies with respect to some or all applicable requirements as described in sections V.B and C of the preamble.

When calculating the costs associated with each stage of the phaseout policy described in the preamble, we use wage information from the Bureau of Labor Statistics Occupational Employment and Wage Statistics. Specifically, we use wage information for a specific industry: medical and diagnostic laboratories.

The remainder of this section discusses the estimated cost of the phaseout policy by stage of the phaseout policy. Section II.F.6 discusses additional cost considerations that we do not quantify.

###### 1. Costs Under Stage 1

Beginning 1 year after the publication date of this final rule, FDA will expect laboratories to comply with MDR requirements (requirements for adverse event reporting) under 21 U.S.C. 360i(a)-(c) and 21 CFR part 803, correction and removal reporting requirements under 21 U.S.C. 360i(g) and 21 CFR part 806, and QS requirements under § 820.198 (complaint files). During the first year following issuance of the final rule, laboratories will face costs associated with compliance with Stage 1, as well as costs associated with reading and understanding the rule in its entirety.

a. Reading and Understanding the Rule

We expect that laboratories affected by the phaseout policy will incur costs to read and understand the rule. We assume an average of one medical laboratory manager and one attorney at each entity will read the rule. Consistent with guidelines from the Department of Health and Human Services, we assume that the reading speed of reviewers ranges from 200 to 250 words per minute. The final rule has approximately 150,000 words. The overall burden in hours (per reader) to read the rule ranges from 10.00 hours (= (150,000 words / 250 words per minute) / 60 mins per hour) to 12.50 hours (= (150,000 words / 200 words per minute) / 60 mins per hour). The mean hourly wages for managers and lawyers in this industry are $57.60 and $80.30, respectively. Fully loaded wage rates are $115.20 an hour for managers and $160.60 an hour for lawyers (average: $137.90). We assume that one to three employees will read the rule. The estimated learning costs per entity would range from $1,379.00 (=10.00 hours x $137.90 per hour x 1 employee) to $5,171.25 (=12.50 hours x $137.90 per hour x 3 employees), with a primary cost of $3,064.44 (=11.11 hours x $137.90 per hour x 2 employees). Multiplying this estimate by the total numbers of affected laboratories per year yields a total one-time cost for reading the rule between $0.81 million and $12.21 million, with a primary estimate of $3.62 million. The estimated total recurring cost ranges from $0.07 million to $0.98 million, with a primary estimate of $0.29 million (see Table 20).

**Table 20. Costs of Reading and Understanding the Rule**
|                                | Primary | Low     | High    |
|--------------------------------|---------|---------|---------|
| Average reading speed (words/minute) | 225     | 250     | 200     |
| Length of preamble & codified (words) | 150,000 | 150,000 | 150,000 |
| Hours                          | 11.11   | 10.00   | 12.50   |
| Number of employees to read rule | 2       | 1       | 3       |
| Labor cost of hourly employee  | $137.90 | $137.90 | $137.90 |
| Per-laboratory cost            | $3,064.44 | $1,379.00 | $5,171.25 |
| Number of affected laboratories | 1,181   | 590     | 2,362   |
| Number of new laboratories per year | 94      | 47      | 189     |
| **Total One-time Costs (millions)** | **$3.62** | **$0.81** | **$12.21** |
| **Total Recurring Costs (millions)** | **$0.29** | **$0.07** | **$0.98** |
||

*Note: Product across table may not be exact due to rounding.*

b. Medical Device Reporting

Under Stage 1, FDA expects laboratories to comply with MDR requirements under 21 U.S.C. 360i(a)-(c) and 21 CFR part 803. In estimating the costs of compliance for laboratories, we use a similar approach to the *Medical Device Reporting: Electronic Submission Requirements* final regulatory impact analysis (Ref. [55]). We expect that laboratories will face one-time costs associated with establishing a reporting system for laboratories for which, at baseline, the requirement to have such systems generally has not been enforced. We also expect new laboratories to enter the market each year, so we assume that the new entities will incur recurring costs associated with establishing a reporting system.

We expect laboratories to establish standard operating procedures (SOPs) in response to the MDR requirements. We estimate it will take 1 – 3 management employees with an hourly wage of $61.36 ($122.72 fully-loaded) 8 – 12 hours each to establish a laboratory’s SOP and train the appropriate people on the new procedures. Multiplying these estimates, we estimate the one-time costs of modifying SOPs to be between $0.58 million and $10.43 million, with a primary estimate of $2.90 million. We estimate the recurring costs to range from $0.05 million to $0.83 million, with a primary estimate of $0.23 million. See Table 21.

We expect laboratories to install and validate e-Submitter software for the purposes of complying with MDR requirements. We expect this task to take a single computer and information system manager 48 to 56 hours, working at an hourly wage of $79.72 ($159.44 fully loaded). Multiplying by the number of affected entities, we estimate the one-time costs of installing and validating e-Submitter software to be between $4.52 million and $21.09 million, with a primary estimate of $9.79 million. We estimate the recurring costs to be between $0.36 million to $1.69 million, with a primary estimate of $0.78 million.

We expect 0.6% of covered laboratories to establish Health Level Seven (HL7) Individual Case Study Report (ICSR) capability (Ref. [55]). We expect this task to take a single computer and information system manager 48 to 52 hours, working at an hourly wage of $79.72 ($159.44 fully loaded). Multiplying by the small fraction of laboratories that we expect to establish such capabilities, we estimate the one-time costs to range between $0.03 million to $0.12 million, with a primary estimate of $0.06 million. We estimate the recurring costs to be between $0.002 million to $0.01 million, with a primary estimate of $0.005 million.

We expect laboratories to acquire an e-certificate from a third-party system to commence medical device reporting. We estimate that there is a small one-time search cost of acquiring the e-certificate of $20. Multiplied by the number of affected entities, we estimate the one-time costs of acquiring an e-certificate to range from $0.01 million to $0.05 million,
We estimate the recurring costs to range from $0.001 million to $0.004 million, with a primary estimate of $0.002 million.

We also expect a small recurring cost associated with the payment of an annual fee to maintain e-certification in the reporting system. We anticipate an annual $10 search cost that applies to each affected laboratory. Multiplying by the number of total laboratories, we estimate this recurring cost to range from $0.01 million to $0.02 million, with a primary estimate of $0.01 million.

Finally, we expect a recurring cost associated with filing and submitting MDRs. We estimate it will take computer and information system managers 430 hours, working with an hourly wage of $79.72 ($159.44 fully loaded). Multiplying by the number of affected entities, we estimate this recurring cost to range from $40.47 million to $161.88 million, with a primary estimate of $80.94 million.

Overall, we expect the total one-time costs for complying with MDR requirements in Stage 1 of the phaseout policy to range from $5.36 million to $32.57 million, with a primary estimate of $13.21 million. The estimated total recurring costs range from $40.91 million to $164.55 million, with a primary estimate of $82.03 million. See Table 21.

**Table 21. Costs of Medical Device Reporting**
| One-time/Annual          | Primary | Low | High |
|--------------------------|---------|-----|------|
| **Establish SOPs**       |         |     |      |
| Hours                    | 10      | 8   | 12   |
| Wage                     | $122.72 | $122.72 | $122.72 |
| Employees                | 2       | 1   | 3   |
| Entities affected        | 1,181   | 590 | 2,362 |
| New entities per year    | 94      | 47  | 189  |
| **One-time Subtotal (millions)** | $2.90 | $0.58 | $10.43 |
| **Recurring Subtotal (millions)** | $0.23 | $0.05 | $0.83 |
| **Install and Validate e-Submitter Software** | | | |
| Hours                    | 52      | 48  | 56   |
| Wage                     | $159.44 | $159.44 | $159.44 |
| Employees                | 1       | 1   | 1   |
| Entities affected        | 1,181   | 590 | 2,362 |
| New entities per year    | 94      | 47  | 189  |
| **One-time Subtotal (millions)** | $9.79 | $4.52 | $21.09 |
| **Recurring Subtotal (millions)** | $0.78 | $0.36 | $1.69 |
| **Establish HL7ICSR capability** | | | |
| Hours                    | 50      | 48  | 52   |
| Wage                     | $159.44 | $159.44 | $159.44 |
| Employees                | 1       | 1   | 1   |
| Entities affected        | 7       | 4   | 14   |
| New entities per year    | 1       | 0   | 1   |
| **One-time Subtotal (millions)** | $0.06 | $0.03 | $0.12 |
| **Recurring Subtotal (millions)** | $0.005 | $0.002 | $0.01 |
| **Acquiring e-Certificate** | | | |
| Search cost              | $20.00  | $20.00 | $20.00 |
| Entities affected        | 1,181   | 590 | 2,362 |
| New entities per year    | 94      | 47  | 189  |
| **One-time Subtotal (millions)** | $0.02 | $0.01 | $0.05 |
| **Recurring Subtotal (millions)** | $0.002 | $0.001 | $0.004 |
| **Recurring Annual**     |         |     |      |
| Maintaining Certificates |         |     |      |
| Search cost              | $10.00  | $10.00 | $10.00 |
| Entities affected        | 1,181   | 590 | 2,362 |
| **Recurring Subtotal (millions)** | $0.01 | $0.01 | $0.02 |
| Filing and submitting MDRs | |    |      |
| Hours                    | 430     | 430 | 430 |
| Wage                     | $159.44 | $159.44 | $159.44 |
| Entities affected        | 1,181   | 590 | 2,362 |
| **Recurring Subtotal (millions)** | $80.94 | $40.47 | $161.88 |
| **Total One-time Costs (millions)** | $13.21 | $5.36 | $32.57 |
| **Total Recurring Costs (millions)** | $82.03 | $40.91 | $164.55 |
||

Notes: Total one-time and recurring costs include both costs to industry and FDA. See section II.G for FDA review costs of MDRs.

c. Correction and Removal Reporting

Under Stage 1, FDA expects laboratories to comply with correction and removal reporting requirements under 21 U.S.C. 360i(g) and 21 CFR part 806. In estimating the costs of compliance for laboratories, we use information from the 2023 FDA notice: *Agency Information Collection Activities; Submission for Office of Management and Budget Review; Comment Request; Medical Devices; Reports of Corrections and Removals* (Ref. [56]).

We expect that the majority of correction and removal reporting costs will be recurring costs associated with creating correction and removal reports. At baseline, the requirement to create such reports generally has not been enforced.

We expect 50% of laboratories to purchase a digital verification certificate to assist with correction and removal reporting (Ref. [56]). We expect this certificate to cost $50. Multiplying by the number of affected entities, we expect a one-time cost of purchasing a digital verification certificate to range from $0.01 million to $0.06 million, with a primary estimate of $0.03 million. Multiplying by the number of new entities per year, we expect a recurring cost of purchasing a digital verification certificate to range from $1,181 to $4,723, with a primary estimate of $2,362.

We expect laboratories to incur a recurring cost associated with correction and removal reporting requirements. We assume it will take a single general/operations manager working at an hourly wage of $57.60 ($115.20 fully-loaded) 10 hours to create a single correction and removal report. The 2023 FDA notice *Agency Information Collection Activities; Submission for Office of Management and Budget Review; Comment Request; Medical Devices; Reports of Corrections and Removals* acknowledged 1,033 correction and removal reports per year.

These numbers suggest that there are approximately 0.11 correction and removal reports per year per entity. We assume that ratio is the same for laboratories and apply the ratio to the total number of affected entities. Multiplying all elements together, we estimate the recurring cost of correction and removal reporting to range between $0.07 million to $0.30 million, with a primary estimate of $0.15 million.

Overall, we expect the total one-time costs for correction and removal reporting in Stage 1 of the phaseout policy to range between $0.01 million to $0.06 million, with a primary estimate of $0.03 million. The estimated total recurring costs range from $0.08 million to $0.30 million, with a primary estimate of $0.15 million. See Table 22.

**Table 22. Costs of Correction and Removal Reporting**
|                                | Primary | Low   | High  |
|--------------------------------|---------|-------|-------|
| **One-time/Annual**            |         |       |       |
| Digital Verification Certificate |         |       |       |
| Flat fee                       | $50.00  | $50.00| $50.00|
| Entities affected              | 590     | 295   | 1,181 |
| New entities per year          | 47      | 24    | 94    |
| **One-time Subtotal (millions)** | $0.03  | $0.01 | $0.06 |
| **Recurring Subtotal**         | $2,362  | $1,181| $4,723|
| **Recurring Annual**           |         |       |       |
| Reporting                      |         |       |       |
| Hours per report               | 10      | 10    | 10    |
| Number of reports per entity   | 0.11    | 0.11  | 0.11  |
| Wage                           | $115.20 | $115.20| $115.20|
| Entities affected              | 1,181   | 590   | 2,362 |
| **Recurring Subtotal (millions)** | $0.15  | $0.07 | $0.30 |
| **Total One-time Costs (millions)** | $0.03  | $0.01 | $0.06 |
| **Total Recurring Costs (millions)** | $0.15  | $0.08 | $0.30 |
||

d. Complaint Files

Under Stage 1, FDA will expect laboratories to comply with quality system (QS) requirements under 21 CFR part 820.198 (complaint files).[^56] In estimating the costs of complaint files, we use number of annual labor hours and proportion of types of labor (from vice president to clerical staff) needed to comply with complaint file requirements (Ref. [57]).[^57] We also use appropriate wage rates and number of affected entities to estimate costs of complaint files. The estimated total one-time costs for complaint files range from $0.60 million to $6.04 million, with a primary estimate of $2.11 million. The estimated total recurring costs range from $0.01 million to $0.05 million, with a primary estimate of $0.02 million.

**Table 23. Costs of Complaint Files**
|                                  | Primary | Low  | High |
|----------------------------------|---------|------|------|
| **One-time**                     |         |      |      |
| 820.198 Complaint files          | Hours   | 14   | 8    | 20   |
| Entities affected                | 1,181   | 590  | 2,362|
| **Recurring Annual**             |         |      |      |
| 820.198 Complaint files          | Hours   | 2    | 1    | 2    |
| New entities per year            | 94      | 47   | 189  |
| **Total One-time Costs (millions)** | $2.11  | $0.60| $6.04|
| **Total Recurring Costs (millions)** | $0.02  | $0.01| $0.05|
||

###### 2. Costs Under Stage 2

Under Stage 2, FDA will expect that laboratories comply with requirements not covered during other stages of the phaseout policy beginning 2 years after the publication of the phaseout policy. These requirements include registration and listing requirements (21 U.S.C. 360 and 21 CFR part 807, excluding subpart E), labeling requirements (21 U.S.C. 352 and 21 CFR parts 801 and 809, subpart B), and investigational use requirements (21 U.S.C. 360j(g) and 21 CFR part 812).[^58]

a. Registration and Listing

Under Stage 2, FDA expects laboratories to comply with registration and listing requirements under 21 U.S.C. 360 and 21 CFR part 807 (excluding subpart E). In estimating the costs of compliance for laboratories, we use a similar approach to the 2016 *Requirements for Foreign and Domestic Establishment Registration and Listing for Human Drugs, Including Drugs That Are Regulated Under a Biologics License Application, and Animal Drugs* final regulatory impact analysis (Ref. [58]).[^59] We anticipate one-time costs associated with registration and listing requirements and recurring costs associated with re-registration.

We expect the registration and listing will take a general/operations manager 3 hours, working at a wage of $57.60 ($115.20 fully loaded), to complete registration for a single establishment and to list that establishment’s IVDs offered as LDTs.[^60] We also expect that annual re-registration and listing updates will take a general/operations manager 1 hour. Multiplying by the numbers of affected entities per year, we expect total one-time costs for registration and listing requirements to range between $0.20 million and $0.82 million, with a primary estimate of $0.41 million. The estimated total recurring costs range from $0.08 million to $0.34 million, with a primary estimate of $0.17 million. See Table 24.

**Table 24. Costs of Registration and Listing**
|                               | Primary | Low   | High  |
|-------------------------------|---------|-------|-------|
| **One-time/Annual**           |         |       |       |
| Initial registration and listing of IVDs offered as LDTs |         |       |       |
| Hours                         | 3       | 3     | 3     |
| Wage                          | $115.20 | $115.20 | $115.20 |
| Entities affected             | 1,181   | 590   | 2,362 |
| New entities per year         | 94      | 47    | 189   |
| *One-time Subtotal (millions)*| $0.41   | $0.20 | $0.82 |
| *Recurring Subtotal (millions)* | $0.03 | $0.02 | $0.07 |
| **Recurring Annual**          |         |       |       |
| Re-registration               |         |       |       |
| Hours                         | 1       | 1     | 1     |
| Wage                          | $115.20 | $115.20 | $115.20 |
| Entities affected             | 1,181   | 590   | 2,362 |
| *Recurring Subtotal (millions)* | $0.14 | $0.07 | $0.28 |
| **Total One-time Costs (millions)** | **$0.41** | **$0.20** | **$0.82** |
| **Total Recurring Costs (millions)** | **$0.17** | **$0.08** | **$0.34** |
||

b. **Labeling**

Under Stage 2, FDA expects laboratories to comply with labeling requirements under 21 U.S.C. 352, 21 CFR part 801, and 21 CFR part 809, subpart B. We anticipate one-time and recurring costs associated with revising existing labeling.

We expect it will take a general/operations manager, working at a wage of $57.60 ($115.20 fully-loaded),[^61] 8 to 54 hours (with a primary estimate of 31 hours) for regulatory affairs personnel and production personnel per laboratory to redesign existing labeling for IVDs offered as LDTs to comply with labeling requirements (Ref. [59]).[^62] Manufacturers would spend these hours to revise the labeling, including among other things, to perform internal review of the new content, to prepare and proofread new artwork, to replace labeling in the production system, and to submit the file to the agency. Multiplying by the number of expected entities, we expect the one-time cost of revising existing labeling to range between $0.54 million and $14.69 million, with a primary estimate of $4.22 million.[^63] Multiplying the estimates by the number of new entities per year, we expect the recurring cost to range between $0.04 million to $1.18 million, with a primary estimate of $0.34 million. See Table 25.

**Table 25. Costs of Labeling**
|                             | Primary | Low   | High  |
|-----------------------------|---------|-------|-------|
| Revise existing labeling    |         |       |       |
| Hours                       | 31      | 8     | 54    |
| Wage                        | $115.20 | $115.20 | $115.20 |
| Entities affected           | 1,181   | 590   | 2,362 |
| New entities per year       | 94      | 47    | 189   |
| *One-time Subtotal (millions)* | $4.22 | $0.54 | $14.69 |
| *Recurring Subtotal (millions)* | $0.34 | $0.04 | $1.18 |
| **Total One-time Costs (millions)** | **$4.22** | **$0.54** | **$14.69** |
| **Total Recurring Costs (millions)** | **$0.34** | **$0.04** | **$1.18** |
||

c. Investigational Use Requirements

Under Stage 2, FDA expects laboratories to comply with investigational use requirements under 21 U.S.C. 360j(g) and 21 CFR part 812. Investigational medical devices (i.e., that are the object of a clinical investigation or research involving one or more subjects to determine device safety and/or effectiveness) that have an approved investigational device exemption (IDE) application, that are considered to have an approved IDE under 21 CFR part 812.2(b), or that are exempt from most of the requirements in 21 CFR part 812 under 21 CFR 812.2(c), are exempted from various other requirements under the FD&C Act and FDA’s regulations, such as premarket approval. We anticipate one-time and annual costs associated with complying with investigational device exemption requirements under 21 U.S.C. 360j(g) and 21 CFR part 812.[^64]

We expect the cost of developing an IDE application for an IVD offered as an LDT to be $48,000 (Ref. [60]).[^65] We assume two percent of the existing IVDs offered as LDTs are investigational, based on extrapolation of internal information from NYSDOH regarding the percent of IVD submissions they receive that are for investigational IVDs offered as LDTs (Ref. [28]). NYSDOH receives IVD submission packages for IVDs offered as LDTs that are not “designated as FDA-cleared, approved, or exempt,” (Ref. [29]) and these submission packages include clinical trial tests as well as high, moderate, and low risk tests offered for clinical use, based on NYSDOH criteria. Over a two-year period, approximately two percent of IVD submission packages received by NYSDOH were for clinical trial IVDs per NYSDOH criteria.

Not all investigational IVDs require an IDE application.[^66] Based on the number of IVD IDE submissions and the number of IVD premarket submissions that FDA received over a four-year period, we estimate that we receive about 13.5 IVD IDE submissions for every 100 premarket submissions. Therefore, we estimate that about 13.5% of investigational IVDs offered as LDTs that would later be subject to premarket review would first submit an IDE application. We estimate that 50% of IVDs are exempt from premarket notification and 50% require a premarket submission. Applying these factors, we estimate that 6.75% (which represents 50% x 13.5%) of investigational IVDs would require an IDE application.

The number of IDE applications for IVDs currently offered as an LDT can be estimated by multiplying the percent of investigational IVDs currently offered as an LDT (2%) by the percent of investigational IVDs that would require an IDE application (6.75%) by the number of affected IVDs offered as LDTs.

We also expect there would be new investigational IVDs introduced every year, at a rate of anywhere between 1% and 100% of new IVDs. To account for our uncertainty, we assume that the mean value between 1% and 100% or 50% of the new IVDs would be investigational. As described above, we estimate that 6.75% of investigational IVDs would require an IDE.

Multiplying the cost estimates from literature by the relevant percentages and number of affected IVDs offered as LDTs, we expect the total one-time costs of preparing and submitting IDE applications for the existing IVDs offered as LDTs to range between $2.56 million and $10.25 million, with a primary estimate of $5.13 million. Table 26 shows the estimated annual costs, which range from $6.20 million to $24.79 million, with a primary estimate of $12.40 million.

**Table 26. Costs of Complying with Investigational Use Requirements**
| One-time                            | Primary | Low    | High   |
|-------------------------------------|---------|--------|--------|
| Total cost of preparing/            |         |        |        |
| submitting IDE                      |         |        |        |
| Inflation-adjusted estimate from    | $48,000 | $48,000| $48,000|
| literature                          |         |        |        |
| Percent of IVDs offered as LDTs     | 2       | 2      | 2      |
| that are investigational            |         |        |        |
| Percent of investigational IVDs     | 6.75    | 6.75   | 6.75   |
| offered as LDTs that require        |         |        |        |
| submission of IDE application       |         |        |        |
| IVDs currently offered as LDTs      | 79,114  | 39,557 | 158,227|
| affected                            |         |        |        |
| **One-time Subtotal (millions)**    | **$5.13**| **$2.56**| **$10.25**|
| Annual                              | Primary | Low    | High   |
|-------------------------------------|---------|--------|--------|
| Total cost of preparing/            |         |        |        |
| submitting IDE                      |         |        |        |
| Inflation-adjusted estimate from    | $48,000 | $48,000| $48,000|
| literature                          |         |        |        |
| Percent of IVDs offered as LDTs     | 50      | 50     | 50     |
| that are investigational            |         |        |        |
| Percent of investigational IVDs     | 6.75    | 6.75   | 6.75   |
| offered as LDTs that require        |         |        |        |
| submission of IDE application       |         |        |        |
| New IVDs offered as LDTs per year   | 7,652   | 3,826  | 15,303 |
| **Annual Subtotal (millions)**      | **$12.40**| **$6.20**| **$24.79**|

**Total One-time Costs (millions)**   | **$7.45**| **$3.73**| **$14.90**|
**Total Annual Costs (millions)**     | **$18.02**| **$9.01**| **$36.03**|
||

Notes: Total one-time and recurring costs include both costs to industry and FDA. See section II.G for FDA review costs of IDEs.

###### 3. Costs Under Stage 3

Beginning 3 years after the publication of this final rule, FDA will expect compliance with the device current good manufacturing practices (CGMP) requirements of the QS requirements under 21 U.S.C. 360j(f) and 21 CFR part 820 (other than requirements under § 820.198 (complaint files), which are already addressed in stage 1).

However, for LDTs, FDA expects compliance with some, but not all, of the QS requirements. As described in section V.C.3 of the preamble, for these LDTs, FDA expects compliance with:

- design controls under 21 CFR 820.30;
- purchasing controls (including supplier controls) under 21 CFR 820.50;
- acceptance activities (receiving, in-process, and finished device acceptance) under 21 CFR 820.80 and 21 CFR 820.86;
- corrective and preventative actions (CAPA) under 21 CFR 820.100; and
- records requirements under 21 CFR part 820, subpart M (including requirements regarding complaint files under 21 CFR 820.198, for which FDA expects compliance during stage 1 of the phaseout policy).

As further described in section V.C.3 of the preamble, for IVDs that are within the scope of the phaseout policy but for which all manufacturing activities do not occur within a single laboratory, or which are transferred outside of that single laboratory, FDA also expects compliance with the other QS requirements under 21 U.S.C. 360j(f) and 21 CFR part 820. We lack evidence to quantify the numbers of such IVDs. To account for uncertainty, we consider different assumptions for low, primary, and high estimates. To estimate a lower bound estimate, we first assume that for all IVDs within the scope of the phaseout policy, all manufacturing activities occur within a single laboratory and, therefore, have zero costs associated with the QS requirements other than those listed above. For an upper bound estimate, we assume that all manufacturing activities do not occur within a single laboratory for any IVD within the scope of the phaseout policy and, therefore, have costs associated with all QS requirements. Since we expect there to be a mix of these two extremes within the scope of the phaseout policy, we use an average of the lower and upper bound estimates for our primary estimate.

In estimating the costs of compliance for laboratories, we use number of annual labor hours and proportion of types of labor (from vice president to clerical staff) needed to comply with each relevant provision of 21 CFR part 820. We also use wage rates to estimate costs of...

compliance with these provisions for affected entities (see Table 27). Table 28 shows the number of labor hours for compliance with each provision of Part 820 (Ref. [57]). We multiply the labor hours by appropriate wage rates and number of affected entities to estimate costs of compliance with the QS requirements under this stage.

Since FDA generally intends to exercise enforcement discretion with respect to QS requirements (other than requirements regarding records) for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified as described in the preamble, we anticipate one-time costs of compliance with records requirements other than complaint files from existing IVDs offered as LDTs and recurring costs from new IVDs offered as LDTs under this stage. In addition, we estimate that the number of affected entities for Stage 3 is lower than Stages 1 and 2 as FDA intends to exercise enforcement discretion with respect to QS requirements (other than requirements regarding records) for LDTs manufactured and performed by a laboratory integrated within a health care system to meet an unmet need of patients receiving care within the same healthcare system. Further, it is our understanding, based on consultation with NYS CLEP, that compliance with NYS CLEP’s clinical laboratory standards (which exceed CLIA requirements in certain respects) and its premarket review requirements collectively could generally satisfy these subparts of the QS regulations except as to certain aspects of design control documentation. Therefore, FDA does not anticipate significant additional burden with respect to compliance with these QS requirements for laboratories offering LDTs approved by NYS CLEP. As discussed in appendix A, we estimate 12.1 percent of new premarket submissions for affected tests will be reviewed by NYS CLEP. For the purpose of estimating costs associated with compliance with Quality System requirements, we extrapolate this to the affected laboratories, estimating that 12.1 percent of affected laboratories (1,181) will have their LDTs reviewed by NYS CLEP. We also estimate that 40 to 70 percent of LDTs from the high complexity laboratories integrated within health care systems (459) are likely to be for unmet needs. We therefore estimate the affected laboratories incurring costs under Stage 3 to be 786 (=1,181 – 1,181*0.121 – 459*0.55).

We expect the total one-time costs for QS requirement (records other than complaint files under 21 CFR 820.198) in Stage 3 of the phaseout policy to range from $0.40 million to $4.02 million, with a primary estimate of $1.41 million. The total recurring costs are estimated to range from $1.94 million to $124.01 million, with a primary estimate of $24.66 million. See Table 28.

**Table 27. Medical and Diagnostic Laboratories Industry Wage Rates for Selected Labor Categories**
Labor Category | Wages (/hour) | NAICS | OCC Code
---|---|---|---
Vice president | $59.68 | 621500 | 11-1000
Upper management | $76.38 | 621500 | 11-2000
Middle management | $66.83 | 621500 | 11-3000
Technical | $30.36 | 621500 | 29-0000
Admin support | $32.67 | 621500 | 43-6011
Clerical | $18.37 | 621500 | 43-4000 |
||

**Table 28. Costs of Compliance with Quality System Requirements**
One-time/Annual |  | Primary | Low | High
---|---|---|---|---
820.20(a) Quality Policy | Hours | 8 | 0 | 24
820.20(b) Organization | Hours | 6 | 0 | 20
820.20(d) Quality Planning | Hours | 14 | 0 | 40
820.20(e) Quality System Procedures | Hours | 14 | 0 | 40
820.22 Quality Audit | Hours | 8 | 0 | 24
820.25 Personnel, establish procedures for identifying training needs | Hours | 8 | 0 | 24
820.25 Personnel, train in CGMP revisions | Hours | 50 | 0 | 290
820.40 Document Controls | Hours | 14 | 0 | 40
820.60 Identification and Traceability | Hours | 8 | 0 | 24
820.72, 820.75 Inspection, measuring, and test equipment, process validation | Hours | 23 | 0 | 72
820.70(i) Automated Processes | Hours | 14 | 0 | 40
820.90 Nonconforming Product | Hours | 14 | 0 | 40
820.140 Handling | Hours | 8 | 0 | 24
820.200 Servicing | Hours | 14 | 0 | 40
820.30(a) General | Hours | 200 | 30 | 560
820.50(a) Assessment of Suppliers and Contractors | Hours | 75 | 25 | 125
820.100 Corrective and Preventive Action | Hours | 28 | 16 | 40
820.150 Storage | Hours | 15 | 8 | 24
820.198 Complaint Files | Hours | 14 | 8 | 20
Entities affected | | 786 | 393 | 1,571
New entities per year | | 63 | 31 | 126 |
**Recurring Annual** |  |  |  | |
820.20(a) Quality Policy | Hours | 1 | 0 | 2
820.20(b) Organization | Hours | 1 | 0 | 2
820.20(c) Management Review | Hours | 8 | 0 | 24
820.20(d) Quality Planning | Hours | 4 | 0 | 10
820.20(e) Quality System Procedures | Hours | 4 | 0 | 10
820.22 Quality Audit | Hours | 1 | 0 | 2
820.25 Personnel, maintain procedures | Hours | 1 | 0 | 2
820.40 Document Controls | Hours | 2 | 0 | 4
820.60 Identification and Traceability | Hours | 1 | 0 | 2
820.72, 820.75 Inspection, measuring, and test equipment, process validation | Hours | 4 | 0 | 13
820.70(i) Automated Processes | Hours | 2 | 0 | 4
820.90 Nonconforming Product | Hours | 2 | 0 | 4
820.140 Handling | Hours | 1 | 0 | 2
820.200 Servicing | Hours | 2 | 0 | 4
820.30(a) General | Hours | 20 | 3 | 56
820.30(b) Design and Development Planning | Hours | 216 | 32 | 520
820.30(e) Design Review | Hours | 942 | 82 | 2,574
820.30(f) Design Verification | Hours | 1,681 | 249 | 4,047
820.30(h) Design Transfer | Hours | 43 | 6 | 104
820.30(i) Design Changes | Hours | 378 | 56 | 910
820.30(j) Design History File | Hours | 22 | 3 | 52
820.50(a) and (b) Purchasing control | Hours | 159 | 98 | 233
820.100 Corrective and Preventive Action | Hours | 3 | 2 | 4
820.150 Storage | Hours | 2 | 1 | 2
820.198 Complaint Files | Hours | 2 | 1 | 2
New entities per year | | 63 | 31 | 126
Total One-time Costs (millions) | | $1.41 | $0.40 | $4.02
Total Recurring Costs (millions) | | $24.66 | $1.94 | $124.01 |
||

We note that on February 2, 2024, FDA issued a final rule amending the device QS regulation, part 820, to align more closely with international consensus standards for devices (87 FR 10119). Specifically, FDA withdrew the majority of the current requirements in part 820 and instead incorporated by reference the 2016 edition of the International Organization for Standardization (ISO) 13485, Medical devices – Quality management systems for regulatory purposes, in part 820. As stated in that rule, the requirements in ISO 13485 are, when taken in totality, substantially similar to the requirements of the current part 820, providing a similar level of assurance in a firm’s quality management system, and FDA intends for the phaseout policy to apply with respect to the regulations promulgated through that rulemaking.

The amended QS requirements will take effect on February 2, 2026, before the beginning of Stage 3. Upon the start of Stage 3, or if the laboratory complies with QS requirements prior to the start of Stage 3, FDA expects compliance with the QS requirements that are in effect at that time.69 For further information on the QS requirements established pursuant to the amendments to the QS regulation, please refer to 89 FR 7496. Notably, the requirements relating to design controls, purchasing controls, acceptance activities, CAPA, and records requirements are set forth in the following ISO 13485 clauses as modified by the regulatory text for part 820:

- Clause 4. Quality Management System, Subclause 4.2.5;
- Clause 6. Resource Management;
- Clause 7. Product Realization, Subclause 7.1, Subclause 7.3, Subclause 7.4, and Subclause 7.4.3; and
- Clause 8. Measurement, Analysis, & Improvement, Subclause 8.2.2, Subclause 8.2.5, Subclause 8.2.6, and Subclause 8.3.

To the extent amended QS requirements are in effect, we do not expect the total costs for compliance with QS requirements in Stage 3 to substantially change (89 FR 7496, February 2, 2024).

###### 4. Costs Under Stages 4 and 5

Beginning 3½ years after the publication of this final rule, FDA will expect laboratories to comply with premarket review requirements for high-risk IVDs offered as LDTs (21 U.S.C. 360e and 21 CFR part 814). Laboratories will face costs of preparing and submitting premarket approval (PMA) applications and PMA supplements as well as greater annual reporting burdens associated with premarket approval. FDA will also face additional costs of reviewing the applications. We quantify these costs in the following sections.

Additionally, moderate risk IVDs offered as LDTs (IVDs that may be eligible for classification into class II) and low risk IVDs offered as LDTs (IVDs that may be eligible for classification into class I) that require a premarket submission will be expected to comply with 510(k) requirements or De Novo requirements beginning 4 years after the publication of the phaseout policy. Under this stage, we anticipate costs associated with preparing and submitting 510(k) premarket notifications or De Novo classification requests, and FDA review costs.

FDA generally intends to continue to exercise enforcement discretion with respect to premarket review for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified as described in the preamble. FDA also generally intends to exercise enforcement discretion with respect to premarket review requirements for LDTs that receive approval through NYS CLEP. In addition, FDA generally intends to exercise enforcement discretion with respect to premarket review requirements for LDTs manufactured and performed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system.

a. Number of Premarket Submissions

Due to the variations in the size of laboratories, business models, and types of IVDs, there is no comprehensive database or repository from which we can definitively calculate the number of IVDs offered as LDTs currently available or the rate at which new IVDs offered as LDTs are introduced. Likewise, there is insufficient data to definitively determine what percentage of IVDs offered as LDTs are likely to be in each class of devices. We rely on New York State Department of Health internal data to estimate the number of affected IVDs offered as LDTs (see section II.D.1 and Table 2).

As discussed in section II.D.1, we assume one laboratory offers 67 IVDs as LDTs and will offer 6 new IVDs as LDTs per year. Of the 67 IVDs as LDTs currently offered per laboratory, we assume that, on average, two will be modified in such a way as to require premarket review per year until they all are FDA authorized. As mentioned above, FDA
Because FDA generally intends to exercise enforcement discretion with respect to premarket review for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of this rule and that are not modified, or that are modified as described in the preamble, we only include costs from modified and new IVDs offered as LDTs under Stages 4 and 5.

**Table 29. Number of Premarket Submissions Under Stage 4 and Stage 5**
|                               | Primary | Low  | High |
|-------------------------------|---------|------|------|
| PMA                           | 101     | 51   | 203  |
| PMA supplements               | 31      | 15   | 61   |
| 510(k) Total                  | 2,179   | 1,090| 4,359|
| 510(k) with method comparison study  | 1,286   | 643  | 2,572|
| 510(k) with moderately complex clinical study | 894 | 447  | 1,787|
| De Novo                       | 267     | 134  | 534  |
| **Total number of new premarket submissions per year for affected tests** | **2,578** | **1,289** | **5,157** |
||

*Notes: The numbers of tests include currently marketed tests that would be modified per year and new tests from both affected labs and new labs entering the market per year.*

b. PMA, 510(k), and De Novo requirements

In estimating the costs of compliance for laboratories, we use estimates for the 510(k) and the premarket approval processes derived by Eastern Research Group (ERG) (Ref. [61])[^70]. The estimates by ERG present the representative costs of regulatory-related activities based on semi-structured discussions with project consultants and other information and knowledge about the development process[^71].

Devices subject to premarket approval typically require premarket and post-market procedures that are not typically associated with 510(k) clearance, such as premarket manufacturing site and clinical site inspections and annual report submissions. In addition, the requirements relating to submissions for device modifications are generally different for devices that have received PMAs as compared with other devices. For example, supplements must be approved, such as for the use of a different facility or establishment to manufacture, process, or package the device. We have excluded costs that would already be part of compliance with the QS requirements under Stage 3, including costs of developing design controls, acquiring GMP-compliant manufacture capability, and developing a risk management system.

To estimate cost for submission and preparation of the PMA, IVDs are broken out by complexity of the clinical trial supporting IVD safety and effectiveness due to the different costs. We use the ERG estimates of the PMAs with complex clinical trials for lower bound estimates (Ref. [61]). For upper bound estimates, we use the ERG estimates of the PMAs with complex, extensive clinical trials. We updated the ERG estimates to account for inflation. We expect that most of the PMAs will involve complex clinical trials. We assume that of the PMAs, 95% are complex clinical trials and 5% are complex, extensive clinical trials. We take 95% of the low and 5% of the high estimates to calculate primary estimates. The total cost of submission and preparation per PMA is estimated to range from $4.10 million to $9.29 million, with a primary estimate of $4.36 million. Multiplying the estimates by the numbers of new IVDs per year and IVDs from new entities per year that are subject to premarket approval, excluding those that would be under enforcement discretion policies, we expect recurring cost of submission and preparation for PMAs to range from $207.60 million to $1,881.11 million, with a primary estimate of $441.47 million.

PMA holders are also subject to annual reporting requirements, which impose preparation costs on PMA holders and review costs on FDA. We use a prior estimate from the Microbiology Devices; Reclassification of Nucleic Acid-Based Systems for *Mycobacterium tuberculosis* complex final regulatory impact analysis (Ref. [62]) to estimate the recurring preparation cost[^72]. The current estimate after adjustment for inflation is $11,798 per PMA. Multiplying the estimates by the numbers of PMA submissions per year, we expect total recurring costs of PMA annual reporting requirements to range from $0.60 million to $2.39 million, with a primary estimate of $1.19 million.

Overall, we estimate the total recurring costs to industry of PMA requirements in Stage 4 to range between $208.20 million and $1,883.50 million, with a primary estimate of $442.67 million. See Table 30.

**Table 30. Costs to Industry of Premarket Approval Application**
| **Cost of Submission and Preparation** | Primary | Low | High |
|----------------------------------------|---------|-----|------|
| Develop necessary SOPs                | $39,572   | $37,688   | $75,376   |
| Hold pre-submission meeting with FDA  | $2,513    | $2,513    | $2,513    |
| Prepare indications for use           | $25,125   | $25,125   | $25,125   |
| Perform clinical trials               | $2,832,871| $2,638,150| $6,532,562|
| Preparing labeling                    | $25,125   | $25,125   | $25,125   |
| Pre-approval inspection               | $115,576  | $115,576  | $115,576  |
| Prepare regulatory submission         | $1,319,075| $1,256,262| $2,512,524|
| **Subtotal cost per submission**      | $4,359,857| $4,100,439| $9,288,800|
| No. PMA submissions per year for affected tests | 101 | 51 | 203 |
| **Recurring Subtotal (millions)**     | **$441.47** | **$207.60** | **$1,881.11** |
| **Recurring Annual**                  |           |           |           |
| Annual Report preparation for existing PMAs | $11,798 | $11,798 | $11,798 |
| No. PMA submissions per year for affected tests | 101 | 51 | 203 |
| **Recurring Subtotal (millions)**     | **$1.19** | **$0.60** | **$2.39** |
| **Total Recurring Costs (millions)**  | **$442.67** | **$208.20** | **$1,883.50** |
||

Notes:
Unless otherwise specified, line-item estimates are inflation-adjusted estimates from Eastern Research Group, Inc. 2012: Economic Analysis of CDRH Submission Requirements. Totals may not add due to rounding. The numbers of PMA submissions per year include currently marketed tests that would be modified per year and new tests from both affected entities currently on the market and new entities entering the market per year. Total recurring costs do not include costs to FDA. See section II.G for FDA review costs of Q-submissions and PMAs.
*We calculate subtotals by multiplying subtotal cost per submission by the number of affected IVDs.

Some IVDs with PMAs might require a PMA supplement under 21 CFR 814.39 when certain modifications are made.[^73] There are several types of PMA supplements (see Table 31; each row is a type of PMA supplement). We first estimate the expected number of PMA supplements by supplement type by multiplying the number of expected PMAs by the number of expected PMA supplements per PMA[^74] and the share of supplements by supplement type.[^75] We also multiply by the Remaining PMA Rate (Table A.4, Column F) to adjust for potential reclassification of Class III IVDs into Class II IVDs. We assume that entities would submit PMA supplements in year 4. See Table 31 for the expected number of annual PMA supplements.

Next, to estimate the total costs to industry of PMA supplement preparation, we multiply the number of PMA supplements by an estimated full-time equivalent (FTE) cost[^76] associated with each supplement type and the cost of preparing a PMA from the previous section. This approach assumes the cost of preparing a PMA supplement for a laboratory is proportional to the FTE required for FDA to review the supplement type. Overall, we estimate the total recurring costs to industry of PMA supplements to range from $6.28 million to $56.89 million, with a primary estimate of $13.35 million. See Table 32.

**Table 31. Number of PMA Supplements by Submission Type**
| Submission Type | Cumulative share of supplements by type | Primary | Low | High |
|-------------------------------------------------------|----------------------------------------|---------|-----|------|
| 135 Review Track                                       | 0.053                                  | 2       | 1   | 5    |
| Normal 180-day track                                   | 0.205                                  | 5       | 3   | 11   |
| Normal 180-day track - No user fee                     | 0.128                                  | 4       | 2   | 7    |
| Panel-Track                                            | 0.067                                  | 2       | 1   | 4    |
| Real-Time Process                                      | 0.374                                  | 14      | 7   | 28   |
| Special CBE                                            | 0.095                                  | 4       | 2   | 7    |
||

**Table 32. Costs to Industry of PMA Supplements**
|     Submission Type     | Adjusted FTE weights over PMA | Primary | Low   | High  |
|--------------------------------------------------------|-------------------------------|---------|-------|-------|
| 135 Review Track                                        | 0.033                         | $0.42   | $0.20 | $1.79 |
| Normal 180-day track                                    | 0.033                         | $0.98   | $0.46 | $4.18 |
| Normal 180-day track - No user fee                      | 0.033                         | $0.67   | $0.32 | $2.88 |
| Panel Track                                             | 1.000                         | $9.85   | $4.63 | $41.98|
| Real Time Process                                       | 0.010                         | $0.79   | $0.37 | $3.35 |
| Special CBE                                             | 0.033                         | $0.64   | $0.30 | $2.71 |
| **Total Recurring Costs (millions)**                    |                               | **$13.35** | **$6.28** | **$56.89** |
||

Note: Total recurring costs do not include costs to FDA. See section II.G for FDA review costs of PMA supplements.

Similar to the PMA, we use the ERG estimates of the 510(k) process to estimate the one-time submission and preparation cost of 510(k)s, adjusting for inflation.[^77] We use the ERG estimates of 510(k) with small or simple clinical trials for 510(k) submissions with method comparison studies (see Table 33) (Ref. [61]). We use the ERG estimates of 510(k) with moderately complex clinical trials for 510(k) submissions with moderately complex clinical studies (see Table 34).[^78]

For any 510(k) submission (or De Novo request[^79]), we expect it will take one operations specialist manager, working at a wage of $66.83 ($133.66 fully loaded), 1 to 2 hours (with a primary estimate of 1.5 hours) to identify a predicate device (or determine that no predicate device exists, in the case of a De Novo). The other one-time submission and preparation costs are derived from the ERG estimates. The total cost of submission and preparation per 510(k) with method comparison studies is estimated to range from $215,457 to $279,157, with a primary estimate of $247,307.

FDA also anticipates that laboratories may seek to utilize FDA’s 510(k) Third Party Review Program. Multiple Third Party Review Organizations (3P510k Review Organizations) are accredited to conduct reviews of 510(k) submissions for certain IVDs. Manufacturers who submit to 3P510k Review Organizations pay the 3P510k Review Organization but do not pay FDA user fees for those submissions. Each 3P510k Review Organization sets their own rates, which are generally comparable to FDA user fees. Due to lack of data, we assume that laboratories will pay the same amount of FDA user fees to 3P510k Review Organizations. FDA assumes that at least 50% of the IVDs offered as LDTs being submitted for 510(k) review will be reviewed under the 510(k) Third Party Review Program.

Multiplying by the numbers of modified and new IVDs per year and IVDs from new entities that are subject to 510(k) with method comparison studies, excluding those that would be under enforcement discretion policies per year yields the recurring submission and preparation costs for 510(k)s with method comparison studies is estimated between $142.10 million and $732.21 million, with a primary estimate of $325.15 million. See Table 33.

**Table 33. Costs to Industry of 510(k)s (Method Comparison Study)**
| Cost of Submission and Preparation | Primary   | Low      | High     |
|------------------------------------|-----------|----------|----------|
| Identify predicate device          |           |          |          |
| Hours                              | 1.5       | 1        | 2        |
| Wage                               | $133.66   | $133.66  | $133.66  |
| Develop necessary SOPs             | $37,688   | $37,688  | $37,688  |
| Hold pre-submission meeting with FDA| $2,136   | $1,759   | $2,513   |
| Prepare indications for use        | $25,125   | $25,125  | $25,125  |
| Perform method comparison          | $62,813   | $62,813  | $62,813  |
| Preparing labeling                 | $25,125   | $25,125  | $25,125  |
| Prepare regulatory submission      | $94,220   | $62,813  | $125,626 |
| **Subtotal cost per submission**   | $247,307  | $215,457 | $279,157 |
| No. 510(k) submissions with method comparison study per year for affected tests | 1,286 | 643 | 2,572 |
| **Subtotal (millions)**            | **$317.98** | **$138.52** | **$717.87** |
| **Fees to 3P510k Review Organizations** |       |          |          |
| MDUFA Review Fee                   | $21,760   | $21,760  | $21,760  |
| (Adjusted fee for small entities)  | ($5,440)  | ($5,440) | ($5,440) |
| IVDs Affected, non-small*          | 225       | 113      | 450      |
| IVDs affected, small*              | 418       | 209      | 836      |
| **Subtotal (millions)**            | $7.17     | $3.58    | $14.34   |
| **Total Recurring Costs (millions)** | **$325.15** | **$142.10** | **$732.21** |
||

*Notes:*  
Unless otherwise specified, line-item estimates are inflation-adjusted estimates from Eastern Research Group, Inc. 2012: Economic Analysis of CDRH Submission Requirements.  
Totals may not add due to rounding.  
The numbers of 510(k) submissions per year include currently marketed tests that would be modified and new tests from both affected entities currently on the market and new entities entering the market per year. Total recurring costs does not include costs to FDA. See section II.G for FDA review costs of Q-submissions and 510(k) submissions.  
*We calculate subtotals by multiplying the subtotal cost per submission by the number of affected IVDs.*

Table 34 presents costs of 510(k) submissions with a moderately complex clinical study. We calculate the costs using the exact same methods as in Table 33. The estimated subtotal cost of submission and preparation per submission ranges from $466,709 to $530,410, with a primary estimate of $498,560. Multiplying the estimates by the numbers of modified and new IVDs per year and IVDs from new entities per year that are subject to 510(k) with moderately complex clinical studies, excluding those that would be under enforcement discretion policies, we expect recurring submission and preparation cost to range from $211.00 million to $957.82 million, with a primary estimate of $450.45 million.

**Table 34. Costs to Industry of 510(k)s (Moderately Complex Clinical Study)**
| Cost of Submission and Preparation | Primary   | Low      | High     |
|------------------------------------|-----------|----------|----------|
| Identify predicate device          |           |          |          |
| Hours                              | 1.5       | 1        | 2        |
| Wage                               | $133.66   | $133.66  | $133.66  |
| Develop necessary SOPs             | $37,688   | $37,688  | $37,688  |
| Hold pre-submission meeting with FDA| $2,136   | $1,759   | $2,513   |
| Prepare indications for use        | $25,125   | $25,125  | $25,125  |
| Perform clinical study             | $314,065  | $314,065 | $314,065 |
| Preparing labeling                 | $25,125   | $25,125  | $25,125  |
| Prepare regulatory submission      | $94,220   | $62,813  | $125,626 |
| **Subtotal cost per submission**   | $498,560  | $466,709 | $530,410 |
| No. 510(k) submissions with moderately complex clinical study per year | 894 | 447 | 1,787 |
| **Subtotal (millions)**            | **$445.47** | **$208.51** | **$947.86** |
| **Fees to 3P510k Review Organizations** |       |          |          |
| MDUFA Review Fee                   | $21,760   | $21,760  | $21,760  |
| (Adjusted fee for small entities)  | ($5,440)  | ($5,440) | ($5,440) |
| IVDs Affected, non-small*          | 156       | 78       | 313      |
| IVDs affected, small*              | 290       | 145      | 581      |
| **Subtotal (millions)**            | $4.98     | $2.49    | $9.96    |
| **Total Recurring Costs (millions)** | **$450.45** | **$211.00** | **$957.82** |
||

*Notes:*  
Unless otherwise specified, line-item estimates are inflation-adjusted estimates from Eastern Research Group, Inc. 2012: Economic Analysis of CDRH Submission Requirements.  
Totals may not add due to rounding.  
The numbers of 510(k) submissions per year include currently marketed tests that would be modified and new tests from both affected entities currently on the market and new entities entering the market per year. Total recurring costs does not include costs to FDA. See section II.G for FDA review costs of Q-submissions and 510(k) submissions.  
*We calculate subtotals by multiplying subtotal cost per submission by the number of affected IVDs.*

Table 35 shows costs of a De Novo classification request. We use the ERG estimates of 510(k) with moderately complex clinical trial for upper bound and use the ERG estimates of 510(k) with a method comparison study for lower bound estimates (Ref. [61]). We assume that most De Novo requests would have data from clinical trials. We take 99% of the high and 1% of the low estimates to calculate primary estimates. We calculate costs of De Novo classification requests using the exact same methods as in Table 34. The estimated subtotal cost of submission and preparation per submission ranges from $216,211 to $530,410, with a primary estimate of $527,202. Multiplying the estimates by the numbers of modified and new IVDs per year and IVDs from new entities per year that are subject to De Novo, excluding those that would be under enforcement discretion policies, we expect recurring submission and preparation cost to range from $28.88 million to $283.39 million, with a primary estimate of $140.84 million.

**Table 35. Costs to Industry of De Novo Classification Request**
| Cost of Submission and Preparation | Primary   | Low      | High     |
|------------------------------------|-----------|----------|----------|
| Determine that no predicate devices exist |       |          |          |
| Hours                              | 1.50      | 1.00     | 2.00     |
| Wage                               | $133.66   | $133.66  | $133.66  |
| Develop necessary SOPs             | $37,688   | $37,688  | $37,688  |
| Hold pre-submission meeting with FDA| $2,513   | $2,513   | $2,513   |
| Prepare indications for use | $25,125 | $25,125 | $25,125 |
| Perform method comparison or clinical study | $311,553 | $62,813 | $314,065 |
| Preparing labeling | $25,125 | $25,125 | $25,125 |
| Prepare regulatory submission | $124,998 | $62,813 | $125,626 |
| **Subtotal cost per submission** | **$527,202** | **$216,211** | **$530,410** |
| No. De Novo submissions per year for affected tests | 267 | 134 | 534 |
| **Total Recurring Costs (millions)\*** | **$140.84** | **$28.88** | **$283.39** |
||

Notes:  
Unless otherwise specified, line-item estimates are inflation-adjusted estimates from Eastern Research Group, Inc. 2012: Economic Analysis of CDRH Submission Requirements. Totals may not add due to rounding. The numbers of De Novo submissions per year include currently marketed tests that would be modified and new tests from both affected entities currently on the market and new entities entering the market per year. Total recurring costs does not include costs to FDA. See section II.G for FDA review costs of Q-submissions and De Novo requests.  
*We calculate subtotals by multiplying subtotal cost per submission by the number of affected IVDs.

###### 5. Summary of Costs

Table 36 summarizes our estimates of the one-time and recurring costs by stage of the phaseout policy. These include costs to FDA and costs to industry. We estimate the total one-time costs to range between $11.67 million and $85.30 million, with a primary estimate of $32.45 million. We estimate the total recurring costs to range between $0.72 billion and $4.54 billion, with a primary estimate of $1.65 billion.

**Table 36. Total Costs to FDA and Industry (millions 2022$)**
|  | Primary | Low | High |
|---|---|---|---|
| **Stage 1** |  |  |  |
| Reading and Understanding the Rule | $3.62 | $0.81 | $12.21 |
| Medical Device Reporting | $13.21 | $5.36 | $32.57 |
| Correction and Removal Reporting | $0.03 | $0.01 | $0.06 |
| Complaint Records | $2.11 | $0.60 | $6.04 |
| Registration and Listing Requirements | $0.41 | $0.20 | $0.82 |
| **Stage 2** |  |  |  |
| Labeling Requirements | $4.22 | $0.54 | $14.69 |
| Investigational Use Requirements | $7.45 | $3.73 | $14.90 |
| **Stage 3** |  |  |  |
| Quality System Requirements | $1.41 | $0.40 | $4.02 |
**Total One-time Costs (millions)** | **$32.45** | **$11.67** | **$85.30** 
**Recurring Annual** |  |  |  |
| **Stage 1** |  |  |  |
| Reading and Understanding the Rule | $0.29 | $0.07 | $0.98 |
| Medical Device Reporting | $82.03 | $40.91 | $164.55 |
| Correction and Removal Reporting | $0.15 | $0.08 | $0.30 |
| Complaint Records | $0.02 | $0.01 | $0.05 |
| Registration and Listing Requirements | $0.17 | $0.08 | $0.34 |
| **Stage 2** |  |  |  |
| Labeling Requirements | $0.34 | $0.04 | $1.18 |
| Investigational Use Requirements | $18.02 | $9.01 | $36.03 |
| **Stage 3** |  |  |  |
| Quality System Requirements | $24.66 | $1.94 | $124.01 |
| **Stage 4** |  |  |  |
| Premarket Approval Application | $503.91 | $238.82 | $2,005.97 |
| Premarket Approval Application Supplements | $14.76 | $6.98 | $59.71 |
| **Stage 5** |  |  |  |
| 510(k) Submission | $825.47 | $378.03 | $1,789.77 |
| De Novo Classification Request | $177.50 | $47.21 | $356.72 |
**Total Recurring Costs (millions)** | **$1,647.32** | **$723.18** | **$4,539.61** |
||

Notes:  
The estimated costs include both costs to industry and FDA. The MDRs review costs for Stage 1, the IDEs review costs for Stage 2, and the Q-submission and premarket review costs to FDA for Stages 4 and 5 are reported in section II.G.

Table 37 presents a summary of the estimated twenty-year stream of costs. We expect that total costs for Stage 1 associated with reading and understanding the rule, medical device reporting, correction and removal reporting, and complaint records would occur in the first year after publication of the final rule. In the first year after publication of the final rule, we estimate total costs to range from $47.85 million to $216.75 million, with a primary estimate of $101.46 million.

We expect that total costs for Stage 2 associated with registration and listing requirements, labeling requirements, and investigational use requirements would occur in the second year after publication of the final rule. In year 2, total costs are estimated to range between $54.67 million to $233.83 million, with a primary estimate of $113.09 million.

In the third year after publication of the final rule, we expect that costs for Stage 3 associated with Quality System requirements except for complaint files would occur. We also expect that half of costs for Stage 4 associated with premarket approval applications would occur in year 3. Total costs in year 3 are estimated to range between $175.44 million to $1.36 billion, with a primary estimate of $386.41 million.

In subsequent years, we expect that costs for Stages 4 and 5 associated with PMAs, PMA supplements, 510(k) submissions or De Novo classification requests would occur. The recurring cost for year 4 to year 20 is estimated to range between $723.18 million and $4.54 billion, with a primary estimate of $1.65 billion. We estimate the total costs over 20 years to range from $12.57 billion to $78.99 billion, with a primary estimate of $28.61 billion.

The present value of total estimated costs is $20.41 billion at a 3 percent discount rate and $13.64 billion at a 7 percent discount rate over 20 years. The annualized value of costs is $1.37 billion at a 3 percent discount rate and $1.29 billion at a 7 percent discount rate.

**Table 37. Twenty-Year Timing of the Costs (millions 2022$)**
|  | Primary | Low | High |
|---|---|---|---|
| Year 1 | $101.46 | $47.85 | $216.75 |
| Year 2 | $113.09 | $54.67 | $233.83 |
| Year 3 | $386.41 | $175.44 | $1,364.29 |
| Year 4-20 (costs for each year) | $1,647.32 | $723.18 | $4,539.61 |
**Total Costs** | **$28,605.32** | **$12,571.97** | **$78,988.22**
**Present Value of Total Costs (3%)** | **$20,407.00** | **$8,972.00** | **$56,376.49**
**Present Value of Total Costs (7%)** | **$13,637.63** | **$5,999.18** | **$37,699.79**
**Annualized Value of Costs (3%)** | **$1,371.67** | **$603.06** | **$3,789.39**
**Annualized Value of Costs (7%)** | **$1,287.30** | **$566.28** | **$3,558.59**
||

###### 6. Other Unquantified Costs

Other unquantified social costs associated with the phaseout policy (or consequences of the costs that have been quantified) may include the impact on prices, access to diagnostics if many laboratories exit the market or discontinue offering certain IVDs rather than incur the costs of compliance with FDA requirements, and/or a decrease in the number of new LDTs due to the
increased operation costs of the phaseout policy. There may be instances in which a laboratory may choose to exit the market or discontinue certain IVDs offered as LDTs due to compliance costs. Without information on the revenues or costs of production of IVDs offered as LDTs, however, we are unable to estimate the impact associated with compliance costs on the prevalence of laboratories exiting the market or discontinuing manufacturing of certain IVDs offered as LDTs.

Our analysis in section III (Final small entity analysis) shows that 22% of estimated receipts from IVDs offered as LDTs come from small laboratories (laboratories with annual receipts of less than $41,500,000), which are more likely to reduce operations or exit the market than large laboratories. However, the enforcement discretion policies discussed here -and in the preamble- make it less likely that these smaller laboratories would reduce operations or exit the market.

However, to the extent that some small laboratories might reduce operations or exit the market, it is possible that larger laboratories might take over the production of certain IVDs offered as LDTs, reducing potential impacts on IVD availability. This might concentrate production in a few large laboratories. Under this scenario, prices for certain IVDs offered as LDTs could increase, reducing overall net social benefits. According to economic theory, production concentration under a few laboratories could increase the risk of supply chain contractions, risking shortages for certain IVDs offered as LDTs and therefore affecting prices and access. Although under monopolistic competition, production of more IVDs offered as LDTs in large laboratories could also result in lower production costs due to the economies of scale associated with the operations of such laboratories, they do not produce at the minimum of their average costs curve and may charge prices higher than their marginal cost.

While we recognize that some laboratories might pass the costs of compliance to their customers by raising prices for IVDs offered as LDTs, increased FDA oversight might also help reduce social costs by helping to support coverage and reimbursement determinations and increasing patient accessibility to IVDs for which there is a reasonable assurance of safety and effectiveness.

We also understand that the increased cost to laboratories under the phaseout policy may reduce the amount of revenue a laboratory can invest in creating and/or modifying IVDs offered as LDTs. This could lead to a reduction in the number of new IVDs offered as LDTs and/or modifications of these which incorporate the most up-to-date scientific knowledge. While this may occur, the increased FDA oversight under the phaseout policy may provide more assurance that new and/or modified IVDs offered as LDTs will provide accurate results.

Finally, it is also possible that some laboratories might decide to switch from an IVD offered as an LDT to an FDA-authorized test or to outsource their testing to other laboratories. According to comments, an FDA-authorized test could cost an additional $6 to $35 per test performed. One comment also stated that outsourcing some testing instead of offering IVDs as LDTs could cost them an additional $3,000 to $6,000 per test, while another stated that outsourcing could cost an additional $760,000 annually. An unknown number of laboratories may pursue outsourcing their testing needs or switching to use of an FDA-authorized test rather than introducing a new test that does not fall within an enforcement discretion policy in the phaseout policy. However, we assume that the cost of switching to an FDA-authorized test when available, would cost less than the cost of submitting a premarket submission. We assume that a laboratory would switch to FDA-authorized tests or outsource their testing needs only if submitting a premarket submission was more costly than either of these alternatives. Either way, the decision would be a private decision made according to their business plan. To the extent that any number of laboratories switch to any number of FDA-authorized tests, or outsource their testing needs, our estimated costs for submitting a PMA or a 510(k) would be overestimated.

##### G. Budgetary Impacts

In addition to the cost to industry of preparing and submitting various submissions to FDA, including MDRs, IDEs, Q-submissions, PMAs, PMA supplements, 510(k)s, and De Novo requests, there would be incremental review costs for FDA to review these additional submissions. FDA is excluding from new review costs LDTs that are expected to be reviewed by NYS CLEP. For LDTs approved by NYS CLEP, FDA intends to exercise enforcement discretion and generally not enforce premarket review requirements. As discussed in appendix A, we estimate that 12.1% (ranging from 6.1% to 24%) of new submissions for IVDs offered as LDTs would not experience new review costs for FDA as a result of FDA’s general enforcement discretion policy with respect to the premarket review requirements for LDTs approved by NYS CLEP.

As mentioned in section II.F.4, FDA assumes that at least 50% of the IVDs offered as LDTs being submitted for 510(k) review will be reviewed under the 510(k) Third Party Review Program. Manufacturers who submit to 3P510k Review Organizations pay the 3P510k Review Organization but do not pay FDA user fees for those submissions. Under the MDUFA V agreement, FDA is currently working to enhance the program with the objective of eliminating routine re-review by FDA of third-party reviews.

To estimate the review costs for FDA, we first use average costs per-page based on premarket submission type used in a prior estimate from the Microbiology Devices; Reclassification of Nucleic Acid-Based Systems for Mycobacterium tuberculosis complex final regulatory impact analysis (Ref. [62]). The current estimate after adjustment for inflation is $864,057 per PMA and $20,565 per 510(k) (or per De Novo).[^81] We also use labor costs from estimated FTEs for FDA review of different submission types, including MDRs, Q-submissions, IDEs, and premarket submissions.[^82] The 3-year average cost of all personnel compensation and benefits paid per FTE at FDA is $315,403 (Ref. [63]). We then multiply this by the estimated FTEs by submission type to estimate the review cost per submission. We use an average of the two estimates for the premarket review cost per submission (we only use the FTEs information for the MDR review cost per listing, IDE cost per submission, and Q-submission cost per submission).

We expect MDR review cost would occur in Stage 1. Multiplying the MDR review cost per listing by the number of MDR submissions yields a total one-time review cost of MDRs between $0.22 million and $0.88 million, with a primary estimate of $0.44 million. Multiplying the review cost per listing by the number of MDR submissions per year yields a total recurring review cost of MDRs between $0.03 million and $0.11 million, with a primary estimate of $0.06 million.

Under stage 2, we expect IDE review cost would incur in year 2. Multiplying the IDE review cost per listing by the number of IDE submissions yields a total one-time review costs of IDEs between $1.16 million and $4.65 million, with a primary estimate of $2.32 million.

The estimated recurring review cost of IDEs range from $2.81 million and $11.24 million, with a primary estimate of $5.62 million.

We expect review costs of Q-submission, PMA, 510(k), and De Novo submissions would incur under stages 4 and 5. To estimate the review cost of Q-submissions, we calculate the number of modified and new IVDs per year subject to premarket review by adding the number of PMA, 510(k), and De Novo submissions.[^83] Multiplying the review cost per Q-submission by the number of new premarket submissions per year yields a total recurring review cost of Q-submissions between $18.63 million and $74.50 million, with a primary estimate of $37.25 million.

The total recurring review cost of PMAs is estimated to range between $29.33 million to $117.30 million, with a primary estimate of $58.65 million. The total recurring review cost of PMA supplements is estimated to range between $0.71 million and $2.82 million, with a primary estimate of $1.41 million. The total recurring review cost of 510(k)s is estimated to range from $11.02 million to $44.06 million, with a primary estimate of $22.03 million. The recurring review cost of De Novo classification requests is estimated to range from $14.92 million to $59.67 million, with a primary estimate of $29.84 million.

Overall, we estimate the total one-time FDA review costs to range between $1.38 million and $5.53 million, with a primary estimate of $2.77 million. We estimate the total recurring FDA review costs to range between $77.43 million and $309.72 million, with a primary estimate of $154.86 million. See Table 38.[^84]

**Table 38. FDA Review Costs by Submission Type**
|  | Primary | Low | High |
|---|---|---|---|
| **MDR** | | | |
| FDA review costs using FTE | $5.58 | $5.58 | $5.58 |
| Affected IVDs offered as LDTs on the market | 79,114 | 39,557 | 158,227 |
| Subtotal, one-time (millions)* | $0.44 | $0.22 | $0.88 |
| MDR submissions per year for affected tests | 10,013 | 5,007 | 20,026 |
| Subtotal, recurring (millions)* | $0.06 | $0.03 | $0.11 |
| **IDE** | | | |
| FDA review costs using FTE | $21,763 | $21,763 | $21,763 |
| Affected IVDs offered as LDTs currently on the market | 107 | 53 | 214 |
| Subtotal, one-time (millions)* | $2.32 | $1.16 | $4.65 |
| New IDE submissions per year for affected tests | 258 | 129 | 516 |
| Subtotal, recurring (millions)* | $5.62 | $2.81 | $11.24 |
| **Q-submission** | | | |
| FDA review costs using FTE | $25,548 | $25,548 | $25,548 |
| Premarket submissions per year (PMAs, 510(k)s, and De Novos) for affected tests | 1,454 | 727 | 2,908 |
| Subtotal, recurring (millions)* | $37.14 | $18.57 | $74.28 |
| **PMA** | | | |
| FDA review costs using page numbers | $864,057 | $864,057 | $864,057 |
| FDA review costs using FTE | $294,429 | $294,429 | $294,429 |
| Average FDA review costs | $579,243 | $579,243 | $579,243 |
| PMA Submissions per year for affected tests | 101 | 51 | 203 |
| Subtotal, recurring (millions)* | $58.65 | $29.33 | $117.30 |
| **PMA Supplements** | | | |
| Average FDA review costs | $110,319 | $110,319 | $110,319 |
| Supplements per year for affected tests | 31 | 15 | 61 |
| Subtotal, recurring (millions)** | $1.41 | $0.71 | $2.82 |
| **510(k)** | | | |
| FDA review costs using page numbers | $20,565 | $20,565 | $20,565 |
| FDA review costs using FTE | $19,870 | $19,870 | $19,870 |
| Average FDA review costs | $20,218 | $20,218 | $20,218 |
| 510(k) submissions per year for affected tests | 1,090 | 545 | 2,179 |
| Subtotal, recurring (millions)* | $22.03 | $11.02 | $44.06 |
| **De Novo** | | | |
| FDA review costs using page numbers | $20,565 | $20,565 | $20,565 |
| FDA review costs using FTE        | $202,804 | $202,804 | $202,804 |
| **Average FDA review costs**     | **$111,685** | **$111,685** | **$111,685** |
| De Novo submissions per year for affected tests | 267      | 134      | 534      |
| **Subtotal, recurring (millions)** | **$29.84** | **$14.92** | **$59.67** |
| **Total one-time costs (millions)** | **$2.77** | **$1.38** | **$5.53** |
| **Total recurring costs (millions)** | **$154.86** | **$77.43** | **$309.72** |
||

Notes: The number of submissions per year include currently marketed tests that would be modified and new tests from both affected entities currently on the market and new entities entering the market per year. The number of premarket submissions per year for Q-submission review costs include all premarket submission types (PMA, 510(k), and De Novo). Totals may not add due to rounding.
*We calculate subtotals by multiplying average FDA review costs by the number of submissions per year.
**We multiply the average FDA review cost per PMA by the FTE weights to calculate the review cost per PMA supplement.

##### H. Transfers

With this phaseout policy, laboratories will pay fees to FDA for establishment registration, premarket submissions (where applicable), and periodic reporting for IVDs with a PMA. While these fees are paid by laboratories, they are revenue for FDA. The approach to estimating fee effects is distinct from the approaches for either benefits or costs, so they will be presented as transfers. Another perspective on the user fees is that they indicate industry bearing costs that are otherwise more simplistically presented as being experienced by FDA; hypothetically, adding the user fee estimates into the cost accounting would double-count effects on net social benefits. 

See Table 39 for the estimated transfers associated with the phaseout policy. All anticipated fees are public information published by FDA. Each laboratory is expected to pay an annual registration fee, at a cost of $7,653 per laboratory. Laboratories will also pay for submission of a report annually to FDA for each IVD that has received premarket approval, which costs $16,925 per report. Laboratories will pay $483,560 to FDA for each PMA they submit. For PMA supplements, they will pay $72,534 for each 180-day supplement, $386,848 for each panel-track supplement, and $33,849 for each real-time supplement they submit. They will pay $21,760 for each 510(k) they submit and $145,068 for each De Novo request they submit.

As mentioned in sections II.F.4 and II.G, we assume that at least 50% of the IVDs offered as LDTs being submitted for 510(k) review will be reviewed under the 510(k) Third Party Review Program. We consider the fees paying to 3P510k Review Organizations as costs to industry, as described in section II.F.4 (see Table 33 and Table 34).

Small businesses that have gross receipts or sales of $100 million or less for the most recent tax year (including their affiliates) are eligible to pay a reduced fee for certain submissions, including:

- 510(k) submissions ($5,440 per submission),
- De Novo requests ($36,267 per submission),
- PMAs ($120,890 per submission),
- PMA supplements ($18,134 for each 180-day supplement, $96,712 for each panel-track supplement, and $8,462 for each real-time supplement), and
- PMA annual reports ($4,231 per submission).

Small businesses with sales of $30 million or less are eligible to have the fee waived on their first PMA.

We assume 40 to 90 percent of the laboratories would have gross receipts or sales of $100 million or less, and we use 65 percent (average of 40 and 90 percent) to estimate the number of small business IVDs. Multiplying these fees by the relevant number of laboratories and IVDs, we expect total annual transfers to range from $34.13 million to $137.09 million, with a primary estimate of $68.54 million.

**Table 39. Transfers**
|                   | Primary | Low  | High  |
|--------------------------------------|---------|------|-------|
| **Recurring Annual**                 |         |      |       |
| Registration Annual Fee              | $7,653  | $7,653 | $7,653 |
| Entities affected                    | 1,275   | 638  | 2,551 |
| **Subtotal (millions)**              | $9.76   | $4.88 | $19.52 |
| Annual reporting on PMA              |         |      |       |
| Fee (Adjusted fee for small entities)| $16,925 | $16,925 | $16,925 |
|  ( $4,231)                           | ($0)    | ($4,231) | ($0) |
| IVDs affected, non-small*            | 35      | 18   | 71    |
| IVDs affected, small*                | 66      | 33   | 132   |
| **Subtotal (millions)**              | $0.88   | $0.30 | $1.76 |
| **One-time/Annual**                  |         |      |       |
| PMA                                  |         |      |       |
| MDUFA Review (Adjusted fee for small entities) | $483,560 | $483,560 | $483,560 |
| ($120,890)                           | ($120,890) | ($120,890) | ($120,890) |
| IVDs affected, non-small*            | 35      | 18   | 71    |
| IVDs affected, small*                | 66      | 33   | 132   |
| **Subtotal (millions)**              | $25.09  | $12.55| $50.19 |
| PMA Supplements –180-day track       |         |      |       |
| MDUFA Review (Adjusted fee for small entities) | $72,534 | $72,534 | $72,534 |
| ($18,134)                            | ($18,134) | ($18,134) | ($18,134) |
| IVDs affected, non-small*            | 2       | 1    | 4     |
| IVDs affected, small*                | 4       | 2    | 7     |
| **Subtotal (millions)**              | $0.20   | $0.10| $0.40 |
| PMA Supplements – Panel-track        |         |      |       |
| MDUFA Review (Adjusted fee for small entities) | $386,848 | $386,848 | $386,848 |
| ($96,712)                            | ($96,712) | ($96,712) | ($96,712) |
| IVDs affected, non-small*            | 1       | 0    | 1     |
| IVDs affected, small*                | 1       | 1    | 2     |
| **Subtotal (millions)**              | $0.36   | $0.18| $0.71 |
| PMA Supplements – Real-Time          |         |      |       |
| MDUFA Review (Adjusted fee for small entities) | $33,849 | $33,849 | $33,849 |
| ($8,462)                             | ($8,462) | ($8,462) | ($8,462) |
| IVDs affected, non-small*            | 5       | 2    | 10    |
| IVDs affected, small*                | 9       | 4    | 18    |
| **Subtotal (millions)**              | $0.24   | $0.12| $0.48 |
| 510(k)                               |         |      |       |
| MDUFA Review (Adjusted fee for small entities) | $21,760 | $21,760 | $21,760 |
| ($5,440)                             | ($5,440) | ($5,440) | ($5,440) |
| IVDs Affected, non-small*            | 381     | 191  | 763   |
| IVDs affected, small*                | 708     | 354  | 1,417 |
| **Subtotal (millions)**              | $12.15  | $6.08| $24.30 |
| De Novo                              |         |      |       |
| MDUFA Review (Adjusted fee for small entities) | $145,068 | $145,068 | $145,068 |
| ($36,267)                            | ($36,267) | ($36,267) | ($36,267) |
| IVDs affected, non-small* | 94 | 47 | 187 |
| IVDs affected, small* | 174 | 87 | 347 |
| Subtotal (millions) | $19.86 | $9.93 | $39.72 |
| Total Recurring Transfers (millions) |  $68.54 | $34.13 | $137.09 |
||

*The numbers of tests include currently marketed tests that would be modified in such a way as to require premarket review as well as new tests subject to premarket review from existing and new entities per year.

##### I. Stream of Benefits, Costs, and Transfers

We describe how we estimate the benefits, costs, and transfers in sections II.E, II.F, II.G and II.H. See Table 40 for a summary of the timing of expected benefits, costs, and transfers over a twenty-year time frame, in millions of 2022 U.S. dollars. Only primary estimates are presented. For each year, we present the undiscounted benefits, costs to industry, costs to FDA, and transfers.

**Table 40. Undiscounted Twenty-year Flow of Benefits, Costs, and Transfers (millions 2022 USD)**
| Year | Benefits VSLY based on 3% discounting | VSLY based on 7% discounting | Costs to Industry | Costs to FDA | Transfers |
|------|--------------------------------------|------------------------------|------------------|-------------|-----------|
| 1    | $0.00                                | $0.00                        | $101             | $0          | $0        |
| 2    | $0.00                                | $0.00                        | $105             | $8          | $10       |
| 3    | $3,111                               | $2,725                       | $332             | $54         | $23       |
| 4    | $3,151                               | $2,760                       | $1,492           | $155        | $68       |
| 5    | $4,420                               | $3,871                       | $1,492           | $155        | $69       |
| 6    | $4,613                               | $4,041                       | $1,492           | $155        | $69       |
| 7    | $4,786                               | $4,192                       | $1,492           | $155        | $69       |
| 8    | $4,939                               | $4,326                       | $1,492           | $155        | $69       |
| 9    | $5,075                               | $4,446                       | $1,492           | $155        | $69       |
| 10   | $5,197                               | $4,552                       | $1,492           | $155        | $69       |
| 11   | $5,305                               | $4,647                       | $1,492           | $155        | $69       |
| 12   | $5,403                               | $4,732                       | $1,492           | $155        | $69       |
| 13   | $5,490                               | $4,808                       | $1,492           | $155        | $69       |
| 14   | $5,568                               | $4,877                       | $1,492           | $155        | $69       |
| 15   | $5,638                               | $4,938                       | $1,492           | $155        | $69       |
| 16   | $5,701                               | $4,994                       | $1,492           | $155        | $69       |
| 17   | $5,758                               | $5,044                       | $1,492           | $155        | $69       |
| 18   | $5,810                               | $5,089                       | $1,492           | $155        | $69       |
| 19   | $5,856                               | $5,130                       | $1,492           | $155        | $69       |
| 20   | $5,899                               | $5,167                       | $1,492           | $155        | $69       |
||

Table 40 shows that for most years in the twenty-year time horizon, FDA review costs are greater than transfers. The total annualized values of FDA review costs, transfers, and the differences are presented in Table 41. These estimates are conducted using our current fiscal year 2024 Medical Device User Fee program (MDUFA) fee structure. We note that user fee payments are only intended to cover a portion of FDA review costs for premarket submissions.

Under the phaseout policy, FDA does not intend to phase out the general enforcement discretion approach for premarket review requirements for high risk IVDs offered as LDTs (Stage 4) before October 1, 2027, or for other IVDs offered as LDTs that require a premarket submission (Stage 5) before April 1, 2028. October 1, 2027, is the start of the next medical device user fee program (i.e., MDUFA VI).^87

**Table 41. Summary of FDA Review Costs and Transfers (Annualized over 20 years, in millions 2022 USD)**
| Discount rate | Primary  | Low    | High    |
|---------------|----------|--------|---------|
| FDA Review Costs 3% | $129.30 | $64.65 | $258.59 |
| FDA Review Costs 7% | $121.39 | $60.69 | $242.77 |
| Transfers 3%        | $57.50  | $28.65 | $115.05 |
| Transfers 7%        | $40.60  | $20.24 | $81.27  |
| Difference (3%)     | $71.80  | $36.00 | $143.54 |
| Difference (7%)     | $80.78  | $40.45 | $161.51 |
||

After calculating the expected benefits, costs, and transfers for each year in a twenty-year time horizon, we calculate the present and annualized values using three and seven percent discount rates. See Table 42.

**Table 42. Summary of Present and Annualized Values (in millions 2022$)**
|                                | Benefits    | Costs*     | Transfers |
|--------------------------------|-------------|------------|-----------|
| Present Value 7%               | $37,176.54  | $13,637.63 | $430.15   |
| Present Value 3%               | $64,583.82  | $20,407.00 | $855.46   |
| Annualized Value 7%            | $3,509.20   | $1,287.30  | $40.60    |
| Annualized Value 3%            | $4,341.05   | $1,371.67  | $57.50    |
||

*The estimated costs include both costs to industry and FDA.

##### J. Analysis of Regulatory Alternatives to the Final Phaseout Policy

We consider four different regulatory alternatives as described below. In our analysis of alternatives, we compare total costs, benefits, and transfers with two options that would be more stringent and one option that would be less stringent. We also consider one alternative of taking no new action. Table 43 summarizes our analysis of the alternatives of the phaseout policy.

*** BERT STOPPED HERE ***

**Table 43. Annualized Values of Regulatory Alternatives Over a 20 Year Period (in billions 2022$)**
| | Final Phaseout Policy | | Alternative 2 | | Alternative 3 | | Alternative 4 | |
|-|---------------------|--|--------------|--|--------------|--|--------------|--|
| | 3% | 7% | 3% | 7% | 3% | 7% | 3% | 7% |
| Total Benefits | $4.34 | $3.51 | $4.39 | $3.57 | $3.94 | $3.10 | $5.14 | $4.21 |
| Total Costs | $1.37 | $1.29 | $1.45 | $1.39 | $1.20 | $1.09 | $5.58 | $5.83 |
| Net Benefits | $2.97 | $2.22 | $2.94 | $2.18 | $2.74 | $2.01 | -$0.44 | -$1.62 |
| Transfers | $0.06 | $0.04 | $0.06 | $0.06 | $0.05 | $0.03 | $0.28 | $0.23 |
||

*Notes: We report primary estimates. There would be no additional costs or benefits under Alternative 1.*

1. We treat one alternative of taking no new action as the baseline for determining the costs and benefits of other alternatives. Under this option, there will be no additional costs or benefits relative to the status quo.

2. The second regulatory alternative reduces the phaseout period to three years following the publication date of the final rule:

   - **Stage 1**: beginning 1 year after the publication date of this final rule, FDA will expect compliance with MDR requirements, correction and removal reporting requirements, and QS requirements under § 820.198 (complaint files).

   - **Stage 2**: beginning 2 years after the publication date of this final rule, FDA will expect compliance with requirements not covered during other stages of the phaseout policy, including registration and listing, labeling, investigational use requirements, and QS requirements under part 820 (other than requirements under § 820.198 (complaint files), which are already addressed in stage 1).

   - **Stage 3**: beginning 3 years after the publication date of this final rule, FDA will expect compliance with premarket review requirements for high-risk IVDs and other premarket review requirements (for moderate-risk and low-risk IVDs that require premarket submissions).

   Under this alternative, we assume that one-time and recurring costs of the QS requirements would occur in year 2 and costs of the PMA, 510(k), and De Novo submissions would occur in year 3. The estimated annualized costs of this alternative would be $1.39 billion, which is $99 million higher than the estimated costs associated with the phaseout policy. The estimated annualized transfers of this alternative would be $58 million, which is $17 million higher than the estimated transfers associated with the phaseout policy. The shorter phaseout period would result in higher annualized benefits because they would begin earlier than under the phaseout policy. The estimated annualized benefits of this alternative would be $3.57 billion, which is $56 million higher than the benefits associated with the phaseout policy. However, a shorter phaseout period means that, among other things, affected laboratories, including small laboratories, would have less time to prepare and it might be less feasible for them to come into compliance.

3. The third alternative extends the phaseout period to ten years for small entities (i.e., laboratories that have their annual receipts and sales less than $100 million) and six years for other entities:

   - **Stage 1**: beginning 1 year after the publication date of this final rule, FDA will expect compliance with MDR requirements, correction and removal reporting requirements, and QS requirements under § 820.198 (complaint files).

   - **Stage 2**: beginning 4 years after the publication date of this final rule, FDA will expect compliance with requirements other than MDR, correction and removal reporting, and QS requirements under part 820 (other than requirements under § 820.198 (complaint files), which are already addressed in stage 1).

   - **Stage 3**: beginning 5 years (7 years for small laboratories) after the publication date of this final rule, FDA will expect compliance with respect to premarket review requirements for high-risk IVDs.

   - **Stage 4**: beginning 6 years (10 years for small laboratories) after the publication date of this final rule, FDA will expect compliance with respect to premarket review requirements for moderate risk and low risk IVDs that require premarket submissions.

   Compared to the final phaseout policy, having a longer phaseout period would reduce the burden on the affected laboratories by shifting costs into the future. Costs for Stage 2 under the phaseout policy (including compliance with registration and listing, labeling, and investigational use requirements) would occur in year 2, and costs for Stage 3 under the phaseout policy (relating to compliance with QS requirements) would occur in year 3, but we assume that costs for Stage 2 under this alternative (which would include the costs for Stage 2 and Stage 3 under the phaseout policy) would occur in year 4. We assume that costs for Stage 3 under this option would occur in year 5 (year 7 for small entities). We finally assume that costs for Stage 4 under this option would occur in year 6 (year 10 for small entities). The affected laboratories would thus have lower costs under Stages 2 to 4, except that the costs for Stage 1 would still occur in the first year after issuance of the final phaseout policy. The estimated annualized costs of this alternative would be approximately $1.09 billion, which is $202 million less than the estimated costs associated with the phaseout policy. Out of the estimated annualized costs, the estimated annualized costs to FDA would be approximately $96 million under this alternative, which is $25 million less than the estimated FDA review costs with the phaseout policy. In addition, the longer phaseout period for small laboratories would mean that these entities would have more time to prepare premarket submissions, potentially making it more feasible for them to come into compliance. However, this option would also reduce annualized benefits by $411 million because extending the phaseout period to six years (and ten years for small laboratories) will reduce the number of avoided harms from problematic IVDs.

4. In the fourth alternative, we assume the same phaseout policy as proposed in the preamble to the proposed rule. Under this alternative, there would be one-time costs of the QS requirements and premarket review requirements because FDA would be phasing out the general enforcement discretion approach for currently marketed IVDs offered as LDTs under stages 3 through 5.
The affected laboratories would thus have higher total costs. The estimated annualized costs of this alternative would be approximately $5.83 billion, which is $4.55 billion higher than the estimated costs associated with the final phaseout policy. This alternative would also increase annualized transfers by $194 million. Since this alternative does not consider premarket reviews by third parties or NYSDOH, the costs to FDA would be higher than the estimated costs of the final phaseout policy. We estimate that costs to FDA with this alternative would increase by $552 million, from $121 million to $673 million.

The benefits would increase because the number of affected tests under stages 3 through 5 would be higher than under the final phaseout policy. We estimate that the benefits associated with this alternative would be approximately $4.21 billion, which is $703 million higher than the estimated benefits of the final phaseout policy. Table 44 below summarizes primary estimates of the costs by stage of the phaseout policy and alternative 4. The cost reduction of the final phaseout policy compared to this alternative is primarily due changes in stages 3 through 5 due to the enforcement discretion policies. See Table 19 for the impact of the enforcement discretion policies on the benefits.

**Table 44. Costs of the Final Phaseout Policy and Alternative 4 (millions 2022$)**
| Stage  | Description                                            | Final Phaseout Policy | Alternative 4  |
|--------|--------------------------------------------------------|-----------------------|----------------|
| Stage 1 | Reading and Understanding the Phaseout Policy         | $3.91                 | $3.91          |
|        | Medical Device Reporting                               | $95.24                | $95.24         |
|        | Correction and Removal Reporting                       | $0.18                 | $0.18          |
|        | Complaint Records                                      | $2.13                 | -              |
| Stage 2 | Registration and Listing Requirements                 | $0.58                 | $0.58          |
|        | Labeling Requirements                                  | $4.55                 | $4.55          |
|        | Investigational Use Requirements                       | $25.47                | $25.47         |
| Stage 3 | Quality System Requirements                           | $26.06                | $472.19        |
| Stage 4 | Premarket Approval Application                        | $503.91               | $21,589.11     |
|        | Premarket Approval Application Supplements             | $14.76                | $289.24        |
| Stage 5 | 510(k) Submission                                      | $825.47               | $13,746.59     |
|        | De Novo Classification Request                         | $177.50               | $2,882.49      |
||

*Notes: We only report primary estimates. The estimated costs include one-time costs and recurring costs per year. The estimated costs include both costs to industry and FDA.*

##### K. Distributional Effects

Phasing out the general enforcement discretion approach for LDTs might generate benefits and costs that accrue differentially to establishments and segments of society. In this section, we discuss health equity effects for populations on which IVDs offered as LDTs are used. We address differential effects for small entities in section III of this analysis.

As described in section II.E, we expect the phaseout policy to increase the accuracy of laboratory test results, reducing the incidence of patient diagnostic error and resulting in more appropriate treatments and improved health outcomes, among other benefits. While we would not expect the benefits of the phaseout policy – in isolation – to differentially affect certain population segments, existing inequities in healthcare access might result in differential accrual of benefits across the general population. For example, there is evidence of disparities in access to tests (Ref. [64]) which might impact the patient populations that the benefits of the phaseout policy would reach. FDA also recognizes that IVDs offered as LDTs might serve communities in rural, medically underserved areas with disparities in access to diagnostic tests.

However, the benefits of test access depend on the ability of tests to work as intended, and the harms of unsafe or ineffective IVDs offered as LDTs might disproportionately occur among medically underserved patient populations that such tests might aim to reach. Without appropriate oversight, IVDs offered as LDTs might exacerbate health disparities. Research reports higher rates of inaccurate results among underrepresented patient populations, particularly racial and ethnic minorities undergoing genetic tests (Refs. [65, 66, 67, 68, 69]).

Additionally, some IVDs offered as LDTs have not been validated for use in all patient populations who may have the relevant health condition, across ages or ethnicities, meaning that it is unknown how well the test might perform across diverse patient populations expected to use the test, and tests might be less accurate in underrepresented patient populations, which could contribute to health disparities (Ref. [70]).

The role of IVDs offered as LDTs in either ameliorating or exacerbating existing health inequity ultimately depends on the safety and effectiveness of IVDs offered as LDTs, which the phaseout policy is intended to help assure. By increasing its oversight, FDA might better prevent and mitigate harms disproportionately realized among underrepresented, medically underserved populations. As such, the benefits of phasing out the general enforcement discretion approach for LDTs might differentially reach these populations.

When IVDs are subject to increased FDA oversight, FDA will help ensure that information is available pertaining to device safety and effectiveness for specific demographic characteristics if performance differs within the target population, through the enforcement of applicable labeling requirements. In addition, when FDA conducts premarket review of a device, FDA may ask that sponsors provide data for different intended patient populations. With new FDA authorities under the Food and Drug Omnibus Reform Act of 2022 (FDORA), sponsors will generally be required to submit diversity action plans to FDA, including the sponsor’s goals for enrollment in device clinical studies, to help improve the generalizability of the results to the intended use population. In contrast, with limited oversight over these IVDs, FDA does not know whether validation studies for these IVDs include diverse patient populations. FDA believes increased oversight for these IVDs will help ensure adequate representation of the intended use population in validation studies and transparency regarding potential differential performance, helping to advance health equity.

Nonetheless, while phasing out the general enforcement discretion approach for LDTs might help to advance health equity, we have no specific data showing that increased FDA oversight of IVDs offered as LDTs will necessarily reduce health disparities.

As described in section II.F.6, pass-through of costs to provide IVDs offered as LDTs might in turn create additional costs to society. If laboratories pass through the cost of compliance to the costs of IVDs offered as LDTs, test frequency might decrease for areas that rely on IVDs offered as LDTs for easy, rapid access to tests. If laboratories or healthcare facilities respond to increased compliance costs by increasing the prices of IVDs offered as LDTs or reducing the availability of IVDs offered as LDTs, there might thus be an increase in health inequity. Vulnerable populations that rely on IVDs offered as LDTs for diagnosis might have less access to diagnostic tests in general after the implementation of the phaseout policy.

However, in the absence of assurances about the safety and effectiveness of these tests, the value of access is uncertain. We further note that in the event any currently marketed tests for underserved populations are withdrawn from the market due to their inability to meet regulatory requirements, other manufacturers may fill the need with appropriately designed and validated tests.

We do not expect the phaseout policy to result in an increase in health inequity in isolation. Though we do have evidence of existing health inequities in diagnostic tests and clinical trials across sociodemographic populations, we lack the evidence to quantify the effect of the phaseout policy on these existing health inequities, and thus cannot determine whether the phaseout policy will ameliorate or exacerbate health inequity. By increasing oversight over IVDs offered as LDTs, FDA may better prevent and mitigate harm to patients that might result from inaccurate and unreliable tests, including patients in underserved populations.

##### L. International Effects

While the phaseout policy will generate benefits that accrue to the domestic population, some laboratories that are located outside the United States would be expected to comply with applicable device requirements, as a result of the phaseout of the general enforcement discretion approach for LDTs, if those laboratories offer IVDs as LDTs to patients within the United States. This section estimates the cost of compliance for international laboratories. These costs are not included in section II.F, which only assesses domestic costs.

As of January 2024, there are 74 international laboratories certified under CLIA to perform non-waived testing. Based on available information and professional judgment, we assume that 100% of CLIA-certified international laboratories are performing high complexity testing and have IVDs offered as LDTs. While our historical experience indicates that these laboratories likely offer a smaller number of IVDs offered as LDTs, as they are typically offering more specialized tests, for the cost estimates, we use the same assumption described in section II.D that each laboratory would have 67 IVDs offered as LDTs, and thus we expect 496 (= 74 x 67) international IVDs offered as LDTs to be affected by the phaseout policy. We also assume 592 (74 x (6+2)) modified (an additional 2 annually) and new (an additional 6 annually) international IVDs offered as LDTs to be affected annually, consistent with assumptions in section II.D.

We also adjust wages to reflect the fact that international laboratories may not offer the same wages as those in the United States. Specifically, we create a list of the unique countries that appear in our data on the 74 international laboratories, then search the National Bureau of Economic Research (NBER) Occupational Wages around the World (OWW) database for wage information for the relevant countries. The most recent year with complete data is 2007. We observe the average hourly wage rate across all sectors for the relevant countries in U.S. dollars, then divide by the same measure for U.S. wages to get a relative measure of wages as percent deviation from the U.S. hourly wage rate for the same period. We then take the average percent deviation across the relevant countries and find that wages for the relevant international countries are 73% that of U.S. wages for the same time period. We therefore adjust the wages we use in the domestic cost analysis by 0.73 to assess international costs.

Aside from coverage and wage rates, the costs for international laboratories are calculated using the exact same methods as in section II.F. Because there are significantly fewer laboratories and tests, and wages are slightly lower, international costs are much lower than domestic costs of compliance. See Table 45 for a summary of international costs, organized by stage and part of the phaseout policy.

**Table 45. International Costs**
| | Primary | Low | High |
|--|---------|-----|------|
| One-time |  |  |  |
| Stage 1 | | | |
| Reading and Understanding the Rule | $165,541 | $74,494 | $279,351 |
| Medical Device Reporting | $612,171 | $498,064 | $752,796 |
| Correction and Removal Reporting | $1,850 | $1,850 | $1,850 |
| Complaint Records | $96,644 | $55,225 | $138,062 |
| Stage 2 | | | |
| Registration and Listing Requirements | $18,669 | $18,669 | $18,669 |
| Labeling Requirements | $192,916 | $49,785 | $336,048 |
| **Investigational Use Requirements** | $466,943   | $466,943  | $466,943  |
| Stage 3 | | | |
| Quality System Requirements  | $96,644    | $55,225   | $138,062  |
| **Total One-time Costs**             | **$1,651,379** | **$1,220,255** | **$2,131,781** |
| **Recurring Annual**                 |            |           |           |
| Stage 1 | | | |
| Medical Device Reporting     | $3,707,098 | $3,707,090 | $3,707,106 |
| Correction and Removal Reporting     | $6,845     | $6,845    | $6,845    |
| Re-registration                      | $6,223     | $6,223    | $6,223    |
| Stage 2 | | | |
| Investigational Use          | $1,129,027 | $1,129,027 | $1,129,027 |
| Premarket Approval Application       | $59,172,917 | $56,088,281 | $117,781,004 |
| Stage 4 | | | |
| Premarket Approval Application Supplements | $1,733,605 | $1,640,318 | $3,506,047 |
| Stage 5 | | | |
| 510(k) Submission            | $104,382,781 | $96,222,502 | $112,543,060 |
| De Novo Classification Request       | $20,842,241 | $11,086,707 | $20,942,297 |
| **Total Recurring Costs**            | **$190,980,737** | **$169,886,993** | **$259,621,610** |
||

See Table 46 for a summary of the expected timing and annualized value of international costs. At a three percent discount rate, we expect the annualized value of international costs to range from $140.29 million to $214.98 million, with a primary estimate of $157.50 million. At a seven percent discount rate, we expect the annualized value of international costs to range from $131.23 million to $201.26 million, with a primary estimate of $147.26 million.

**Table 46. Twenty-Year Timing of International Costs (millions 2022$)**
|                          | Primary | Low     | High    |
|--------------------------|---------|---------|---------|
| Year 1                   | $4.59   | $4.34   | $4.89   |
| Year 2                   | $5.53   | $5.38   | $5.67   |
| Year 3                   | $35.40  | $33.77  | $65.63  |
| Year 4-20 (costs for each year) | $190.98 | $169.89 | $259.62 |
| **Total Costs of the Phaseout Policy** | **$3,292.19** | **$2,931.58** | **$4,489.76** |
| Present Value of Total Costs (3%) | $2,343.16 | $2,087.14 | $3,198.30 |
| Present Value of Total Costs (7%) | $1,560.07 | $1,390.28 | $2,132.20 |
| **Annualized Value of Costs (3%)** | **$157.50** | **$140.29** | **$214.98** |
| **Annualized Value of Costs (7%)** | **$147.26** | **$131.23** | **$201.26** |
||

#### III. Final Small Entity Analysis

The Regulatory Flexibility Act requires agencies to analyze regulatory options that would minimize any significant impact of a rule on small entities. Because most facilities that will be affected by the phaseout of the general enforcement discretion approach for LDTs are defined as small businesses and the phaseout policy is likely to impose a substantial burden on the affected small entities, we find that the phaseout policy will have a significant economic impact on a substantial number of small entities. This analysis, as well as other sections in this document and the final rule, serves as the Final Regulatory Flexibility Analysis, as required under the Regulatory Flexibility Act.

##### A. Description and Number of Affected Small Entities

We used detailed data from 2017 Statistics of U.S. Businesses on U.S. 6-digit NAICS detailed employment sizes and revenues to analyze the potential impacts of the phaseout policy on small entities. The Small Business Administration (SBA) considers Medical Laboratories (NAICS code 621511) to be small if their annual receipts are less than $41.5 million. Since not all laboratories in this NAICS code offer IVDs as LDTs, we use the number of affected laboratories and distribute them proportionally across the revenue distribution from the Economic Census to estimate breakdown of the laboratories by revenue size (see Table 47). Of the 1,181 laboratories, 1,085 laboratories (those with less than $41.5 million in annual receipts), or 92 percent of the total, would be small according to the 2023 SBA size standard. We estimate that small businesses also manufacture 22% of IVDs offered as LDTs currently on the market. We provide more detail on these estimates in Appendix B.

**Table 47. Distribution of Revenues for Laboratories Offering IVDs as LDTs**
| Receipts Size ($1,000) | Number of Laboratories Under NAICS Code 621511 | Number of Laboratories Offering IVDs as LDTs | Number of IVDs offered as LDTs |
|------------------------|-----------------------------------------------|---------------------------------------------|----------------------------------|
| < $150                 | 438                                           | 166                                         | 56                               |
| $151 - $999            | 933                                           | 327                                         | 625                              |
| $1000 - $1,999         | 413                                           | 145                                         | 754                              |
| $2,000 - $3,999        | 481                                           | 169                                         | 1,948                            |
| $4,000 - $5,999        | 343                                           | 120                                         | 3,061                            |
| $6,000 - $9,999        | 146                                           | 51                                          | 2,150                            |
| $10,000 - $14,999      | 77                                            | 27                                          | 1,489                            |
| $15,000 - $19,999      | 115                                           | 40                                          | 3,114                            |
| $20,000 - $24,999      | 79                                            | 28                                          | 2,920                            |
| $25,000 - $29,999      | 21                                            | 7                                           | 951                              |
| $30,000 - $39,999      | 43                                            | 15                                          | 2,270                            |
| $40,000 - $49,999      | 15                                            | 5                                           | 1,151                            |
| $50,000 - $99,999      | 67                                            | 24                                          | 5,475                            |
| $100,000 +             | 194                                           | 68                                          | 63,163                           |
| **Total**              | 3,365                                         | 1,193                                       | 89,127                           |
| < $41.5 million        | 3,091                                         | 1,097                                       | 19,510                           |
| **Percent Small**      | **92%**                                       | **92%**                                     | **22%**                          |
||

##### B. Description of the Potential Impacts of the Phaseout Policy on Small Entities

We compiled the costs and transfers associated with the phaseout policy and compared them to the estimated share of annual receipts of the laboratories offering IVDs as LDTs. In Table 48, we estimate the total annualized costs per entity at a 7 percent discount rate over 20 years and the costs as a percent of revenue by receipts size. The estimated annualized cost per small entity ranges from $4,395 to $3,045,766 per laboratory, depending on its size.
As shown in Table 48, the annualized costs per small entity averages $232,618 represent 5.8 percent of receipts for the small laboratories (with annual receipts of less than $41,500,000). Because we don’t know how costs would be distributed across entities, we estimate average costs per laboratory by their receipt size categories using the assumption that costs are distributed proportionally to receipts and for this reason costs as a percent of receipts appear to be constant across all receipt size categories.

The extent in which smaller laboratories may be disproportionately impacted by the phaseout of the general enforcement discretion approach for LDTs, is dependent on the number of IVD's offered as LDTs per lab. FDA anticipates that the enforcement discretion policies discussed in the preamble of the final rule will moderate these concerns and help to avoid complete disruption to the test market. As noted in Appendix B-Table 8, the average costs per LDT are smallest for stages 1 through 3 of the phaseout policy representing 10% of costs and up to 59% of affected tests, whereas average costs per LDT for stages 4 and 5 represent 90% of costs affecting 3% of tests. The percentage of tests that may experience costs under stages 4 and 5 will increase as new laboratories and tests enter the market during and after stages 4 and 5, as they will fall within the enforcement discretion policy for currently marketed tests. However, they may still fall within the scope of other enforcement discretion policies described in the preamble to the final rule, including those for unmet needs and LDTs approved by NYS CLEP. However, in the event that a new lab does not fall within the scope of other enforcement discretion policies, costs under stages 4 and 5 could present as a potential barrier to entry in the LDT market for new laboratories. In Table B7 of Appendix B, total costs and transfers of the phaseout policy are estimated to be on average anywhere between about 2.54, 5.8 and 16 percent for all entities. We do not have the information about labs to determine how the average estimates are distributed among the firms according to their size categories. Depending on profit margins with respect to revenue, the costs of this rule may be prohibitive for some small labs, making it likely for some small entities in this size category to exit the market, reduce operations, sell the business, be subject to acquisitions by larger firms or not enter the market. If profit margins were too small for many small firms considering the costs, it is possible that selling to large entities, would further cause industry consolidation and contribute to the growth of monopolies in the industry which would hinder competition.

While we do not have the data on profit margins to properly estimate the number of labs that would be adversely impacted by this rule, we estimate that small laboratories make fewer IVDs offered as LDTs than large firms. We estimate that small labs make up 92 percent of all labs, and that they also hold a 24 percent share of IVDs offered as LDTs. With the low number of IVDs offered as LDTs per small lab, it is more likely that the percent of costs over receipts per lab would be closer to our low average estimate 2.5 percent. Furthermore, for the final rule, as explained in comments 17 and 18, FDA intends to exercise enforcement discretion with respect to premarket review and QS requirements (except for applicable requirements under 21 CFR 820, subpart M (Records)) for currently marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the final rule and that are not modified, or that are modified as described in the preamble and for LDTs developed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system (as described in section V.B.3 of the preamble), we do not expect significant market concentration or market exit to result from the phaseout policy.

Small businesses that have gross receipts or sales of $100 million or less for the most recent tax year (including their affiliates) are eligible to pay a reduced fee for certain submissions, including 510(k) submissions, De Novo classification requests, PMAs, and PMA annual reports. The estimated average recurring transfer for small businesses is $4,069. As seen in Table 48, the percentage of receipts that are additional transfers associated with the phaseout policy are estimated to be 0.10 percent for 166 laboratories (15 percent of the small entities) with their annual receipts less than $150,000.

**Table 48. Small Business Costs and Transfers as a Percentage of Receipts**
| Receipts Size ($1,000) | Labs | Average Receipts | Total Costs per Lab | Costs as a % of Receipts | Total Transfers per Lab | Transfers as a % of Receipts |
|------------------------|------|------------------|---------------------|--------------------------|------------------------|-----------------------------|
| < $150                 | 166  | $75,755          | $4,395              | 5.8%                     | $76                    | 0.10%                       |
| $151 - $999            | 327  | $430,532         | $24,978             | 5.8%                     | $434                   | 0.10%                       |
| $1000 - $1,999         | 145  | $1,172,533       | $68,026             | 5.8%                     | $1,181                 | 0.10%                       |
| $2,000 - $3,999        | 169  | $2,601,807       | $150,947            | 5.8%                     | $261                   | 0.10%                       |
| $4,000 - $5,999        | 120  | $5,733,410       | $332,631            | 5.8%                     | $5,775                 | 0.10%                       |
| $6,000 - $9,999        | 51   | $9,459,407       | $548,799            | 5.8%                     | $9,528                 | 0.10%                       |
| $10,000 - $14,999      | 27   | $12,425,958      | $720,908            | 5.8%                     | $12,516                | 0.10%                       |
| $15,000 - $19,999      | 40   | $17,395,652      | $1,009,231          | 5.8%                     | $17,521                | 0.10%                       |
| $20,000 - $24,999   | 28   | $23,750,766      | $1,377,930    | 5.8%                     | $23,922           | 0.10%                         |
| $25,000 - $29,999   | 7    | $29,082,108      | $1,687,235    | 5.8%                     | $29,292           | 0.10%                         |
| $30,000 - $39,999   | 15   | $33,924,383      | $1,968,166    | 5.8%                     | $34,169           | 0.10%                         |
| $40,000 - $49,999   | 5    | $49,317,674      | $2,861,227    | 5.8%                     | $92,157           | 0.19%                         |
| $50,000 - $99,999   | 24   | $52,498,495      | $3,045,766    | 5.8%                     | $106,083          | 0.20%                         |
| $100,000 +          | 68   | $209,184,656     | $12,136,110   | 5.8%                     | $422,695          | 0.20%                         |
| Total               | 1,193| $16,841,846      | $977,101      | 5.8%                     | $30,296           | 0.18%                         |
| < $41.5 M           | 1,097| $4,009,525       | $232,618      | 5.8%                     | $4,069            | 0.10%                         |
| Percent small       | 92%  | 24%              | 24%           | 100%                     | 13%               | 56.42%                        |
||

##### C. Alternatives to Minimize the Burden on Small Entities

Regulatory alternative 3, described in section II.J, would reduce costs for all laboratories by extending the phaseout period to ten years for small entities and six years for other entities. Below we show how the reduction in cost under the alternative would reduce the cost on small laboratories, if it were implemented.

The alternative that could reduce the impact to small entities would be an extended phaseout policy from 4 years to 10 years for small laboratories as discussed in section II.J.3 ("third alternative"). Compared with the final phaseout policy, small laboratories would have lower one-time and recurring costs for Stage 2 of the third alternative because they generally would have an additional one to two years before FDA would expect compliance with these requirements (e.g., labeling, registration and listing, investigational use, and QS requirements). There would also be an additional 3.5 years for the compliance expectations for PMA requirements and 6 years for the compliance expectations for 510(k) and De Novo requirements. The costs associated with Stage 1 would be unimpacted by the extended phaseout policy as the costs would still occur in the first year after issuance of the final phaseout policy.

We estimate this option would reduce total costs by $690 to $478,302 per small entity. For all laboratories, total recurring costs are estimated to be 4.9 percent of their average receipts. This alternative would also reduce transfers for all laboratories offering IVDs as LDTs from an average of $4,069 to $2,055 per entity for laboratories with their annual receipts below $41.5 million, which is $2,014 less than the estimated transfers of the phaseout policy. For the smallest laboratories (with annual receipts lower than $150,000), total transfers would be 0.05% percent of receipts. See Table 49.

**Table 49. Small Business Costs and Transfers as a Percentage of Receipts under Regulatory Alternative 3**
| Receipts Size ($1,000) | Labs | Average Receipts | Costs Per Lab (7%) | Costs as a % of Receipts | Transfers Per Lab (7%) | Transfers as a % of Receipts |
|------------------------|------|------------------|--------------------|--------------------------|------------------------|------------------------------|
| < $150                 | 166  | $75,755          | $3,705             | 4.9%                     | $38                    | 0.05%                        |
| $151 - $999            | 327  | $430,532         | $21,055            | 4.9%                     | $217                   | 0.05%                        |
| $1000 - $1,999         | 145  | $1,172,533       | $57,343            | 4.9%                     | $591                   | 0.05%                        |
| $2,000 - $3,999        | 169  | $2,601,807       | $127,243           | 4.9%                     | $1,312                 | 0.05%                        |
| $4,000 - $5,999        | 120  | $5,733,410       | $280,395           | 4.9%                     | $2,890                 | 0.05%                        |
| $6,000 - $9,999        | 51   | $9,459,407       | $462,617           | 4.9%                     | $4,769                 | 0.05%                        |
| $10,000 - $14,999      | 27   | $12,425,958      | $607,697           | 4.9%                     | $6,264                 | 0.05%                        |
| $15,000 - $19,999      | 40   | $17,395,652      | $850,743           | 4.9%                     | $8,769                 | 0.05%                        |
| $20,000 - $24,999      | 28   | $23,750,766      | $1,161,542         | 4.9%                     | $11,973                | 0.05%                        |
| $25,000 - $29,999      | 7    | $29,082,108      | $1,422,274         | 4.9%                     | $14,660                | 0.05%                        |
| $30,000 - $39,999      | 15   | $33,924,383      | $1,659,088         | 4.9%                     | $17,101                | 0.05%                        |
| $40,000 - $49,999      | 5    | $49,317,674      | $2,411,904         | 4.9%                     | $71,741                | 0.15%                        |
| $50,000 - $99,999      | 24   | $52,498,495      | $2,567,464         | 4.9%                     | $85,176                | 0.16%                        |
| $100,000 +             | 68   | $209,184,656     | $10,230,325        | 4.9%                     | $339,391               | 0.16%                        |
| Total                  | 1,193| $16,841,846      | $823,658           | 68.5%                    | $23,202                | 0.14%                        |
| < $41.5 M              | 1,097| $4,009,525       | $196,088           | 4.9%                     | $2,055                 | 0.05%                        |
||

#### IV. References

[1] "Preliminary Regulatory Impact Analysis; Initial Regulatory Flexibility Analysis; Unfunded Mandates Reform Act Analysis," 2023. [Online]. Available: https://www.fda.gov/about-fda/reports/economic-impact-analyses-fda-regulations.

[2] J. Rychert, R. Schmidt and J. Genzen, "Laboratory-Developed Tests Account for a Small Minority of Tests Ordered in an Academic Hospital System," *American journal of clinical pathology*, vol. 160, no. 3, p. 297–302, 2023.

[3] J. Hooftman, A. Dijkstra, I. Suurmeijer, A. van der Bij, E. Paap and L. Zwaan, "Common contributing factors of diagnostic error: A retrospective analysis of 109 serious adverse event reports from Dutch hospitals," *BMJ Quality & Safety*, 2023.

[4] L. Zwaan, M. de Bruijne, C. Wagner, A. Thijs, M. Smits, G. van der Wal and D. Timmermans, "Patient record review of the incidence, consequences, and causes of diagnostic adverse events," *Archives of internal medicine*, vol. 170, no. 12, pp. 1015-1021, 2010.

[5] C. van Moll, T. Egberts, C. Wagner, L. Zwaan and M. Berg, "The Nature, Causes, and Clinical Impact of Errors in the Clinical Laboratory Testing Process Leading to Diagnostic Error: A Voluntary Incident Report Analysis," *Journal of Patient Safety*, vol. 19, no. 8, pp. 573-579, 2023.

[6] A. K. Stroobants, H. M. J. Goldschmidt and M. Plebani, "Error budget calculations in laboratory medicine: linking the concepts of biological variation and allowable medical errors," *Clinica chimica acta*, vol. 333, no. 2, pp. 169-176, 2003.

[7] G. D. Schiff and L. L. Leape, "Commentary: how can we make diagnosis safer?," *Academic Medicine*, vol. 87, no. 2, pp. 135-138, 2012.

[8] G. D. Schiff, S. Kim, R. Abrams, K. Cosby, B. Lambert, A. Elstein, S. Hasler, N. Krosnjar, R. Odwazny, M. Wisniewski and R. McNutt, "Diagnosing diagnosis errors: lessons from a multi-institutional collaborative project," *Advances in patient safety: from research to implementation*, vol. 2: concepts and methodology, 2005.

[9] ECRI, "Deep Dive: Laboratory Testing," [Online]. Available: https://web.archive.org/web/20140714191534/https://www.ecri.org/Press/Pages/ECRI-Institute-PSO-Study-on-Lab-Test-related-Safety-Events-Finds-Most-Occurred-Outside-the-Laboratory.aspx.

[10] M. L. Graber, N. Franklin and R. Gordon, "Diagnostic error in internal medicine," *Archives of internal medicine*, vol. 165, no. 13, pp. 1493-1499, 2005.
[11] H. Singh, A. Meyer and E. Thomas, "The frequency of diagnostic errors in outpatient care: estimations from three large observational studies involving US adult populations," *BMJ Quality and Safety*, vol. 23, no. 9, pp. 727-731, 2014.

[12] New York State Department of Health, "Public Comment: Re: Food and Drug Administration Docket No. FDA-2023-N-2177 (88 FR 68006)," 22 November 2023. [Online]. Available: https://www.regulations.gov/comment/FDA-2023-N-2177-4963. [Accessed 22 December 2023].

[13] J. C. Hopewell, S. Parish, R. Clarke, J. Armitage, L. Bowman, J. Hager, M. Lathrop and R. Collins, "No Impact of KIF6 Genotype on Vascular Risk and Statin Response Among 18,348 Randomized Patients in the Heart Protection Study," *Journal of the American College of Cardiology*, vol. 57, no. 20, pp. 2000-2007, 2011.

[14] T. Assimes , H. Hólm , S. Kathiresan, M. P. Reilly, G. Thorleifsson, B. F. Voight, J. Erdmann, C. Willenborg, D. Vaidya, C. Xie, C. C. Patterson, T. M. Morgan, M. S. Burnett, M. Li, M. A. Hlatky, J. W. Knowles, J. R. Thompson, D. Absher, C. Iribarren, A. Go, S. P. Fortmann, S. Sidney and N. T. Risch, "Lack of Association Between the Trp719Arg Polymorphism in Kinesin-Like Protein-6 and Coronary Artery Disease in 19 Case-Control Studies," *J American College of Cardiology*, pp. 1552-1563, 2010.

[15] S. Kliff and A. Bhatia, "When They Warn of Rare Disorders, These Prenatal Tests Are Usually Wrong," *New York Times*, 1 January 2022.

[16] Centers for Disease Control and Prevention, "Division of Laboratory Systems (DLS): Strengthening Clinical Laboratories," 15 November 2018. [Online]. Available: https://www.cdc.gov/csels/dls/strengthening-clinical-labs.html. [Accessed 21 August 2023].

[17] Grand View Research, "Laboratory Developed Tests Market Size, Share & Trends Analysis Report By Technology (Immunoassay, Molecular Diagnostics), By Application (Oncology, Nutritional & Metabolic Disease), By Region, And Segment Forecasts, 2023 - 2030," Grand View Research, San Francisco, 2023.

[18] Centers for Medicare & Medicaid Services, "Clinical Laboratory Improvement Amendments (CLIA) Post-Public Health Emergency (PHE) Guidance," 11 May 2023. [Online]. Available: https://www.cms.gov/files/document/qso-23-15-clia.pdf. [Accessed 21 August 2023].

[19] Centers for Medicare and Medicaid Services, "CLIA Overview," 22 October 2013. [Online]. Available: https://www.cms.gov/regulations-and-guidance/legislation/clia/downloads/ldt-and-clia_faqs.pdf. [Accessed 18 January 2024].

[20] U.S. Food and Drug Administration, "Memorandum to File from Brittany Schuck Re: Examples of In Vitro Diagnostic Products (IVDs) Offered as Laboratory Developed Tests (LDTs) that Raise Public Health Concerns," 2023. [Online]. Available: https://www.regulations.gov/document/FDA-2023-N-2177-0076.

[21] J. D. Pfeifer, R. Loberg, C. Lofton-Day and B. A. Zehnbauer, "Reference Samples to Compare Next-Generation Sequencing Test Performance for Oncology Therapeutics and Diagnostics," *American Journal of Clinical Pathology*, pp. 157(4):628-638, 2022.

[22] U.S. Food and Drug Administration, "Letter to deCODEme Complete Scan," 2010. [Online]. Available: https://www.fda.gov/media/79216/download.

[23] U.S. Food and Drug Administration, "Letter to 23andMe Personal Genome Service," 2010. [Online]. Available: http://web.archive.org/web/20191214010336/https://www.fda.gov/media/79205/download.

[24] American Association for Cancer Research, "Re: Request for Information Regarding 21st Cures – Request for Feedback: A Modernized Framework for Innovative Diagnostic Tests," 5 January 2015. [Online]. Available: https://www.aacr.org/wp-content/uploads/2019/11/AACR-Comments-Modernized-Framework-Diagnostics-Jan-5-2015.pdf. [Accessed 24 August 2023].

[25] U.S. Government Publishing Office, "Laboratory Testing in the Era of Precision Medicine," 20 September 2016. [Online]. Available: https://www.govinfo.gov/content/pkg/CHRG-114shrg21906/pdf/CHRG-114shrg21906.pdf. [Accessed 24 August 2023].

[26] U.S. Food and Drug Administration, "Framework for Regulatory Oversight of Laboratory Developed Tests (LDTs); Draft Guidance for Industry, Food and Drug Administration Staff, and Clinical Laboratories," 3 October 2014. [Online]. Available: https://www.fda.gov/media/89841/download. [Accessed 23 August 2023].

[27] Pew Research Center, "The Role of Lab-Developed Tests in the In Vitro Diagnostics Market.," 22 10 2021. [Online]. Available: https://www.pewtrusts.org/en/research-and-analysis/reports/2021/10/the-role-of-lab-developed-tests-in-the-in-vitro-diagnostics-market.

[28] U.S. Food and Drug Administration, "Memo to File: Information from the New York State Department of Health (NYSDOH) for the Preliminary Regulatory Impact Analysis for the Laboratory Developed Tests Proposed Rule and Certain Related Analysis," 2023.

[29] Department of Health, Wadsworth Center, "Clinical Laboratory Evaluation Program: Clinical Laboratories: Obtain a Permit: Test Approval," [Online]. Available: https://www.wadsworth.org/regulatory/clep/clinical-labs/obtain-permit/test-approval. [Accessed 21 August 2023].

[30] Congressional Budget Office, "Federal Policies in Response to Declining Entrepreneurship," 29 December 2020. [Online]. Available: https://www.cbo.gov/publication/56945. [Accessed 21 August 2023].

[31] AstuteAnalytica India Pvt. Ltd., "United States Clinical Laboratory Services Market to Generate Revenue of US$ 125.6 Billion by 2030: Astute Analytica," 13 March 2023. [Online]. Available: https://www.globenewswire.com/news-release/2023/03/13/2625730/0/en/United-States-Clinical-Laboratory-Services-Market-to-Generate-Revenue-of-US-125-6-Billion-by-2030-Astute-Analytica.html. [Accessed 21 August 2023].

[32] U.S. Food and Drug Administration, "Performance Report to Congress: Medical Device User Fee Amendments, FY 2022, page 21," 2022. [Online]. Available: www.fda.gov/media/167825/download. [Accessed 21 August 2023].

[33] D. E. Newman-Toker, Z. Wang, Y. Zhu, N. Nassey, A. S. Saber Tehrani, A. C. Schaffer, C. W. Yu-Moe, G. D. Clemens, M. Fanai and D. Siegal, "Rate of diagnostic errors and serious misdiagnosis-related harms for major vascular events, infections, and cancers: toward a national incidence estimate using the “Big Three”," *Diagnosis*, vol. 8, no. 1, pp. 67-84, 2021.

[34] D. Newman-Toker, N. Nassey, A. Schaffer, C. W. Yu-Moe, G. D. Clemens, Z. Wang, Y. Zhu, A. S. S. Tehrani, M. Fanai, A. Hassoon and D. Siegal, "Burden of serious harms from diagnostic error in the USA," *BMJ Quality & Safety*, 2023.

[35] D. Newman-Toker, A. Schaffer, W. Yu-Moe, N. Nassey, A. S. Tehrani, G. Clemens, Z. Wang, Y. Zhu, M. Fanai and D. Siegal, "Serious misdiagnosis-related harms in malpractice claims: The "Big Three" - vascular events, infections, and cancers," *Diagnosis*, vol. 6, no. 3, pp. 227-240, 2019.

[36] U.-P. Rohr, C. Binder, T. Dieterle, F. Giusti, C. G. M. Messina, E. Teorien, H. Moch and H. H. Schafer, "The Value of In Vitro Diagnostic Testing in Medical Practice: A Status Report," *PLOS One*, vol. 11, no. 3, pp. 1-16, 2016.

[37] M. M. Ward, F. Ullrich, K. Matthews, G. Rushton, M. A. Goldstein, D. F. Bajorin, A. Hanley and C. F. Lynch, "Who Does Not Receive Treatment for Cancer?," *Journal of Oncology Practice*, vol. 9, no. 1, pp. 20-6, 2013.

[38] National Cancer Institute, "Cancer Stat Facts: Common Cancer Sites," 2023. [Online]. Available: https://seer.cancer.gov/statfacts/html/common.html. [Accessed 21 August 2023].

[39] National Cancer Institute, "Age and Cancer Risk," National Institutes of Health, 5 March 2021. [Online]. Available: https://www.cancer.gov/about-cancer/causes-prevention/risk/age. [Accessed 18 August 2023].

[40] P. A. Johnstone, M. S. Norton and R. H. Riffenburgh, "Survival of patients with untreated breast cancer," *Journal of Surgical Oncology*, vol. 73, no. 4, pp. 273-277, 2000.

[41] H. Wao, R. Mhaskar, A. Kumar, B. Miladinovic and B. Djulbegovic, "Survival of patients with non-small cell lung cancer without treatment: a systematic review and meta-analysis," *Systematic Reviews*, vol. 2, no. 10, 2013.

[42] U.S. Department of Health and Human Services, "Guidelines for Regulatory Impact Analysis," 12 Jan 2017. [Online]. Available: https://aspe.hhs.gov/reports/guidelines-regulatory-impact-analysis.

[43] American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee, "Heart Disease and Stroke Statistics-2023 Update: A Report From the American Heart Association," *American Heart Association*, 2023.

[44] Centers for Disease Control and Prevention, "Heart Disease Facts," 15 May 2023. [Online]. Available: https://www.cdc.gov/heartdisease/facts.htm#print. [Accessed 27 November 2023].

[45] Z. Raisi-Estabragh, O. Kobo, A. Elbadawi, P. Velagapudi, G. Sharma, R. Bullock-Palmer, S. Petersen, L. Mehta, W. Ullah, A. Roguin, L. Sun and M. Mamas, "Differential Patterns and Outcomes of 20.6 Million Cardiovascular Emergency Department Encounters for Men and Women in the United States," *Journal of the American Heart Association*, vol. 11, no. 19, p. e026432, 2022.

[46] J. Conly, F. Clement, M. Tonelli, B. Hemmlgarn, S. Klarenbach, A. Lloyd and F. McAlister, "Cost-effectiveness of the use of low- and high-potency statins in people at low cardiovascular risk.," *CMAJ*, 2011.

[47] U.S. Department of Health and Human Services, "Contractor Project Report: International Prescription Drug Price Comparisons: Estimates Using 2022 Data," U.S. Department of Health and Human Services, 2024.

[48] Centers for Disease Control and Prevention, "National Notifiable Diseases Surveillance System. 2019 annual tables of infectious disease data," 2021. [Online]. Available: https://www.cdc.gov/nchs/hus/contents2020-2021.htm#Table-IDNotifiable. [Accessed 31 January 2024].

[49] Y. Li, S. You, K. Lee, R. Yeasoubi, K. Hsu, T. Gift, H. Chesson, A. Berruti, J. Salomon and M. Ronn, "The Estimated Lifetime Quality-Adjusted Life-Years Lost Due to Chlamydia, Gonorrhea, and Trichomoniasis in the United States in 2018," *The Journal of Infectious Diseases*, vol. 227, no. 8, pp. 1007-1018, 2023.

[50] A. Malek, C.-C. Chang, D. Clark and R. Cook, "Delay in Seeking Care for Sexually Transmitted Diseases in Young Men and Women Attending a Public STD Clinic," *Open AIDS Journal*, vol. 14, no. 7, pp. 7-13, 2013.

[51] Blue Cross and Blue Shield of Minnesota, "Blue Cross and Blue Shield of Minnesota Files Lawsuit Against GS Labs," 1 March 2022. [Online]. Available: https://www.bluecrossmn.com/about-us/newsroom/news-releases/blue-cross-and-blue-shield-minnesota-files-lawsuit-against-gs-labs. [Accessed 21 August 2023].

[52] APM Reports, "GS Labs' responses to questions from APM Reports," APM Reports, 28 7 2022. [Online]. Available: https://www.apmreports.org/story/2022/07/28/gs-labs-responses-to-questions-from-apm-reports. [Accessed 10 1 2024].

[53] T. Scheck, J. Lu, W. Callan and A. Canny, "Private COVID lab created headaches for Minnesota consumers, health officials," 28 July 2022. [Online]. Available: https://www.mprnews.org/story/2022/07/28/gs-labs-covid-testing-minnesota. [Accessed 21 August 2023].

[54] D. McKnight and P. Hinton, "Tort Costs in America: An Empirical Analysis of Costs and Compensation of the U.S. Tort System," 22 November 2022. [Online]. Available: https://instituteforlegalreform.com/research/tort-costs-in-america-an-empirical-analysis-of-costs-and-compensation-of-the-u-s-tort-system/. [Accessed 21 August 2023].

[55] U.S. Food and Drug Administration, "Summary; Medical Device Reporting: Electronic Submission Requirements Final Regulatory Impact Analysis," 14 Feb 2014. [Online]. Available: https://www.fda.gov/about-fda/economic-impact-analyses-fda-regulations/summary-medical-device-reporting-electronic-submission-requirements-final-rule.

[56] U.S. Food and Drug Administration, "Agency Information Collection Activities; Submission for Office of Management and Budget Review; Comment Request; Medical Devices; Reports of Corrections and Removals," 11 Aug 2023. [Online]. Available: https://www.regulations.gov/document/FDA-2023-N-1006-0002.

[57] U.S. Food and Drug Administration, "Medical Devices; Quality System Regulation Amendments (Proposed Rule) Preliminary Regulatory Impact Analysis," 02 03 2022. [Online]. Available: https://www.fda.gov/about-fda/economic-impact-analyses-fda-regulations/medical-devices-quality-system-regulation-amendments-proposed-rule-preliminary-regulatory-impact. [Accessed 25 01 2024].

[58] U.S. Food and Drug Administration, "Requirements for Foreign and Domestic Establishment Registration and Listing for Human Drugs, Including Drugs that Are Regulated Under a Biologics License Application, and Animal Drugs," Aug 2016. [Online]. Available: https://www.fda.gov/about-fda/economic-impact-analyses-fda-regulations/requirements-foreign-and-domestic-establishment-registration-and-listing-human-drugs-including-drugs.

[59] U.S. Food and Drug Administration, "Summary: Content and Format of Labeling for Human Prescription Drug and Biological Products; Requirements for Pregnancy and Lactation Labeling Final Regulatory Impact Analysis," 9 Dec 2014. [Online]. Available: https://www.fda.gov/about-fda/economic-impact-analyses-fda-regulations/summary-content-and-format-labeling-human-prescription-drug-and-biological-products-requirements.

[60] J. V. Batavia and S. J. Goldenberg, "Strategic Planning and Costs of FDA Regulation," *Academic Entrepreneurship for Medical and Health Scientists*, vol. 1, no. 2, 2019.

[61] I. Eastern Research Group, "Economic Analysis of CDRH Submission Requirements," 2012.

[62] U.S. Food and Drug Administration, "Microbiology Devices; Reclassification of Nucleic Acid-Based Systems for Mycobacterium tuberculosis complex Final Regulatory Impact Analysis," 2014. [Online]. Available: https://www.fda.gov/about-fda/economic-impact-analyses-fda-regulations/summary-microbiology-devices-reclassification-nucleic-acid-based-systems-mycobacterium-tuberculosis.

[63] U.S. Food and Drug Administration, "Food Safety Modernization Act Domestic and Foreign Facility Reinspection, Recall, and Importer Reinspection Fee Rates for Fiscal Year 2024," 28 July 2023. [Online]. Available: https://www.federalregister.gov/documents/2023/07/28/2023-15927/food-safety-modernization-act-domestic-and-foreign-facility-reinspection-recall-and-importer. [Accessed 20 October 2023].

[64] U.S. Food and Drug Administration, "Discussion Paper on Laboratory Developed Tests (LDTs)," 13 January 2017. [Online]. Available: https://www.fda.gov/media/102367/download. [Accessed 30 March 2023].

[65] G. S. Gerhard, S. G. Fisher and A. M. Feldman, "Genetic Testing for Inherited Cardiac Diseases in Underserved Populations of Non-European Ancestry: Double Disparity," *JAMA Cardiology*, pp. 3(4):273-274, 2018.

[66] A. K. Manrai, B. H. Funke, H. L. Rehm, M. S. Olesen, B. A. Maron, P. Szolovits, D. M. Margulies, J. Loscalzo and I. S. Kohane, "Genetic Misdiagnoses and the Potential for Health Disparities," *New England Journal of of Medicine*, pp. 375:655-665, 2016.

[67] A. R. Martin, M. Kanai, Y. Kamatani, Y. Okada, B. M. Neale and M. J. Daly, "Clinical Use of Current Polygenic Risk Scores May Exacerbate Health Disparities," *Nature Genetics*, pp. 51,584-591, 2019.

[68] Y. Wang, K. Tsuo, M. Kanai, B. M. Neale and A. R. Martin, "Challenges and Opportunities for Developing More Generalizable Polygenic Risk Scores," *Annual Review of Biomedical Data Science*, pp. 5:293-320, 2022.

[69] L. Duncan, H. Shen, B. Gelaye, J. Meijsen, K. Ressler, M. Felman, R. Peterson and B. Domingue, "Analysis of Polygenic Risk Score Usage and Performance in Diverse Human Populations," *Nature Communications*, p. 10(1):3328, 2019.
[70] K. F. Hoskins, O. C. Danciu, N. Y. Ko and G. S. Calip, "Association of Race/Ethnicity and the 21-Gene Recurrence Score With Breast Cancer–Specific Mortality Among US Women," *JAMA Oncology*, pp. 7(3):370-378, 2021.

[71] J. Gruber, "Chapter2," in *PublicFinanceandPublicPolicy*, WorthPublishers., 2016.

[72] Federal Reserve Bank of San Francisco, "Federal Reserve Bank of San Francisco Education Publications," November 2002. [Online]. Available: <https://www.frbsf.org/education/publications/doctor-econ/2002/november/private-social-costs-pollution-production/#fn>. [Accessed 21 August 2023].

[73] Congressional Research Service, "FDA Regulation of Laboratory-Developed Tests (LDTs)," 7 December 2022. [Online]. Available: <https://crsreports.congress.gov/product/pdf/IF/IF11389>. [Accessed 23 August 2023].

[74] J. K. Hammitt and J.-T. Liu, "Effects of Disease Type and Latency on the Value of Mortality Risk," *Journal of Risk and Uncertainty*, vol. 28, pp. 73-95, 2004.

[75] J. Hickner, P. Thompson, T. Wilkinson, P. Epner, M. Shaheen, A. Pollock, J. Lee, C. Duke, B. Jackson and J. Taylor, "Primary Care Physicians' Challenges in Ordering Clinical Laboratory Tests and Interpreting Results," *The Journal of the American Board of Family Medicine*, vol. 27, no. 2, pp. 268-274, 2014.

#### Appendix A. Estimation of the Number of Affected Labs and Tests

##### Number of Affected Labs

To obtain the number of laboratories and tests affected by the phaseout policy, we use data from the Centers for Medicare & Medicaid Services (CMS) that shows the number of laboratories by laboratory type (column A in Table A.1).[^96] Only laboratories that are certified under CLIA and meet the regulatory requirements under CLIA to perform high complexity testing are affected by the phaseout policy since LDTs are high complexity. To determine the number of affected high complexity laboratories, we must first determine the number of non-waived laboratories (column C in Table A.1) and then estimate how many of those can perform high complexity testing versus only moderate complexity testing. We determine the number of non-waived laboratories by excluding the number of laboratories whose certificate type is microscopy or waiver (column B in Table A.1). Since laboratories are certified only as waived or non-waived, without specific notation on whether they meet the regulatory requirements for high complexity testing or only moderate complexity testing, the estimate for how many of the non-waived laboratories can perform high complexity testing is based on FDA professional judgement and historical knowledge. Therefore, we estimated the percent of each laboratory type that was likely to be high complexity to determine the high complexity rate (column D). We then estimate the number of high complexity laboratories affected by the phaseout policy by multiplying column C times high complexity rate in column D (column E). We estimate that there are 11,808 high complexity CLIA laboratories. This is close to our original estimate of 12,000 used in the preliminary analysis (Ref. [1]) and a 2021 report from the Pew Charitable Trust (Ref. [12]) that also estimated that there are approximately 12,000 CLIA-certified laboratories performing high complexity testing.

As explained in section II.D.1, we rely on the information about laboratories and tests in NYS to estimate the percent of high complexity laboratories that make IVDs offered as LDTs. From the NYSDOH data, we calculate that approximately 10% of laboratories located in NYS that are certified under CLIA and that meet the regulatory requirements under CLIA to perform high complexity testing are developing IVDs offered as LDTs. We assume that NYS is representative of the U.S. laboratory community. We therefore estimate that approximately 10% of 11,808 (or 1,181) laboratories in the U.S. that are certified under CLIA and that meet the regulatory requirements under CLIA to perform high complexity testing currently manufacture IVDs offered as LDTs. To account for potential variability across the country, we estimate the proportion of high complexity laboratories making IVDs offered as LDTs to vary from 5% of 11,808 (or 590) laboratories to a high estimate of 20% of 11,808 (or 2,362) affected laboratories by reducing the primary estimate by 50% and doubling the primary estimate, respectively. In addition, we assume that there would be new laboratories entering the market every year (approximately 8 percent of the affected high complexity laboratories making IVDs offered as LDTs, ranging from 47 to 189, with a primary estimate of 94). While there are also likely laboratories that exit the market each year, we have not subtracted expected departing laboratory numbers from our cost estimates.

**Table A.1 Number of Affected Laboratories by Laboratory Type**
| Laboratory Type     | (A) Total | (B) Waiver and Microscopy | (C) Difference (A) - (B) | (D) High Complexity Rate | (E) No. HC Labs (C) x (D) | (F) Integrated Healthcare System Rate | (G) No. HC Labs within Healthcare System (E) x (F) |
|---------------------|-----------|---------------------------|--------------------------|--------------------------|--------------------------|-------------------------------------|-----------------------------------------------|
| Physician Office ** | 123,726   | 108,696                   | 15,030                   | 0                        | -                        | -                                   | -                                             |
| Other **            | 39,524    | 36,162                    | 3,362                    | 0.5                      | 1,681                    | 0                                   | -                                             |
| Pharmacy**                           | 29,120      | 29,110         | 10   | 0          | -          | -              |
| Intermediate Care Facility           | 28,758      | 28,705         | 53   | 0          | -          | -              |
| Skilled Nursing/Nursing Facility     | 15,154      | 15,109         | 45   | 0          | -          | -              |
| Home Health Agency                   | 13,824      | 13,809         | 15   | 0          | -          | -              |
| Assisted Living Facility             | 12,019      | 12,015         | 4    | 0          | -          | -              |
| Hospital                             | 9,399       | 2,599          | 6,800| 0.5        | 3,400      | 1              | 3,400          |
| Independent                          | 9,198       | 3,602          | 5,596| 0.75       | 4,197      | 0              | -              |
| Community Clinic                     | 8,165       | 6,686          | 1,479| 0.5        | 740        | 0.66           | 488            |
| End Stage Renal Disease Dialysis     | 7,301       | 7,294          | 7    | 0          | -          | -              |
| Ambulatory Surgery Center            | 7,150       | 6,719          | 431  | 0.75       | 323        | 0.5            | 162            |
| Other Practitioner                   | 6,688       | 6,653          | 215  | 0          | -          | -              |
| School/Student Health Service        | 6,729       | 6,599          | 130  | 0          | -          | -              |
| Ambulance                            | 5,968       | 5,917          | 51   | 0          | -          | -              |
| Hospice                              | 5,100       | 5,091          | 9    | 0          | -          | -              |
| Federally Qualified Health Center    | 4,556       | 4,332          | 224  | 0.5        | 112        | 0              | -              |
| Ancillary Test Site                  | 3,700       | 2,811          | 889  | 0.75       | 667        | 0.5            | 333            |
| Mobile Lab                           | 3,219       | 3,131          | 88   | 0.25       | 22         | 0.75           | 17             |
| Rural Health Care Clinic             | 2,726       | 2,512          | 214  | 0.25       | 54         | 0.25           | 13             |
| Industrial                           | 1,771       | 1,751          | 20   | 0.5        | 10         | 0              | -              |
| Comprehensive Outpatient Rehab       | 1,230       | 1,214          | 16   | 0.5        | 8          | 0.5            | 4              |
| Prison                               | 1,204       | 1,168          | 36   | 0.5        | 18         | 0              | -              |
| Public Health Laboratory             | 1,015       | 777            | 238  | 1          | 238        | 0              | -              |
| Health Maintenance Org               | 695         | 536            | 159  | 0.5        | 80         | 0.5            | 40             |
| Health Fair                          | 617         | 608            | 9    | 0          | -          | -              |
| Blood Banks                          | 426         | 167            | 259  | 1          | 259        | 0.5            | 130            |
| Insurance                            | 45          | 42             | 3    | 0          | -          | -              |
| **Total**                            | **349,207** | **313,815**    | **35,392** | | **11,808**     | **4,586**       |
||

##### Number of Affected Tests

We estimate the number of affected tests using information from NYSDOH. We then estimate the number of affected tests subject to different requirements and that would fall under different enforcement policies by adjusting the number of tests to account for the number that are currently marketed, the number expected to be impacted by potential reclassification of Class III IVDs to Class II IVDs, the number of LDTs expected to be for unmet needs in an integrated health care system, and the number of LDTs estimated to be reviewed by NYS CLEP each year.

1) Using NYSDOH Information

Using the NYSDOH information, FDA calculates that each laboratory that manufactures IVDs offers an average of 67 IVDs as LDTs and introduces an average of 6 new IVDs offered as LDTs per year (see section II.D.1 for details of the calculation). Multiplying 67 IVDs per lab by the number of affected laboratories, it is estimated that the number of affected currently marketed IVDs offered as LDTs ranges from 39,557 (=590 x 67) to 158,227 (=2,362 x 67), with a primary estimate of 79,114 (=1,181 x 67). Multiplying 6 new IVDs offered as LDTs per laboratory by the number of affected laboratories, we estimate 3,542, 7,085, or 14,170 new IVDs offered as LDTs may be affected per year. We also estimate new IVDs offered as LDTs from new laboratories entering the market every year (8 percent of affected laboratories). The total number of new IVDs offered as LDTs per year is estimated to range from 3,826 to 15,303, with a primary estimate of 7,652.

FDA generally intends to exercise enforcement discretion with respect to premarket review requirements for currently marketed IVDs offered as LDTs and LDTs manufactured and performed by a laboratory integrated within a healthcare system to meet an unmet need of patients receiving care within the same healthcare system. To estimate the number of high complexity laboratories within integrated healthcare systems which may fall under the enforcement discretion policy for unmet needs (column G in Table A.1 above), we first multiply the number of high complexity laboratories in each laboratory type in each row in Column E by the estimated rate of integrated healthcare system in Column F for that laboratory type. As with the high complexity rate, to determine the estimated rate of integrated healthcare system laboratories, we used professional judgment and historical knowledge to estimate the percent of high complexity laboratories we expect would be part of an integrated healthcare system for each laboratory type. We estimate that there are approximately 4,586 high complexity laboratories that may be part of an integrated healthcare system. Applying the proportion of high complexity laboratories making IVDs offered as LDTs (5%, 10%, 20%) to this estimate, we estimate the number of high complexity laboratories with LDTs within integrated healthcare systems to range from 229 (=5% x 4,586) to 917 (=20% x 4,586), with a primary estimate of 459 (=10% x 4,586). We assume that about 40 to 70 percent of LDTs from these high complexity laboratories integrated within healthcare systems are likely to be for unmet needs. The estimated number of LDTs that are likely to be for unmet needs range from 10,755 to 24,582, with a primary estimate of 16,900 (see Table A.2).

**Table A.2 Number of Affected Laboratories and LDTs For Unmet Needs**
| Number of affected labs and tests | Primary | Low  | High  | Calculation         |
|----------------------------------|---------|------|-------|---------------------|
| Affected labs (A)                | 1,181   | 590  | 2,362 | 11,808 * 5-20%      |
| Affected IVDs offered as LDTs on the market (B) | 79,114  | 39,557 | 158,227 | A * 67             |
| New IVDs offered as LDTs per year (C) | 7,652   | 3,826 | 15,303 | A * 1.08 * 6       |
| Affected labs for unmet needs (D) | 459     | 229  | 917   | 4,586 * 5-20%       |
| Affected LDTs on the market for unmet needs (E) | 16,900  | 10,755 | 24,582 | D * 67 * 40-70%    |
| New LDTs for unmet needs (F)     | 1,635   | 1,040 | 2,377 | D * 1.08 * 6 * 40-70% |
| Affected IVDs offered as LDTs excluding LDTs for unmet needs | 62,213  | 28,802 | 133,645 | B - E             |
| New IVDs offered as LDTs excluding LDTs for unmet needs | 6,017   | 2,786 | 12,926 | C - F             |
||

*Note: Product across table may not be exact due to rounding.*

Table A.3 shows the number of affected IVDs offered as LDTs by submission type, excluding tests for unmet needs, using the primary estimates from Table A.2. As explained in section II.D.1, we also expect that among currently-marketed IVDs offered as LDTs that were first marketed prior to the date of issuance of the rule, 2 IVDs will be modified per laboratory per year in a manner that falls outside of the enforcement discretion policy for currently marketed IVDs offered as LDTs and will thus be expected to undergo premarket review (2,362 = 1,181 affected labs * 2). We then add this number to the number of new tests (10,013 = 2,362 + 7,652) to estimate total number of new tests in column A of Table A.3. We further break down these estimates by submission type to estimate compliance costs. As mentioned in section II.F.2, we estimate that approximately 50 percent of IVDs currently undergo premarket review (5,007 = 10,013 * 0.5). In addition, we assume that about 40 percent are offered after 510(k) clearance (4,005 = 10,013 * 0.40), 5 percent after De Novo classification (501 = 10,013 * 0.05), and 5 percent after premarket approval (501 = 10,013 * 0.05). We assume these estimated percentages also apply to IVDs offered as LDTs and apply these shares to the estimated total number of affected tests, minus the number of affected tests expected to be offered under an enforcement policy without premarket submission, to estimate the number of IVDs offered as LDTs by submission type. For 2,139 LDTs for unmet needs (2,139 = 459 * 2 * 55% + 1,635)98, we estimate that approximately 10 percent are offered after premarket approval (214 = 2,139 * 0.10), 80 percent after 510(k) clearance (1,711 = 2,139 * 0.80), and 10 percent after De Novo classification (214 = 2,139 * 0.10). We then subtract the number of new LDTs for unmet needs (column B of Table A.3) from the number of new IVDs offered as LDTs (column A of Table A.3) to calculate the number of new IVDs offered as LDTs excluding LDTs for unmet needs.

**Table A.3 Number of IVDs offered as LDTs Per Year Excluding LDTs For Unmet Needs**
|| No. IVDs offered as LDTs, total (A) | No. LDTs for unmet needs (B) | No. IVDs offered as LDTs excluding LDTs for unmet needs (A - B) |
|-------------------------------------|-----------------------------|-------------------------------------------------------------|-------------------------------------------------------------|
| Total                               | 10,013                      | 2,139                                                       | 7,874                                                       |
| Exempt                              | 5,007                       | -                                                           | 5,007                                                       |
| PMA                                 | 501                         | 214                                                         | 287                                                         |
| 510 (k)                             | 4,005                       | 1,711                                                       | 2,294                                                       |
| De Novo                             | 501                         | 214                                                         | 287                                                         |
||

2) **Reclassification adjustment**

On January 31, 2024, FDA announced its intent to initiate the reclassification process for most IVDs that are currently Class III into Class II. The majority of these tests are infectious disease and companion diagnostic IVDs. Reclassification would allow manufacturers of certain types of tests to seek marketing clearance through the less burdensome 510(k) pathway rather than the PMA pathway. FDA also intends to continue taking a risk-based approach in the initial classification of IVDs to determine the appropriate level of regulatory controls and whether a new test may be classified into Class II through De Novo classification (and special controls established), rather than being class III and subject to the PMA pathway. Based on our experience, we believe that special controls could be developed that, along with general controls, could provide a reasonable assurance of safety and effectiveness for most future companion diagnostic and infectious disease IVDs. As such, first of a kind submission for such tests would be expected to submit a De Novo, with follow-on IVDs submitting 510(k) notifications. Therefore, our estimates for the number of affected IVDs that will be subject to certain requirements were adjusted based on the anticipated number of submissions following the potential reclassifications and considering that most future companion diagnostic and infectious disease IVDs would be reviewed through the De Novo and 510(k) pathways.

To estimate anticipated rates of submissions going forward, we rely on historical data for total IVD original and supplemental PMA submissions where FDA review started in the last ten fiscal years (FY 2014-2023). We use information for IVD PMA submissions received per fiscal year for certain CDRH-regulated infectious disease and companion diagnostic tests that have either been recently down classified or for which FDA intends to propose reclassification (Table A.4). The number of estimated De Novo submissions per year going forward is estimated based on the number of “first of a kind (FOAK)” original PMA submissions since, going forward, we would expect new FOAK tests to submit a De Novo instead of a PMA. The number of estimated 510(k) submissions per year going forward is estimated by considering the number of non-FOAK original PMA submissions as well as PMA supplements since, going forward, we would expect these submissions to be 510(k)s instead. (We excluded “135-Day Review Track for 30-Day Notice” and “Special CBE” PMA supplement types from our counts since no 510(k) would be expected for these types of modifications for Class II devices).

We use these data to calculate the reduction in anticipated PMA submissions and the increase in anticipated De Novo and 510(k) submissions going forward, following potential reclassification and considering that most future companion diagnostic and infectious disease IVDs would be Class II. This is represented by the last three columns of Table A.4, showing the rate of remaining PMAs, the rate of new De Novos, and the rate of new 510(k)s anticipated per year, respectively. Then we use these rates to determine the new estimates, as shown in Table A.5. For example, we multiply the remaining PMA rate (Column F of Table A.4) of 0.40 by the estimated number of PMAs before potential reclassification (Table A.5) to obtain the number of PMAs expected after potential reclassification (287 * 0.40 = 115) (Table A.5). We then multiply the De Novo rate of 0.06 (Column G of Table A.4) by the number of PMAs before potential...
reclassification and add to the number of De Novo submissions before potential reclassification to estimate the number of De Novo submissions after potential reclassification (287 * 0.06 + 287 = 304) (Table A.5).

**Table A.4 Rate of Potential Reclassification of PMA Submission Per Fiscal Year (FY 2014-2023)**

| Review Track                    | (A) Average Historical IVD submissions per Year | (B) # from column A that fall into proceeds considered for reclassification | (C) Average FOAK used to estimate future # De Novo per Year | (D) Average non-FOAK used to estimate future # 510(k) per Year | (E) Estimated # Remaining PMA per year in future | (F) Remaining PMA Rate =(E)/(A) | (G) De Novo Rate =(C)/(A) | (H) 510(k) Rate =(D)/(A) |
|---------------------------------|-----------------------------------------------|-------------------------------------------------------------|-------------------------------------------------------------|-------------------------------------------------------------|---------------------------------|---------------------------|-----------------------------|---------------------------|
| PMA Original                    | 11.7                                          | 7.0                                                         | 0.7                                                         | 6.3                                                         | 4.7                             | 0.40                      | 0.06                        | 0.54                      |
| 135 Review Track For 30-Day Notice | 13.7                                          | 5.3                                                         | 0.0                                                         | 0.0                                                         | 8.4                             | 0.61                      | 0.00                        | 0.00                      |
| Normal 180 Day Track            | 27.4                                          | 17.3                                                        | 0.0                                                         | 17.3                                                        | 10.1                            | 0.37                      | 0.00                        | 0.63                      |
| Normal 180 Day Track No User Fee| 15.8                                          | 9.4                                                         | 0.0                                                         | 9.4                                                         | 6.4                             | 0.41                      | 0.00                        | 0.59                      |
| Panel Track                     | 7.7                                           | 4.8                                                         | 0.0                                                         | 4.8                                                         | 2.9                             | 0.38                      | 0.00                        | 0.62                      |
| Real-Time Process               | 53.3                                          | 25.9                                                        | 0.0                                                         | 25.9                                                        | 27.4                            | 0.51                      | 0.00                        | 0.49                      |
| Special CBE                     | 11.3                                          | 5.5                                                         | 0.0                                                         | 0.0                                                         | 5.8                             | 0.51                      | 0.00                        | 0.00                      |
||

**Table A.5 Number of Submissions for IVDs offered as LDTs Before and After Potential Reclassification**
| Submission Type    | Before Potential Reclassification | After Potential Reclassification |
|--------------------|-----------------------------------|----------------------------------|
|                    | Primary | Low  | High            | Primary | Low  | High            |
| PMA                | 287     | 143  | 574             | 115     | 58   | 230             |
| 510(k)             | 2,294   | 1,147| 4,588           | 2,479   | 1,239| 4,958           |
| De Novo            | 287     | 143  | 574             | 304     | 152  | 608             |
| PMA supplements    | 67      | 33   | 133             | 35      | 17   | 70              |
||

3) **NYS CLEP review adjustment**

As discussed in the preamble, FDA intends to exercise enforcement discretion and generally not enforce premarket review requirements for LDTs that are approved by NYS CLEP. Therefore, we reduce the estimated number of premarket submissions to FDA by the number of premarket submissions expected to go to NYS CLEP. We use an estimate of the number of premarket submissions expected to be reviewed by NYS CLEP, rather than an estimate of the number of LDTs expected to be approved by NYS CLEP, due to the assumption that neither NYS CLEP nor FDA will approve/authorize 100% of the submissions reviewed, and workload costs are based on the number of submissions reviewed, not the number approved/authorized.

NYSDOH provided information indicating that they review an average of 888 LDTs per year, including high risk, moderate risk, low risk, and clinical trial IVDs based on NYSDOH criteria. Accounting for FDA’s assumption that 50% of IVDs offered as LDTs will be low risk and not subject to premarket review and that 21% of new IVDs offered as LDTs per year will fall within the enforcement discretion policy for unmet needs (0.21 = 1,635 new LDTs for unmet needs per year / 7,652 new IVDs offered as LDTs per year; see Table A.2), we estimate that an average of 351 (= [100% - 21%] x 50% x 888) new IVDs offered as LDTs per year that would normally be submitted to FDA for premarket review would not undergo FDA premarket review as a result of the enforcement discretion policy with respect to LDTs approved by NYS CLEP.

We then estimate that 12.1% of premarket submissions for IVDs offered as LDTs would be reviewed by NYS CLEP (and that these would not be submitted to FDA for review) by dividing 351 by 2,898 (the total number of new PMAs, 510(k)s, and De Novo submissions per year for affected tests, see Table A.5), excluding those that would be under the unmet needs enforcement discretion policy (0.121 = 351 / 2,898).

We thus exclude 12.1% from the number of affected tests after potential reclassification (Table A.5). Table A.6 below shows the final number of premarket submissions for IVDs offered as LDTs, excluding tests for unmet needs and submissions for LDTs expected to be reviewed by NYS CLEP, when estimating the costs of Stages 4 and 5 in section II.F.4. For example, we estimate the number of PMAs to be 101 (=115 - 115*0.121).

**Table A.6 Number of IVD Submissions After Potential Reclassification and Excluding Tests for Unmet Needs and LDTs Reviewed by NYS CLEP**
| Submission Type    | Primary | Low  | High |
|--------------------|---------|------|------|
| PMA                | 101     | 51   | 203  |
| 510(k)             | 2,179   | 1,090| 4,359|
| De Novo            | 267     | 134  | 534  |
| PMA supplements    | 31      | 15   | 61   |
||

### Appendix B. Final Small Entity Analysis Estimates

The purpose of this Appendix is to explain the steps for calculating the number of laboratories and existing IVDs offered as LDTs per receipt size category that was used for Section III. Final Small Entity Analysis. In Table 48 of Section III. Final Small Entity Analysis, we used detailed data from 2017 Statistics of U.S. Businesses on U.S. 6-digit NAICS detailed employment sizes and revenues to analyze the potential impacts of the phaseout policy on small entities. We initially use our estimated total market revenue for IVDs offered as LDTs of $14 billion in the bottom of column E in Table B.1 as our total annual receipts and extrapolate the share of annual receipt by enterprise size from the 2017 Census data corresponding to NAICS code 621511. This estimate is based on the assumption that 35% of revenue for this NAICS category is from IVDs offered as LDTs. ^99^ We also re-classify enterprise size categories given our new estimated average receipts per lab (Table B.2).

|**Table B.1 Growth Adjusted Annual Receipts from IVDs Offered as LDTs by Enterprise Size (2022, U.S. Dollars)**| ||
|----|--------------------------|-------|
| Number of Firms and Receipts by Enterprise Receipt Size 2017 | Receipts Only IVDs Offered as | Calculation |
||

| I | Enterprise Size ($1,000) | Firms | Receipts ($1,000) D | LDTs ($1,000) in 2022 dollars adjusted for growth since 2017 E | |
|---|---------------------|-------|-----------------|------------------------------------------|---------------------------|
| 1 | < $150 | 438 | $22,315 | $12,575 | E Total * D1/D Total |
| 2 | $151 - $999 | 933 | $250,134 | $140,954 | E Total * D2/D Total |
| 3 | $1000 - $1,999 | 413 | $301,551 | $169,929 | E Total * D3/D Total |
| 4 | $2,000 - $3,999 | 481 | $779,302 | $439,148 | E Total * D4/D Total |
| 5 | $4,000 - $5,999 | 343 | $1,224,596 | $690,078 | E Total * D5/D Total |
| 6 | $6,000 - $9,999 | 146 | $860,008 | $484,627 | E Total * D6/D Total |
| 7 | $10,000 - $14,999 | 77 | $595,808 | $335,747 | E Total * D7/D Total |
| 8 | $15,000 - $19,999 | 115 | $1,245,731 | $701,988 | E Total * D8/D Total |
| 9 | $20,000 - $24,999 | 79 | $1,168,397 | $658,409 | E Total * D9/D Total |
| 10 | $25,000 - $29,999 | 21 | $380,304 | $214,307 | E Total * D10/D Total |
| 11 | $30,000 - $39,999 | 43 | $908,377 | $511,884 | E Total * D11/D Total |
| 12 | $40,000 - $49,999 | 15 | $460,659 | $259,588 | E Total * D12/D Total |
| 13 | $50,000 - $99,999 | 67 | $2,190,319 | $1,234,278 | E Total * D13/D Total |
| 14 | $100,000 + | 194 | $25,270,700 | $14,240,421 | E Total * D14/D Total |
| | Total | 3,365 | $35,658,201 | $20,093,935 | |
||

We estimate the number of labs by receipt size category by the same proportion as the number of firms by receipt category from the Census data. For example, for firms with annual receipts less than <$150,000 we divided 438 by 3,365 and multiply by 1,275 (total LDT labs) to obtain 166 (438/3,365*1,275 = 166). We repeat this calculation for the rest of the rows. We then estimate the average receipts per laboratory by receipt size category.

**Table B.2 Estimated Number of LDT Laboratories and Average Annual Receipts per Laboratory (2022 U.S. dollars)**
| i  | Enterprise Size ($1,000) | Firms | Receipts ($1,000) | Receipts IVDs Offered as LDTs Only ($1,000) in 2022 dollars adjusted for growth since 2017 | LDT Labs (1,181) | Average Receipts per lab |
|----|--------------------------|-------|------------------|------------------------------------------------------------------------------------------|------------------|--------------------------|
| 1  | < $150                   | 438   | $22,315          | $12,575                                                                                  | 166              | $76                     |
| 2  | $151 - $499              | 933   | $250,134         | $140,954                                                                                 | 354              | $399                    |
| 3  | $500 - $999              | 413   | $301,551         | $169,929                                                                                 | 157              | $1,086                  |
| 4  | $1,000 - $2,999          | 481   | $779,302         | $439,148                                                                                 | 182              | $2,409                  |
| 5  | $3,000 - $5,999          | 343   | $1,224,596       | $690,078                                                                                 | 130              | $5,309                  |
| 6  | $6,000 - $7,999          | 146   | $860,008         | $484,627                                                                                 | 55               | $8,759                  |
| 7  | $8,000 - $9,999          | 77    | $595,808         | $335,747                                                                                 | 29               | $11,506                 |
| 8  | $10,000 - $14,999        | 115   | $1,245,731       | $701,988                                                                                 | 44               | $16,107                 |
| 9  | $15,000 - $19,999        | 79    | $1,168,397       | $658,409                                                                                 | 30               | $21,991                 |
| 10 | $20,000 - $23,999        | 21    | $380,304         | $214,307                                                                                 | 8                | $26,928                 |
| 11 | $24,000 - $29,999        | 43    | $908,377         | $511,884                                                                                 | 16               | $31,411                 |
| 12 | $30,000 - $35,999        | 15    | $460,659         | $259,588                                                                                 | 6                | $45,665                 |
| 13 | $36,000 - $99,999        | 67    | $2,190,319       | $1,234,278                                                                               | 25               | $48,610                 |
| 14 | $100,000 +               | 194   | $25,270,700      | $14,240,421                                                                              | 74               | $193,689                |
| Total | 3,365                | $35,658,201      | $20,093,935                                                                              | 1,275            | $15,757                 |
||

We obtain the number of IVDs offered as LDTs per receipts size category in column D by multiplying column C times 79,114 (which is our estimated number of affected IVDs offered as LDTs currently on the market, as described in section II.D.1). See Table B.3.

**Table B.3 Share of LDTs Offered as LDTs and IVDs Offered as LDTs per Receipt Category**
| i  | Enterprise Size ($1,000) | Percent Firms by Receipt Size A | LDT LABS B = A x 1,181 | Percent Receipts by Receipt Size C | IVDs offered as LDTs per receipt category * D |
|----|--------------------------|--------------------------------|------------------------|-----------------------------------|-----------------------------------------------|
| 1  | < $150                   | 13%                            | 166                    | 0.06%                             | 56                                          |
| 2  | $151 - $999              | 28%                            | 354                    | 0.70%                             | 625                                         |
| 3  | $1000 - $1,999           | 12%                            | 157                    | 0.85%                             | 754                                         |
| 4  | $2,000 - $3,999          | 14%                            | 182                    | 2.19%                             | 1,948                                       |
| 5  | $4,000 - $5,999          | 10%                            | 130                    | 3.43%                             | 3,061                                       |
| 6  | $6,000 - $9,999          | 4%                             | 55                     | 2.41%                             | 2,150                                       |
| 7  | $10,000 - $14,999        | 2%                             | 29                     | 1.67%                             | 1,489                                       |
| 8  | $15,000 - $19,999        | 3%                             | 44                     | 3.49%                             | 3,114                                       |
| 9  | $20,000 - $24,999        | 2%                             | 30                     | 3.28%                             | 2,920                                       |
| 10 | $25,000 - $29,999        | 1%                             | 8                      | 1.07%                             | 951                                         |
| 11 | $30,000 - $39,999        | 1%                             | 16                     | 2.55%                             | 2,270                                       |
| 12 | $40,000 - $49,999        | 0%                             | 6                      | 1.29%                             | 1,151                                       |
| 13 | $50,000 - $99,999        | 2%                             | 25                     | 6.14%                             | 5,475                                       |
| 14 | $100,000 +               | 6%                             | 74                     | 70.87%                            | 63,163                                      |
| Total | 100%                  | 1,275                 |  |  | 89,127                                      |
||

\* Column D is the product of each row in Column C and 79,114. 

The following four tables show detailed calculations leading to estimating cost as a percent of receipts using the primary, low and high estimates.

**Table B.4 Detailed Calculations for Percent Receipts and Average Receipts per Lab**
| i  | Enterprise Size ($1,000) | LDT Labs A | Receipts by Size Category B | Percent Receipts by Receipt Category C = B / B_total | Average Receipts per Lab D = B / A |
|----|--------------------------|------------|-----------------------------|-----------------------------------------------------|-----------------------------------|
| 1  | < $150                   | 166        | $12,574,839                 | 0.06%                                               | $75,755                           |
| 2  | $151 - $999              | 327        | $140,954,286                | 0.70%                                               | $430,532                          |
| 3  | $1000 - $1,999           | 145        | $169,928,542                | 0.85%                                               | $1,172,533                        |
| 4  | $2,000 - $3,999          | 169        | $439,148,446                | 2.19%                                               | $2,601,807                        |
| 5  | $4,000 - $5,999          | 120        | $690,078,339                | 3.43%                                               | $5,733,410                        |
| 6  | $6,000 - $9,999          | 51         | $484,627,495                | 2.41%                                               | $9,459,407                        |
| 7  | $10,000 - $14,999        | 27         | $335,746,086                | 1.67%                                               | $12,425,958                       |
| 8  | $15,000 - $19,999        | 24         | $701,988,231                | 3.49%                                               | $17,395,652                       |
| 9  | $20,000 - $23,999        | 28         | $658,409,354                | 3.28%                                               | $23,750,766                       |
| 10 | $25,000 - $29,999        | 7          | $214,307,047                | 1.07%                                               | $29,082,108                       |
| 11 | $30,000 - $39,999        | 15         | $511,884,157                | 2.55%                                               | $33,924,383                       |
| 12 | $40,000 - $49,999        | 5          | $259,588,803                | 1.29%                                               | $49,317,674                       |
| 13 | $50,000 - $99,999        | 24         | $1,234,277,833              | 6.14%                                               | $52,498,495                       |
| 14 | $100,000 +               | 68         | $14,240,421,072             | 70.87%                                              | $209,184,656                      |
| | Total | 1,193                | $20,093,934,751            |  | $16,841,846                      |
||

**Table B.5 Annualized Costs by Receipt Category - Primary, Low and High Estimates**
| i  | Enterprise Size ($1,000) | Annualized Cost by Receipt Category Primary, Low and High Estimates (7%) |||
|----|--------------------------|----------------------------------------------------------------------------|--------------------------|----------------------------------------------------------------------------|
|    | Calculation              | E (primary) = $1,165,774,775 x C_i     | E (low) = $510,386,965 x C_i     | E (high) = $3,222,662,070 x C_i  |
| 1  | $151 - $999              | $729,545                              | $319,402                          | $2,016,751                      |
| 2  | $1000 - $1,999           | $8,177,639                            | $3,580,246                        | $22,606,226                     |
| 3  | $2,000 - $3,999          | $9,858,617                            | $4,316,194                        | $27,253,113                     |
| 4  | $4,000 - $5,999          | $25,477,747                           | $11,154,393                       | $70,430,558                     |
| 5  | $6,000 - $9,999          | $40,035,759                           | $17,528,025                       | $110,674,655                    |
| 6  | $10,000 - $14,999        | $28,116,271                           | $12,309,563                       | $77,724,481                     |
| 7  | $15,000 - $19,999        | $19,478,771                           | $8,527,986                        | $53,847,019                     |
| 8  | $20,000 - $24,999        | $40,726,726                           | $17,830,537                       | $112,584,761                    |
| 9  | $25,000 - $29,999        | $38,198,443                           | $16,723,631                       | $105,595,588                    |
| 10 | $30,000 - $39,999        | $12,433,292                           | $5,443,410                        | $34,370,530                     |
| 11 | $40,000 - $49,999        | $29,697,600                           | $13,001,884                       | $82,095,900                     |
| 12 | $50,000 - $99,999  | $15,060,340  | $6,593,556  | $41,632,731  |
| 13 | $100,000 +        | $71,608,173  | $31,350,720 | $197,953,283 |
| 14 | Total             | $826,175,853 | $361,707,420| $2,283,876,474|
|    | Total             | $1,165,774,775| $510,386,965| $3,222,662,070
||

**Table B.6 Annualized Costs per Lab Receipt Category - Primary, Low and High Estimates**
| Enterprise Size ($1,000) | Annualized Cost per lab by Receipt Category | Primary, Low and High Estimates (7%) |||
|--------------------------|---------------------------------------------|--------------------------------------|---------------------------------------------|--------------------------------------|
| i | Calculation | F (primary) = F (low) = F (high) = |
|   |             | E (primary) / Aᵢ | E (low) / Aᵢ | E (high) / Aᵢ |
| 1 | <$150       | $4,395     | $1,924     | $12,150   |
| 2 | $151 - $999 | $24,978    | $10,936    | $69,049   |
| 3 | $1000 - $1,999 | $68,026 | $29,782    | $188,051  |
| 4 | $2,000 - $3,999 | $150,947 | $66,086    | $417,277  |
| 5 | $4,000 - $5,999 | $332,631 | $145,629   | $919,523  |
| 6 | $6,000 - $9,999 | $548,799 | $240,269   | $1,517,098|
| 7 | $10,000 - $14,999 | $720,908 | $315,620 | $1,992,873|
| 8 | $15,000 - $19,999 | $1,009,231 | $441,850 | $2,789,912|
| 9 | $20,000 - $24,999 | $1,377,930 | $603,271 | $3,809,144|
| 10 | $25,000 - $29,999 | $1,687,235 | $738,687 | $4,664,184|
| 11 | $30,000 - $39,999 | $1,968,166 | $861,681 | $5,440,787|
| 12 | $40,000 - $49,999 | $2,681,227 | $1,252,671 | $7,909,561|
| 13 | $50,000 - $99,999 | $3,045,766 | $1,333,464 | $8,419,700|
| 14 | $100,000 +        | $12,136,110 | $5,313,301 | $33,549,002|
|    | Total             | $977,101    | $427,784   | $2,701,093 |
||

**Table B.7 Costs as a Percent of Receipts - Primary, Low and High Estimates**
| Enterprise Size ($1,000) | Annualized Cost as a Percent of Receipts per lab | Primary, Low and High Estimates (7%) | ||
|--------------------------|-------------------------------------------------|--------------------------------------|-------------------------------------------------|--------------------------------------|
| i | Calculation | G (primary) = F (primary) / Dᵢ | G (low) = F (low) / Dᵢ | G (high) = F (high) / Dᵢ |
| 1 | $151 - $999 | 5.80% | 2.54% | 16.04% |
| 2 | $1000 - $1,999 | 5.80% | 2.54% | 16.04% |
| 3 | $2,000 - $3,999 | 5.80% | 2.54% | 16.04% |
| 4 | $4,000 - $5,999 | 5.80% | 2.54% | 16.04% |
| 5 | $6,000 - $9,999 | 5.80% | 2.54% | 16.04% |
| 6 | $10,000 - $14,999 | 5.80% | 2.54% | 16.04% |
| 7 | $15,000 - $19,999 | 5.80% | 2.54% | 16.04% |
| 8 | $20,000 - $24,999 | 5.80% | 2.54% | 16.04% |
| 9 | $25,000 - $29,999 | 5.80% | 2.54% | 16.04% |
| 10 | $30,000 - $39,999 | 5.80% | 2.54% | 16.04% |
| 11 | $40,000 - $49,999 | 5.80% | 2.54% | 16.04% |
| 12 | $50,000 - $99,999 | 5.80% | 2.54% | 16.04% |
| 13 | $100,000 +        | 5.80% | 2.54% | 16.04% |
|    | Total             | 5.80% | 2.54% | 16.04% |
||

Table B.8 below, describes total discounted costs to industry along with the percent share of costs by stage, LDTs affected, costs per LDT along with percent affected tests, entities affected and cost per lab.

**Table B.8 Discounted Costs to Industry and Percentage, by affected LDTs, by Entity and by Stage**
| Stage  | Discounted Costs ($ Millions, 7%, 20 years) | Percent Costs | LDTs Affected | Costs per LDT | Percent Tests Affected | Entities Affected | Costs per Lab |
|--------|--------------------------------------------|---------------|---------------|---------------|------------------------|-------------------|---------------|
| Stage 1 | $83  | 7%  | 10,013 | $8,335  | 11% | 1,275 | $65,457  |
| Stage 2 | $12  | 1%  | 10,013 | $1,187  | 11% | 1,275 | $9,326   |
| Stage 3 | $21  | 2%  | 52,641 | $391    | 59% | 849   | $24,216  |
| Stage 4 | $361 | 31% | 132    | $2,735,305 | 0.1% | 849 | $424,763 |
| Stage 5 | $689 | 59% | 2,446  | $281,807 | 3%  | 849   | $812,045  |
| Total   | $1,166 | 100% |  | $3,027,025 |   |  | $1,335,807 |
||

## CIVIL COVER SHEET

The JS 44 civil cover sheet and the information contained herein neither replace nor supplement the filing and service of pleadings or other papers as required by law, except as provided by local rules of court. This form, approved by the Judicial Conference of the United States in September 1974, is required for the use of the Clerk of Court for the purpose of initiating the civil docket sheet. (SEE INSTRUCTIONS ON NEXT PAGE OF THIS FORM.)

I:
(a) PLAINTIFFS: American Clinical Laboratory Association, et al.
    DEFENDANTS: US Food and Drug Administration, et al.

(b) County of Residence of First Listed Plaintiff: Washington, DC
    (EXCEPT IN U.S. PLAINTIFF CASES)
    County of Residence of First Listed Defendant: [blank]
    (IN U.S. PLAINTIFF CASES ONLY)
    NOTE: IN LAND CONDEMNATION CASES, USE THE LOCATION OF THE TRACT OF LAND INVOLVED.

(c) Attorneys (Firm Name, Address, and Telephone Number):
    King & Spalding LLP, 500 W 2nd Street, Ste 1800, Austin, TX 78701, (512) 457-2000
    Attorneys (If Known): [blank]

II. BASIS OF JURISDICTION (Place an "X" in One Box Only):
□ 1 U.S. Government Plaintiff
[X] 2 U.S. Government Defendant
□ 3 Federal Question (U.S. Government Not a Party)
□ 4 Diversity (Indicate Citizenship of Parties in Item III)

III. CITIZENSHIP OF PRINCIPAL PARTIES (Place an "X" in One Box for Plaintiff and One Box for Defendant):
PTF    DEF
□ 1    □ 1    Citizen of This State
□ 2    □ 2    Citizen of Another State
□ 3    □ 3    Citizen or Subject of a Foreign Country
□ 4    □ 4    Incorporated or Principal Place of Business In This State
□ 5    □ 5    Incorporated and Principal Place of Business In Another State
□ 6    □ 6    Foreign Nation

IV. NATURE OF SUIT (Place an "X" in One Box Only):

CONTRACT:
□ 110 Insurance
□ 120 Marine
□ 130 Miller Act
□ 140 Negotiable Instrument
□ 150 Recovery of Overpayment & Enforcement of Judgment
□ 151 Medicare Act
□ 152 Recovery of Defaulted Student Loans (Excludes Veterans)
□ 153 Recovery of Overpayment of Veteran's Benefits
□ 160 Stockholders' Suits
□ 190 Other Contract
□ 195 Contract Product Liability
□ 196 Franchise

REAL PROPERTY:
□ 210 Land Condemnation
□ 220 Foreclosure
□ 230 Rent Lease & Ejectment
□ 240 Torts to Land
□ 245 Tort Product Liability
□ 290 All Other Real Property

TORTS
PERSONAL INJURY:
□ 310 Airplane
□ 315 Airplane Product Liability
□ 320 Assault, Libel & Slander
□ 330 Federal Employers' Liability
□ 340 Marine
□ 345 Marine Product Liability
□ 350 Motor Vehicle
□ 355 Motor Vehicle Product Liability
□ 360 Other Personal Injury
□ 362 Personal Injury - Medical Malpractice

PERSONAL INJURY:
□ 365 Personal Injury - Product Liability
□ 367 Health Care/Pharmaceutical Personal Injury Product Liability
□ 368 Asbestos Personal Injury Product Liability

PERSONAL PROPERTY:
□ 370 Other Fraud
□ 371 Truth in Lending
□ 380 Other Personal Property Damage
□ 385 Property Damage Product Liability

CIVIL RIGHTS:
□ 440 Other Civil Rights
□ 441 Voting
□ 442 Employment
□ 443 Housing/Accommodations
□ 445 Amer. w/Disabilities - Employment
□ 446 Amer. w/Disabilities - Other
□ 448 Education

PRISONER PETITIONS:
Habeas Corpus:
□ 463 Alien Detainee
□ 510 Motions to Vacate Sentence
□ 530 General
□ 535 Death Penalty
Other:
□ 540 Mandamus & Other
□ 550 Civil Rights
□ 555 Prison Condition
□ 560 Civil Detainee - Conditions of Confinement

FORFEITURE/PENALTY:
□ 625 Drug Related Seizure of Property 21 USC 881
□ 690 Other

LABOR:
□ 710 Fair Labor Standards Act
□ 720 Labor/Management Relations
□ 740 Railway Labor Act
□ 751 Family and Medical Leave Act
□ 790 Other Labor Litigation
□ 791 Employee Retirement Income Security Act

BANKRUPTCY:
□ 422 Appeal 28 USC 158
□ 423 Withdrawal 28 USC 157

INTELLECTUAL PROPERTY RIGHTS:
□ 820 Copyrights
□ 830 Patent
□ 835 Patent - Abbreviated New Drug Application
□ 840 Trademark
□ 880 Defend Trade Secrets Act of 2016

SOCIAL SECURITY:
□ 861 HIA (1395ff)
□ 862 Black Lung (923)
□ 863 DIWC/DIWW (405(g))
□ 864 SSID Title XVI
□ 865 RSI (405(g))

FEDERAL TAX SUITS:
□ 870 Taxes (U.S. Plaintiff or Defendant)
□ 871 IRS—Third Party 26 USC 7609

IMMIGRATION:
□ 462 Naturalization Application
□ 465 Other Immigration Actions

OTHER STATUTES:
□ 375 False Claims Act
□ 376 Qui Tam (31 USC 3729(a))
□ 400 State Reapportionment
□ 410 Antitrust
□ 430 Banks and Banking
□ 450 Commerce
□ 460 Deportation
□ 470 Racketeer Influenced and Corrupt Organizations
□ 480 Consumer Credit (15 USC 1681 or 1692)
□ 485 Telephone Consumer Protection Act
□ 490 Cable/Sat TV
□ 850 Securities/Commodities Exchange
□ 890 Other Statutory Actions
□ 891 Agricultural Acts
□ 893 Environmental Matters
□ 895 Freedom of Information Act
□ 896 Arbitration
[X] 899 Administrative Procedure Act/Review or Appeal of Agency Decision
□ 950 Constitutionality of State Statutes

Here is the complete text extraction from the image, going through each section:

V. ORIGIN (Place an "X" in One Box Only)
[X] 1 Original Proceeding
□ 2 Removed from State Court
□ 3 Remanded from Appellate Court
□ 4 Reinstated or Reopened
□ 5 Transferred from Another District (specify)
□ 6 Multidistrict Litigation - Transfer
□ 8 Multidistrict Litigation - Direct File

VI. CAUSE OF ACTION
"Cite the U.S. Civil Statute under which you are filing (Do not cite jurisdictional statutes unless diversity):"
5 U.S.C. § 701 et seq., 21 U.S.C. § 301 et seq.
"Brief description of cause:"
Challenge to FDA final rule under Administrative Procedure Act

VII. REQUESTED IN COMPLAINT:
□ CHECK IF THIS IS A CLASS ACTION UNDER RULE 23, F.R.Cv.P.
DEMAND $
CHECK YES only if demanded in complaint:
JURY DEMAND: □ Yes □  No

VIII. RELATED CASE(S) IF ANY
(See instructions):
□ JUDGE _________________________ DOCKET NUMBER _________________________

DATE: 05/29/2024
SIGNATURE OF ATTORNEY OF RECORD: /s/Edward F. Fernandes

FOR OFFICE USE ONLY:
RECEIPT #
AMOUNT
APPLYING IFP
JUDGE
MAG. JUDGE

---

[^1]: "A laboratory developed test (LDT) is a type of in vitro diagnostic test that is designed, manufactured and used within a single laboratory." https://www.fda.gov/medical-devices/vitro-diagnostics/laboratory-developed-tests (Sept. 27, 2018).
[^2]: Given that FDA has not expressed its view in a rule subject to notice and comment rulemaking, its determination that an LTD is a medical device would not enjoy Chevron deference, but rather lesser Skidmore deference.
[^3]: 21st Century Cures: Examining the Regulation of Laboratory Developed Tests, Before the H. Comm. on Energy and Commerce Health Subcomm. 113 Cong. 91 (2014) (preliminary transcript) ("21st Century Cures hearing") (statement of Dr. Jeffrey Shuren, Director, CDRH. FDA); see also id. at 68.
[^4]: Pursuant to a delegation of authority from the Secretary of HHS, FDA is delegated limited authority under CLIA, to "implement CLIA's complexity categorization provisions as they apply to commercially available tests . . ." https://www.govinfo.gov/content/pkg/FR-2004-04-27/pdf/04-9527.pdf (emphasis added). This authority pertains to categorizing the complexity of such tests for purposes of CLIA. As discussed below, this involves physical test kits that are "commercially available" in interstate commerce, and not the type of laboratory developed testing at issue in this memorandum. This delegation of authority was recognized in statutory language in section 3057 of the 21st Century Cures Act. The Centers for Disease Control and Prevention (CDC) also provides analysis, research, and technical assistance with respect to CLIA and manages the Clinical Laboratory Improvement Advisory Committee (CLIAC).
[^3]: The new estimate of 69 IVDs offered as LDTs per affected entity was derived using comments and the initial PRIA estimate of 67. Specifically, we weighted together a maximum estimate of 310, a high estimate of 192, a middle estimate of 67, a low estimate of 13, and a minimum estimate of 1.
[^4]: FDA - CDRH Announces Intent to Initiate the Reclassification Process for Most High Risk IVDs. [https://www.fda.gov/medical-devices/medical-devices-news-and-events/cdrh-announces-intent-initiate-reclassification-process-most-high-risk-ivds](https://www.fda.gov/medical-devices/medical-devices-news-and-events/cdrh-announces-intent-initiate-reclassification-process-most-high-risk-ivds). As discussed in the preamble to the final rule, FDA notes that the reclassification process will include opportunities for public comment and FDA aims to complete the process before stage 4 of the phaseout policy.
[^5]: Statements in preamble which are not mirrored in the text of the rule are treated as interpretive rules, at best. See Wyeth v. Levine, 555 U.S. 555 (2009).
[^6]: See: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/vitro-diagnostic-multivariate-index-assays-draft-guidance-industry-clinical-laboratories-and-fda. In addition, in 2010, FDA held a widely-attended public meeting to solicit feedback from stakeholders on LDTs.
[^7]: See FDA Notification and Medical Device Reporting for Laboratory Developed Tests (LDTs) (Oct. 3, 2014) and Framework for Regulatory Oversight of Laboratory Developed Tests (LDTs) (Oct. 3, 2014).
[^8]: The regulation would not reach state public health laboratories or state university laboratories, since they are not "persons" within the meaning of the law. See FDCA § 201(e). 14 The term “person” is defined in the Agency’s premarket application rules as including “any individual, partnership, corporation, association, scientific or academic establishment, Government agency, or organizational unit thereof or any other legal entity.” 21 C.F.R. § 814.2. Inasmuch as the regulation is inconsistent with the definition of “person” in the Act, as including governmental entities, we believe that definition is invalid. 15 Counties and local governmental entities are “persons” under the FDCA. See, e.g., *Cook County, Ill. v. U.S. ex rel. Chandler*, 538 U.S. 119 (2003) (holding that a county is a “person” under the False Claims Act).
[^8]: Throughout this document, FDA uses the phrase “LDTs approved by NYS CLEP” to refer to LDTs that are approved, conditionally approved, or within an approved exemption from full technical documentation, under NYS CLEP. These three categories of LDTs are discussed further in section V.B.2 of the preamble. Other LDTs, including “LDTs used in Clinical Trials” and “Tests Not Subject to Evaluation” which are described on NYS CLEP’s website, are not considered “LDTs approved by NYS CLEP” and are not within the enforcement discretion policy with respect to premarket review requirements described in section V.B.2 of the preamble. For additional discussion of the NYS CLEP premarket review program, see section V.B.2 of the preamble.
[^9]: Memorandum to File from Elizabeth Hillebrenner, Associate Director for Scientific and Regulatory Programs, RE: Summary of 2020 Assessment of the First 125 EUA Requests from Laboratories for Molecular Diagnostic Tests for SARS-CoV-2 (September 22, 2023), available in Docket No. FDA-2023-N-2177.
[^12]: These requests resulted in review because FDA did not generally exercise enforcement discretion for LDTs intended for emergencies/potential emergencies/material threats declared under section 564 of the FD&C Act.
[^13]: Memorandum to File from Elizabeth Hillebrenner, Associate Director for Scientific and Regulatory Programs, RE: Summary of 2020 Assessment of the First 125 EUA Requests from Laboratories for Molecular Diagnostic Tests for SARS-CoV-2 (September 22, 2023), available in Docket No. FDA-2023-N-2177
[^14]: See Complaint, _In re Myriad Genetics, Inc. Sec. Litig._, No. 2:19-cv-00707-DBB (D. Utah 2019); Complaint, _Hickok v. Capone_, No. 2021-0686 (Del. Ch. 2021); Complaint, _Davis v. Natera, Inc._, No. 3:22-cv-00985 (N.D. Cal. 2022); Complaint, _Carroll v. Myriad Genetics Inc._, No. 4:22-CV-00739 (N.D. Cal. 2022); _Biesterfeld v. Ariosa Diagnostics, Inc._, No. 1:21-cv-03085, 2022 WL 972281 (N.D. Ill. 2022); and Complaint, _Kogus v. Capone_, No. 2022-0047-SG (Del. Ch. 2022).
15 Market distortions may be associated with events, decisions, or interventions taken by governments, companies, or other agents that influence the market in ways that undermine optimal allocation as modeled under the First Fundamental Theorem of Welfare Economics. Related concepts include market failure, government failure or behavioral bias (Ref. [71]).
[^16]: Social costs are costs incurred from the viewpoint of society (including external costs), beyond just stakeholders (private costs). When laboratories avoid paying for external costs arising from their actions (such as costs to manufacture tests with a reasonable assurance of safety and effectiveness, and if borne by individuals not involved in the decision to order such tests—for example, taxpayers funding government health insurance), the costs to society as a whole (such as non-internalized worsened health outcomes from inaccurate test results) remain. External costs, along with private costs, affect whether society is operating at a socially efficient rate of output (Ref. [72]).
[^17]: This includes laboratories operating under State licensure programs deemed exempt from CLIA.
[^18]: This discussion concerns only laboratories that are certified under CLIA and that meet the regulatory requirements under CLIA to perform high complexity testing (including laboratories operating under State licensure programs deemed exempt from CLIA), as other laboratories would be out of compliance with CLIA regulations if they were developing and performing tests that are not FDA-authorized. As noted in the preamble to the final rule, tests offered by such laboratories have never fallen within FDA’s general enforcement discretion approach.
[^19]: According to its website, CMS regulates all laboratory testing (except research) performed on humans in the U.S. through CLIA. In total, CLIA covers approximately 320,000 laboratories, but we do not know how many of these laboratories meet the regulatory requirements under CLIA to perform high complexity testing. It is worth noting that the number of CLIA certified laboratories, including laboratories that meet the requirements under CLIA for high complexity testing, can vary over time as new laboratories acquire certifications and others may close or lose their certification (Ref. [18]).
[^20]: We also examined census data. According to 2017 Statistics of U.S. Businesses (SUSB) data from the U.S. Census there are 3,365 Medical Laboratories (represented by NAICS code 621511). While data from the Census does not provide information on the number of laboratories under NAICS code 621511 that specifically manufacture IVDs offered as LDTs, if we assumed half of the entities were IVD manufacturers and the other half were laboratories, we would get 1,683 laboratories. The difference between this estimate and our primary estimate (502-[1,683-1,181]) is less than 5% of our primary estimate. We also consider varying our estimates by -5% and +10% to be sufficient for estimating the range of variability between our low and high estimate.
[^27]: From Newman-Toker (2023), annual US incidence was 6.0 M vascular events, 6.2 M infections and 1.5 M cancers. Per ‘Big Three’ dangerous disease case, weighted mean error and serious harm rates were 11.1% and 4.4%, respectively. Extrapolating to all diseases (including non-‘Big Three’ dangerous disease categories), the authors estimated 795,000 total serious harms annually in the USA (plausible range 598 000–1 023 000). Sensitivity analyses using other assumptions estimated 549 000 serious harms. Results were compatible with setting-specific serious harm estimates from inpatient, emergency department and ambulatory care (Ref. [34]).
[^28]: In general, the phaseout policy may reduce the risk of dying earlier or at a certain age (also referred to as the hazard function). This change in the hazard function can be expressed as a reduction in the expected number of deaths in a specified time period (less than one for an individual) or as an increase in the expected number of years lived (Ref. [74]).
[^29]: As noted above, per Newman-Toker et al. (2023), annual US incidence was 6.0 M vascular events, 6.2 M infections and 1.5 M cancers.
[^30]: According to the Census Bureau as of July 1, 2022, the population of NY was 19,677,151, and the population of the U.S. was 333,287,557. We thus assume that about 5.9% (= 19,677,151 / 333,287,557) of new U.S. cancers are in NY. For the population estimates used, refer to: [https://www.census.gov/quickfacts/fact/table/US/PST045222](https://www.census.gov/quickfacts/fact/table/US/PST045222).
[^33]: Calculated from Table 1 “Life table for the total population: United States, 2020” in the report “United States Life Tables, 2020,” available at: [CDC Life Tables](https://www.cdc.gov/nchs/data/nvsr/nvsr71/nvsr71-01.pdf)
[^34]: The approach for valuing mortality risk reductions is generally based on estimates of the value per statistical life (VSL), from which a value per statistical life year (VSLY) is derived. The VSLY values presented are updated to 2022 dollars per HHS guidance (Ref. [42]).
[^35]: We note that the HHS estimates of VSLY depend on the choice of discount rate.
37 Available at: [https://aspe.hhs.gov/sites/default/files/documents/796080c2812365443347c1cea347188/standard-ria-values-2024.xlsx](https://aspe.hhs.gov/sites/default/files/documents/796080c2812365443347c1cea347188/standard-ria-values-2024.xlsx)
[44](https://www.bls.gov/oes/current/naics4_621500.htm)  
[45] NAICS code: 621500  
[46] We use several sources to estimate costs associated with the phaseout policy. We use input from ICR packages for correction and removal reporting under Stage 1, but we rely on other sources such as previous RIAs or ERG reports to estimate costs for the other requirements. A major distinction between regulatory impact analysis (RIA) and Information Collection Request (ICR) is that the RIA bases estimates on external factors as yet unknown to FDA while ICR captures real-time data of the elements prescribed in regulations. In addition, the RIA considers affected entities according to specific requirements of regulations while the ICR includes a broader range for the number of respondents. Our estimates, therefore, may be different from the ICR package estimates because the RIA estimates costs of the rule in a different way compared as the costs of burden in the ICR package.  
[47] In this section, when we use the word “laboratories,” we refer to manufacturers who offer IVDs as LDTs that are within the scope of the phaseout policy.
[^56]: The ICR package of QS regulation (OMB control number 0910-0073) is available at: [https://www.reginfo.gov/public/do/PRAViewDocument?ref_nbr=202309-0910-003](https://www.reginfo.gov/public/do/PRAViewDocument?ref_nbr=202309-0910-003)
[^57]: We assume half of the labor hours would be for complaint files and the other half of the labor hours would be for other activities under subpart M.
[^58]: We anticipate that costs for compliance with any other requirements under Stage 2 such as mandatory recall orders under section 518(e) of the FD&C Act, or notification orders under section 518(a) of the FD&C Act would only be triggered under certain circumstances. Therefore, the costs are likely to be minimal compared to the costs for compliance with the requirements listed below. In addition, if requirements listed below are appropriately satisfied, these other requirements generally should not become applicable.
[^59]: The ICR package of registration and listing (OMB control number 0910-0625) is available at: [https://www.reginfo.gov/public/do/PRAViewDocument?ref_nbr=202206-0910-003](https://www.reginfo.gov/public/do/PRAViewDocument?ref_nbr=202206-0910-003)
[^60]: We assume that each affected laboratory will list an average number of 67 product listings and an average number of 6 new product listings per year, based on our estimates discussed in section II.D.1.
[^61]: NAICS code 621500, occupation codes 11-1021 for general and operations managers. Available from: [https://www.bls.gov/oes/current/naics4_621500.htm](https://www.bls.gov/oes/current/naics4_621500.htm)
[^62]: The ICR package of medical device labeling requirements (OMB control number 0910-0485) is available at: [https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202308-0910-004](https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202308-0910-004)
[^63]: As discussed in section II.D.1, we assume each affected laboratory offers 67 IVDs as LDTs and will offer 6 new IVDs as LDTs per year.
[^64]: The ICR package of investigational device exemptions (OMB control number 0910-0078) is available at: [https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202210-0910-013](https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202210-0910-013)
[^65]: IDE requirements in 21 CFR part 812 include certain requirements distinct from the requirement for approval of an IDE application, such as certain recordkeeping and labeling requirements. We anticipate that costs for compliance with these other requirements, where applicable, would be minimal compared to the costs discussed in this subsection for preparing and submitting an IDE application. These costs may be overestimated as the cited source reflects costs of IDE applications for all devices.
[^66]: Investigations of certain categories of devices are exempt from most requirements in 21 CFR part 812. See 21 CFR 812.2(c). Moreover, certain categories of investigations are considered to have an approved IDE application. See 21 CFR 812.2(b).
[^67] All wage rates are doubled to account for overhead costs. Available from: [https://www.bls.gov/oes/current/naics4_621500.htm](https://www.bls.gov/oes/current/naics4_621500.htm)
[^68] As discussed in section II.D.1, we assume each affected laboratory offers 67 IVDs as LDTs and will offer 6 new IVDs as LDTs per year.
[^70]: Another study by ERG used the same source for an analytical framework for examining the value of antibacterial products: [ASPE Report](https://aspe.hhs.gov/reports/analytical-framework-examining-value-antibacterial-products-0).
[^71]: We may under- or over-estimate the costs of premarket preparation since the estimates by ERG are not specific to IVDs. We have revised some of the estimates based on FDA professional judgement and historical knowledge.
[^72]: The ICR package of premarket approval of medical devices (OMB control number 0910-0231) is available at: [Reginfo](https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202211-0910-010).
[^73]: See the following page for a list of changes that would require a PMA supplement: [https://www.fda.gov/medical-devices/premarket-approval-pma/pma-supplements-and-amendments](https://www.fda.gov/medical-devices/premarket-approval-pma/pma-supplements-and-amendments).
[^74]: As of June 2023, the estimated number of active PMAs for all IVDs is 187 and total number of supplements over 7 years is 928. We divide the total number of supplements by the number of active PMAs and 7 years to calculate the number of PMA supplement per active PMA per year, which is 0.71 (= 928 supplements / 187 active PMAs / 7 years). We assume that the same rates for IVDs overall will apply to IVDs offered as LDTs.
[^75]: We use the FDA internal information on the total number of supplement submissions received by FDA from 2017 to 2023, as of July 2023.
[^76]: This cost reflects hours spent in CDRH substantive review of devices, required to determine whether they meet the standard to be approved. It does not include some of the steps required to complete review of a submission, such as management or time spent on such reviews by staff outside CDRH.
[^77]: The ICR package of premarket notification submission 510(k) (OMB control number 0910-0120) is available at: [https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202308-0910-018](https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202308-0910-018)
[^78]: Distinctions of the estimates used in Table 33 and Table 34 are based on the type of study supporting clinical validation of these tests due to differing costs.
[^79]: In the absence of more detailed information on De Novo costs, we extrapolate 510(k) costs to estimate De Novo costs.
[^81]: We extrapolate 510(k) costs to estimate De Novo costs, noting however that De Novo costs are likely higher than 510(k) costs. However, in the absence of more detailed information, we sometimes rely on such extrapolations to arrive at estimates due to uncertainty.
[^82]: This cost reflects hours spent in CDRH substantive review of devices, required to determine whether they meet the standard to be approved, cleared, or granted marketing authorization. It does not include some of the steps required to complete review of a submission, such as management or time spent on such reviews by staff outside CDRH.
[^83]: Our estimates of review costs of Q-submissions are uncertain as we assume every premarket submission will have a Q-submission, which is unlikely, and we do not account for the fact that some premarket submissions have multiple Q-submissions or that we receive some Q-submissions that are not tied to a premarket submission. Further, we do not account for the likelihood of receiving Q-submissions prior to Stage 3 as laboratories prepare in advance for premarket submissions.
[^84]: The costs could be spread over time depending on the time of submission and review.
[^87]: Note that under the phaseout policy, FDA intends to phase out the general enforcement discretion approach for establishment registration requirements during the current MDUFA V program, such that user fee payments for establishment registrations (which are distinct from user fee payments for premarket submissions) will be subject to the current MDUFA V fee structure.
[^93]: The average annualized cost per small entity ranges from $4,395 per laboratory with annual receipts that are less than $150,000 and to $3,045,766 per laboratory with annual receipts between $40 and $49.99 million.
[^94]: Although businesses with gross receipts of $100 million or less are eligible to pay a reduced fee, we estimate the transfers per firm by receipts size category using the total transfers paid proportional to the share of IVDs offered as LDTs per receipt category in Table B.3 column D in Appendix B. For example, for firms making less than $150,000 in annual revenues, their share of total annual fees would be equivalent to 0.12% or $95 divided by $81,816 of the low estimate of total annualized fees of $2331 million.