METADATA
last updated: 2026-03-02 RT
file_name: _AI_open-euas-open-access-diagnostics-report.md
file_date: 2026-03-02
title: FloodLAMP Open EUAs - Open-Source Protocol and Open-Access Authorization for Diagnostic Tests
category: regulatory
subcategory: open-euas
tags: open-eua, open-source-diagnostics, salivadirect, fda-eua, cdc-test, open-access, ldt-oversight
source_file_type: md
xfile_type: NA
gfile_url: https://docs.google.com/document/d/1EZbWftQkdu0VnPlYXqMveg-ZhTDXMqtYZMz5NR-3Z9A
xfile_github_download_url: https://raw.githubusercontent.com/FocusOnFoundationsNonprofit/floodlamp-archive/main/regulatory/open-euas/_AI_open-euas-open-access-diagnostics-report.md
pdf_gdrive_url: NA
pdf_github_url: NA
conversion_input_file_type: NA
conversion: NA
license: CC BY 4.0 - https://creativecommons.org/licenses/by/4.0/
tokens: 12968
words: 8079
notes: Initially created by ChatGPT Pro 5.2 Extended, edited by Opus 4.6 and RT during archive preparation. **PARTIALLY HUMAN EDITED/VERIFIED - MAY CONTAIN ERRORS** Research report analyzing the "open EUA" concept (open-source protocol combined with open-access authorization) centered on SalivaDirect as the canonical example, with analysis of the CDC test, pre-pandemic antecedents, regulatory barriers to proliferation, and implications for future open-access diagnostics. Key sources include FDA authorization documents, SalivaDirect retrospective literature, policy analyses, and FloodLAMP archive materials.
summary_short: Research report on "open EUAs" -- the combination of open-source diagnostic protocols with open-access FDA authorization -- centered on SalivaDirect as the canonical example. Covers the regulatory mechanics of the steward/designation model, evidence of uptake in press and academia, pre-pandemic antecedents, reasons the model did not proliferate during COVID-19, a detailed analysis of the CDC test as a government "open" diagnostic with structural limitations, and a practical framework for future open-access diagnostics. Includes a comparative openness table across six COVID-19 EUA test categories.


CONTENT

***INTERNAL TITLE:*** Open EUAs for Diagnostic Tests (Open-Source Protocol + Open-Access Authorization)
**A research report centered on the SalivaDirect “open source protocol” EUA model and what happened to the concept (pre‑COVID → March 2026)**

**Archive Files included in context window:**
`regulatory/open-euas/Anne Wyllie Nomination for FDA Foundation 2022 Innovations in Regulatory Science Awards.md`
`regulatory/open-euas/2020-08-15_FDA Website - Press Release FDA Issues EUA for SalivaDirect.md`
Added during Opus 4.6 edits:
`regulatory/fda-euas/2020-12-01_CDC EUA IFU - CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel.md`

---

## Table of contents
1. [Executive summary](#executive-summary)  
2. [What an “open EUA” actually means](#what-an-open-eua-actually-means)  
3. [Case study: SalivaDirect as the canonical open‑protocol/open‑access EUA](#case-study-salivadirect-as-the-canonical-openprotocolopenaccess-eua)  
4. [Evidence of uptake in press, academia, and FDA-facing discussions](#evidence-of-uptake-in-press-academia-and-fda-facing-discussions)  
5. [Did FDA authorize *other* open‑protocol/open‑access EUAs?](#did-fda-authorize-other-openprotocolopenaccess-euas)  
6. [Pre‑pandemic antecedents and “why this wasn’t obviously new”](#prepandemic-antecedents-and-why-this-wasnt-obviously-new)  
7. [Why the open‑protocol/open‑access model didn’t proliferate in COVID‑19](#why-the-openprotocolopenaccess-model-didnt-proliferate-in-covid19)  
8. [Where things stand now (2023–2026): open protocols vs open access](#where-things-stand-now-20232026-open-protocols-vs-open-access)  
9. [A practical framework for future “open access diagnostics”](#a-practical-framework-for-future-open-access-diagnostics)  
10. [The CDC test as the canonical "open" government diagnostic](#the-cdc-test-as-the-canonical-open-government-diagnostic)  
11. [Comparative table: openness characteristics of COVID-19 EUA diagnostic tests](#comparative-table-openness-characteristics-of-covid-19-eua-diagnostic-tests)  
12. [Appendices](#appendices)  
13. [References](#references)  

---

## Executive summary

### The core idea
An **“open EUA”** (as discussed here) is not merely publishing a protocol. It is an **institutional and legal pattern** that combines:

- **Open-source protocol**: the assay design and workflow are fully disclosed and implementable without proprietary equipment; and key components can be sourced from **multiple suppliers**.  
- **Open-access authorization**: *multiple* independent clinical laboratories can legally run the test **under one authorization**, rather than each needing their own EUA/510(k)/De Novo, *because* a single steward (the EUA holder) **designates/authorizes participating labs** and maintains a “single source of truth” for validated configurations and updates.

SalivaDirect is the clearest widely-recognized example of this in FDA’s COVID‑19 test authorization corpus, and FDA itself explicitly described it as an “open source” protocol that *designated* labs could run ([1], [21]).

### What SalivaDirect contributed (beyond “saliva”)
Across public sources, SalivaDirect is repeatedly characterized as:
- **Extraction‑free** (skipping nucleic acid extraction), with supply‑chain benefits ([1], [8]).  
- **Flexible validation** across multiple reagent/instrument combinations (“off‑the‑shelf” components) ([1], [7]).  
- A **novel regulatory framework**: an EUA held centrally, with **quality oversight** by the steward and execution at **designated high‑complexity labs**, leading to a large national lab network and millions of tests ([2], [7]).

A 2024 retrospective paper from the SalivaDirect team describes a “novel regulatory framework” established with FDA—EUA to Yale, protocol offered to designated CLIA high‑complexity labs with quality oversight by Yale—and reports **>200 labs** and **>6.5 million tests as of May 2024** ([7]).

### Was the concept “picked up” in the press and academic retrospectives?
Yes—but unevenly.

- **Press**: Many stories emphasized *saliva* and *skipping extraction*; fewer centered the **regulatory novelty** (open protocol + designated-lab access). Still, multiple outlets explicitly called it “open source” and described designated labs using the protocol ([15]–[19]).  
- **Academic / policy**: A citizen-science policy analysis (2022) framed COVID as revealing structural barriers to open-source diagnostics and argued U.S. approval models are poorly suited to them ([9]). A 2024 SalivaDirect retrospective explicitly positions the model as scalable and replicable for future outbreaks ([7]).

### Was this concept considered before COVID?
Elements existed, but the **exact combination** was unusual:

- **Open protocols** in diagnostics have long existed via publications and CLIA‑governed laboratory-developed tests (LDTs).  
- **Networked testing under centralized authority** also existed in public-health contexts (e.g., reference lab networks), and EUA mechanisms existed pre‑COVID.  
- What appears rarer is FDA’s explicit endorsement of an *“open source protocol”* framing **plus** a centralized steward role enabling broad adoption in many unaffiliated CLIA labs without the steward manufacturing a proprietary kit ([1], [7], [20]).

### Why didn’t the model proliferate?
Public evidence points to several interacting frictions:

1. **Regulatory incentives changed mid‑pandemic**: HHS policy changes in Aug/Oct 2020 reduced FDA review of COVID LDT EUAs, likely reducing the value of building an “open‑access EUA umbrella,” because labs could implement LDTs without EUA in that period ([6], [11]).  
2. **Stewardship burden**: The steward must maintain validated “test systems” (not just “protocols”), manage amendments/bridging studies, and oversee a lab network with reporting obligations ([2], [7]).  
3. **Commercial incentives**: many developers pursued proprietary kits/platforms; open protocols don’t create the same exclusivity.  
4. **Market shift**: U.S. testing demand shifted toward rapid antigen and at‑home tests; an open PCR protocol for high‑complexity labs is not a direct substitute.

### Current state (through March 2026)
- SalivaDirect remains an ongoing example: by mid‑2024 SalivaDirect, Inc. held multiple EUAs and continued updating reagent/instrument options through amendments and bridging studies ([7]).  
- The broader “open access” question is tightly linked to U.S. policy oscillations on **LDT oversight** (and legal challenges to FDA’s attempted expansions), which affect whether “open protocols” propagate via CLIA LDTs versus via centralized FDA authorizations ([6], [12]–[14]).

---

## What an “open EUA” actually means

### EUA reality: FDA authorizes **test systems**, not “ideas”
One of the central observations in the attached nomination letter is that FDA generally authorizes a **defined test system**: sample type, reagents, instruments, workflow, and performance claims must be specified; FDA does not simply approve a “protocol” in the abstract ([20]).

That has two consequences:

1. **Open publication ≠ open access.** Publishing a complete protocol does not automatically allow others to run it “as that FDA‑authorized test” unless they are legally covered under the authorization.  
2. **Openness must be engineered into the regulatory wrapper.** To make a published protocol “open access,” you need a regulatory mechanism that authorizes execution by multiple independent entities.

### A simple taxonomy: “open” has at least 4 layers
| Layer | What’s “open”? | Typical mechanism | What it enables | What it *doesn’t* solve |
|---|---|---|---|---|
| 1. **Open science** | Paper/preprint describes workflow | Publication | Scientific replication | Legal authority to offer clinical testing under FDA authorization |
| 2. **Open protocol** | Step-by-step IFU-like protocol, implementable with commodity parts | Protocol repositories, publications, SOP sharing | Faster adoption by capable labs | A lab is still “on its own” for regulatory posture (LDT vs EUA) |
| 3. **Open supply chain** | Multiple validated suppliers of key reagents/instruments | Validation across alternatives; equivalence/bridging | Resilience against shortages; cost competition | Requires ongoing maintenance as suppliers change |
| 4. **Open access authorization** | Multiple labs can legally run under one authorization | Designated/authorized labs under an EUA held by a steward | Scaling without requiring each lab to submit full EUA | Governance, liability, quality control, and costs shift to steward |

In this framing, **SalivaDirect** is notable because it attempted to combine **all four layers** ([1], [7], [20]).

### Why “open source” is a loaded term in IVD regulation
In software, “open source” implies license terms and freedom to copy/modify. In IVDs, analogous freedoms collide with:
- **Quality System Regulation / manufacturing controls** (for commercial IVDs)  
- **CLIA validation responsibilities** (for LDTs)  
- **FDA’s product-specific authorization posture** (for EUA, 510(k), De Novo, PMA)  
- **Liability and post‑market surveillance expectations**  

Thus, an “open source diagnostic” in the *regulatory* sense usually needs a **governance model** (a steward, a network, contracts, verification, versioning).

---

## Case study: SalivaDirect as the canonical open‑protocol/open‑access EUA

### 1) What FDA publicly said in August 2020
In its August 15, 2020 press release announcing the EUA, FDA emphasized:
- saliva collected in a sterile container  
- the ability to avoid a separate nucleic‑acid extraction step  
- validation across different combinations of commonly used reagents/instruments  
- and explicitly: Yale intended to provide the protocol as an **“open source” protocol** for **designated laboratories** to run under Yale’s instructions ([1], [21]).

That press release is one of the most explicit FDA uses of “open source” framing for an EUA’d diagnostic protocol.

### 2) What the nomination letter claims was novel
The attached nomination letter (June 3, 2022) frames SalivaDirect’s “open source protocol” EUA as unusual because:
- the academic team sought an IVD EUA **without intending to manufacture kits**,  
- they **did not patent** the test, used **CDC primers**, fully disclosed ingredients, and validated multiple suppliers and instruments,  
- FDA created a **designation** process allowing CLIA labs to use the EUA, and  
- the lab network plus multi‑supplier validation produced supply‑chain redundancy and contributed to cost compression ([20]).

Some of these points are corroborated by FDA documents and later retrospectives ([1], [2], [7]); others (e.g., “most unusual submission FDA had ever received”) appear as narrative claims that would require independent confirmation beyond the nomination letter itself.

### 3) How the EUA wrapper operationalized “open access”
FDA authorization materials for SalivaDirect (held by SalivaDirect, Inc.) explicitly limit testing to **laboratories designated by the EUA holder** and impose obligations on both the steward and authorized laboratories (e.g., use only as authorized, reporting/recordkeeping, false result reporting pathways, etc.) ([2], [3]).

This is the “open access” scaffold: **one legal authorization → many performing labs**, with the steward acting as a compliance and quality gate.

### 4) The “steward model” in one picture (textual)
**Traditional kit EUA**
1. Manufacturer validates a fixed kit configuration.  
2. FDA authorizes the kit.  
3. Any qualifying lab buys and runs the kit (no “designation” step).  

**SalivaDirect-style open protocol + open access**
1. Steward validates multiple acceptable configurations (reagents/instruments/workflows).  
2. FDA authorizes the test system(s) as described in authorized labeling.  
3. Steward **designates** labs that can run the protocol under the EUA.  
4. Steward maintains the configuration list and manages amendments/bridging as supply chain or workflows change ([2], [7]).

### 5) Reported scale and evolution
A 2024 retrospective paper from the SalivaDirect team describes:
- the protocol as open-source and extraction-free using off-the-shelf reagents/equipment,  
- a “novel regulatory framework” with EUA to Yale and designated CLIA high‑complexity labs under Yale’s oversight,  
- growth to a U.S. network of **>200 labs**, and  
- **>6.5 million tests as of May 2024**, plus multiple EUA expansions/amendments over time ([7]).

### 6) Why SalivaDirect is often described as “generic-like” diagnostics
The nomination letter’s “generics of diagnostics” analogy is directionally helpful but imperfect ([20]).

| Feature | Small-molecule drug generics (ANDA model) | SalivaDirect-style “generic-like” diagnostics |
|---|---|---|
| IP baseline | Active ingredient and reference product known; patents expire | Assay components may be public, but performance depends on workflow/instrument/reagent interactions |
| Regulatory mechanism | ANDA relies on reference listed drug, bioequivalence | No true “ANDA for diagnostics”; 510(k) is predicate-based but not a “copy this exact spec” pathway |
| Substitutability | High (chemically the same) | Variable (lab execution, instruments, reagents can shift performance) |
| Competition lever | Multiple manufacturers produce same product | Multiple suppliers can provide components; labs can implement; steward can validate combinations |

SalivaDirect’s key “generic-like” move was to pre‑validate multiple supply chain options (and keep adding more) while providing a compliance wrapper for widespread CLIA lab use ([1], [7]).

---

## Evidence of uptake in press, academia, and FDA-facing discussions

This section answers: **Was the open protocol + designation + stewardship concept discussed outside the nomination letter?**

### A) Press coverage: what was emphasized
Below is a non‑exhaustive sample of media coverage that explicitly references **open source** and/or the **designated lab** model.

| Outlet | Date | What it highlighted | Notes |
|---|---:|---|---|
| FDA press release | 2020‑08‑15 | “open source” protocol; designated labs; multi‑reagent/instrument validation | Primary source of FDA’s framing ([1]) |
| ContagionLive | 2020‑08‑15 | “open source system” where designated labs follow protocol | Trade/clinical media ([15]) |
| STAT | 2020‑08‑15 | Open protocol; access and cost angles | General audience ([16]) |
| TIME | 2020‑10‑13 | SalivaDirect as one path to broader testing access | More about access ([17]) |
| LabMedica | 2020‑08‑17 | Open‑source saliva PCR test and scaling | Trade press ([18]) |
| SelectScience | 2022 (interview) | “open-access protocol”; designated labs; amendments for redundancy | Quantitative adoption claims ([19]) |

**Observation:** Even when “open source” is mentioned, press stories often focus on *saliva* and *no extraction* as the innovation, with the open‑access regulatory design treated as secondary. This helps explain why the “open EUA” concept did not become a broadly discussed regulatory category during the pandemic.

### B) Academic and policy literature
A few categories of publications matter for understanding whether the concept “landed” in the literature:

1) **Method and validation papers** (how to implement and how well it performs)  
2) **Implementation/network papers** (how it scaled, governance, real-world operations)  
3) **Regulatory/policy retrospectives** (what it implies for future oversight and innovation)

A sample map:

| Publication type | Example(s) | Key relevance to “open EUA” idea |
|---|---|---|
| Method/validation | Vogels et al. SalivaDirect paper (2020) ([8]) | Simplified/flexible protocol; supports “off-the-shelf” concept |
| Implementation & scaling | SalivaDirect retrospective (2024) ([7]) | Explicitly frames “novel regulatory framework”; lab network, amendments, bridging studies |
| Policy analysis of open-source diagnostics | Monaco & Ware (2022) ([9]) | U.S. approval model poorly suited for open-source diagnostics; COVID as a stress test |
| Broader “open-source diagnostics” models | Emperador et al. (2020) ([10]) | Proposes open-source co-development model for outbreak preparedness (conceptual antecedent) |

### C) FDA-facing discussions and “signals” from regulators
Even if the concept didn’t lead to many additional “open protocol EUAs,” FDA documentation and public-facing materials show the agency had mechanisms and vocabulary for:
- “authorized” vs “designated” high-complexity labs ([2], [5])  
- umbrella EUAs for certain test classes ([4], [5])  
- templates to standardize submissions and accelerate review ([4])

FDA also published after-action and assessment materials describing how it prioritized tests that addressed shortages (e.g., pivot to saliva amid swab constraints) and how policy changes around LDT oversight affected EUA review load ([6]).

---

## Did FDA authorize other open‑protocol/open‑access EUAs?

### 1) “Designated laboratories” are not unique—but the *open protocol* framing is
FDA has long included in EUA scope/conditions the ability to restrict a test to:
- a single laboratory (classic LDT),  
- a named set of laboratories, or  
- “designated” laboratories (designated by an EUA holder) ([5]).

So the *designation tool* itself is not necessarily unique. What stands out about SalivaDirect is:
- FDA’s explicit “open source protocol” framing in a press announcement ([1]); and  
- deliberate design around **multi‑supplier components** and **commodity equipment** to function as a generic-like, supply-chain-resilient assay ([7], [20]).

### 2) Umbrella EUAs: an alternate scaling pathway that looks “open” but isn’t
FDA issued “umbrella EUAs” for certain molecular test categories during COVID. These umbrellas can make it *easier* for many labs to be authorized, but they typically still result in **lab‑specific** authorizations or restrictions, and they do not automatically create a single stewarded open protocol ([4], [5]).

In other words:
- Umbrella EUA = standardized pathway to get many authorizations  
- SalivaDirect model = one authorization + stewarded lab network running a shared protocol

### 3) Attempts to replicate the “open-source protocol EUA” idea
There is evidence that other groups pursued “open-source protocol EUA” submissions. For example, a 2022 meeting summary from the COVID‑19 Diagnostics Evidence Accelerator reports FloodLAMP stated it had submitted multiple open-source protocol EUAs and a pre‑EUA (and sought to resubmit with partners/federal support) ([22]). This suggests aspiration and activity—but not widespread authorization outcomes.

### 4) Bottom line
Based on public evidence reviewed for this report, SalivaDirect is:
- the most prominent FDA-described “open source protocol” EUA for a COVID PCR assay, and  
- a rare example where the steward did not rely on proprietary equipment and instead validated multiple supply chain options and maintained a multi‑lab network under one authorization ([1], [2], [7]).

---

## Pre‑pandemic antecedents and “why this wasn’t obviously new”

### 1) Open protocols existed (but not as “open access FDA authorizations”)
Before COVID, diagnostic methods were routinely published and replicated as:
- research assays  
- CLIA laboratory-developed tests (LDTs), often as “homebrew” assays implemented from literature  

This is “open protocol” in the scientific sense, but it is not necessarily “open access under FDA authorization.”

### 2) Centralized oversight + distributed execution existed in public health
Public health labs and reference networks have long used shared protocols and central coordination. In emergencies, EUA authority existed pre‑COVID.

SalivaDirect’s twist was **pairing that networked operational model** with:
- *explicit* open-source positioning; and  
- a design optimized for commodity supply chains rather than a proprietary kit ([1], [7], [20]).

### 3) “Open-source diagnostics” as preparedness strategy was discussed pre‑/early‑COVID
In 2020, Emperador et al. published a model for open-source co-development partnerships for molecular diagnostics, explicitly tied to outbreak and epidemic preparedness, emphasizing flexible, sustainable approaches (especially relevant to low- and middle-income country settings) ([10]).

This suggests the idea of open-source diagnostics was “in the air,” but the FDA EUA instantiation (as open access) was still unusual.

---

## Why the open‑protocol/open‑access model didn’t proliferate in COVID‑19

This section synthesizes the most plausible explanations supported by public evidence, plus reasonable inferences.

### 1) Mid‑pandemic policy whiplash around LDT EUAs
In August 2020 HHS announced a policy position affecting FDA’s ability to require premarket review/EUAs for LDTs, and FDA subsequently deprioritized or stopped reviewing many COVID LDT EUA requests in the fall of 2020 ([6], [11]).

FDA’s EUA assessment report later described that following the HHS statement, FDA “Declined to Review” a large number of LDT EUA requests, in part because LDTs could be marketed without an EUA during that period and FDA prioritized other categories (e.g., home collection and POC tests) ([6]).

**Implication for open EUAs:** If many labs could offer COVID testing as LDTs without an EUA, the incremental value of building a stewarded open-access EUA network may have dropped—especially for teams without resources to manage ongoing stewardship.

Later, in November 2021 HHS withdrew/reversed the 2020 policy, and FDA updated its COVID test policy accordingly ([12]). But by then, the market had evolved and many labs/tests were already entrenched.

### 2) The steward model is operationally heavy
An “open access” EUA steward must do what a kit manufacturer normally does, but without product revenue:
- maintain authorized labeling and version control  
- onboard and train labs (contracts, verification, quality expectations)  
- add new instruments/reagents via bridging studies and EUA amendments  
- handle post‑market reporting, complaints, performance monitoring  
- coordinate communications across a network  

The SalivaDirect team’s retrospective emphasizes the amendment cadence and bridging work needed to keep expanding equipment/supply options ([7]).

### 3) Misaligned incentives and funding
Open protocols can reduce exclusivity. Without a durable funding source (public funding, philanthropy, fees, or a product business), a steward may be unable to sustain the governance load.

The nomination letter explicitly frames SalivaDirect’s cost compression as a public benefit—but that same dynamic can discourage commercial followers from replicating the model ([20]).

### 4) Technical scope mismatch with public demand shifts
PCR testing in high-complexity labs has different economics and logistics than:
- point-of-care molecular tests  
- at-home antigen tests  
- retail distribution models  

As COVID response shifted, the “open PCR protocol executed by CLIA high-complexity labs” became less central to the consumer-facing narrative, even if it remained important for surveillance and institutional programs.

### 5) Regulatory structure: authorization attaches to an entity
The requestor’s key point is structural: even if every component is disclosed, FDA authorizes a product to a legal entity. Without a mechanism like “designated labs” or a modular authorization framework, open-source publication alone cannot create broad open access.

Monaco & Ware (2022) essentially make this argument from a policy angle: open-source diagnostics face structural hurdles under existing approval models ([9]).

---

## Where things stand now (2023–2026): open protocols vs open access

### SalivaDirect today (as reflected in public materials through 2024–2026)
Public sources indicate SalivaDirect evolved beyond the original Yale EUA to multiple EUAs held by SalivaDirect, Inc., including provisions for direct-to-consumer / at-home collection methods, and continues to rely on designated laboratories and maintained lists of qualified reagents/instruments ([2], [7]).

### LDT oversight turbulence matters for “open access diagnostics”
From 2024–2025, FDA finalized a rule attempting to increase oversight of LDTs, followed by legal challenges and later reporting that FDA would rescind/roll back the rule after a court decision ([13], [14]). This volatility affects the practical landscape:

- If **LDTs remain primarily CLIA-governed**, “open protocol” dissemination may be the dominant path (labs implement locally).  
- If **FDA oversight expands**, there may be more demand for shared regulatory infrastructure—potentially increasing the attractiveness of stewarded open-access models (if legally supported).

### The “open access” gap remains
Even in 2026, the system still largely lacks a routine, non-emergency pathway equivalent to:
- **“one open dossier → many independent manufacturers/labs can legally operate under it”**  
without those labs/entities filing their own marketing submissions.

That is the crux of why the SalivaDirect-style model remains important as a regulatory design pattern: it shows a way to approximate “generic diagnostics” during emergencies by combining open protocol + centralized stewardship + lab designation ([7], [20]).

---

## A practical framework for future open access diagnostics

Below are actionable design patterns, each with tradeoffs.

### Pattern 1: Stewarded Open Protocol EUA (SalivaDirect model)
**Best for:** rapid scale in high-complexity labs using commodity equipment, with supply-chain flexibility.  
**Key requirements:** a capable steward + funding + governance ([7]).

**Enhancements to consider (policy or program design):**
- Public funding for stewardship activities (amendments, bridging, QA)  
- Shared national infrastructure: proficiency testing, reference panels, shared QA dashboards  
- Standardized contract templates and onboarding verification workflows

### Pattern 2: Government- or consortium-operated “reference protocol” with delegated execution
**Best for:** emergencies where equity, cost, and resilience matter more than exclusivity.  
**What changes:** steward role is institutionalized (CDC, NIH, or multi-stakeholder consortium), not dependent on one academic lab.

### Pattern 3: Modular authorization / open master file + right-of-reference
**Goal:** allow many entities to “plug into” an openly available validated dossier (or module) rather than repeat work.  
**Challenge:** today, right-of-reference systems often assume proprietary master files, not public-domain ones; FDA would need policy clarity.

### Pattern 4: Standards-based pathway (consensus methods + minimal bridging)
**Goal:** convert parts of “protocol validation” into standards (e.g., CLSI style), so following the standard reduces evidence burden.  
**Challenge:** standards are slow to develop, and outbreak dynamics are fast.

---

## The CDC test as the canonical "open" government diagnostic

### Why the CDC test matters in the open EUA context
The CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel is, in an important sense, the original "open" COVID diagnostic test in the United States. As a government-developed and government-distributed assay, it embodied several key features that later open-protocol EUAs like SalivaDirect would build on — but it also illustrates the structural limitations that prevented even the most widely available public-sector test from functioning as a truly open-access platform.

The CDC test was authorized for use by any CLIA high-complexity laboratory. Its primer and probe sequences (N1, N2, and the RP internal control) were publicly published. Reagents were distributed through the International Reagent Resource (IRR) and, over time, commercially manufactured primer/probe alternatives were accepted. By the December 2020 revision of the IFU, multiple extraction platforms from QIAGEN, Roche, Promega, and bioMérieux were authorized. In these respects, the CDC test scored well on "openness" — full disclosure of chemicals and reagents, publicly available primer sequences, and multi-vendor sourcing of key components.

However, the CDC test also reveals the limits of openness without intentional design for access and resilience:

- **Single instrument dependency.** The test was authorized only on the Applied Biosystems 7500 Fast Dx Real-Time PCR Instrument. Unlike SalivaDirect, which validated across multiple RT-PCR platforms, the CDC test remained locked to one instrument family, limiting which labs could adopt it and creating a supply constraint.
- **Distribution model, not designation model.** The CDC did not use a "designated laboratory" framework like SalivaDirect. Instead, reagents were distributed through IRR and commercial channels to any qualifying lab. This is simpler than stewardship-with-designation but also means there was no systematic onboarding, quality oversight, or network coordination among implementing labs.
- **Supply chain fragility despite openness.** Despite disclosing reagents and accepting multi-vendor extraction methods, the CDC test suffered acute supply chain failures — most catastrophically, the contaminated initial test kit batch in February 2020, followed by extraction reagent shortages that hobbled testing capacity nationwide for weeks.

### The initial CDC test kit failure
The CDC's initial test kit failure was one of the most consequential early failures of the U.S. pandemic response. The first batch of RT-PCR test kits was distributed to state and local public health laboratories beginning February 5, 2020. Within days, multiple laboratories reported false positive reactivity in negative controls. A CDC internal investigation (published in PLOS ONE, 2021) identified two separate problems ([24]):

1. **N1 component contamination**: The N1 reagent in the distributed kits was contaminated with synthetic genetic material (positive control plasmid), causing approximately 2% false positive results. The contamination was not present in pre-validation material and was determined to have occurred during the post-production quality control process or packaging. The FDA confirmed that CDC manufactured its test in one of its research laboratories rather than in its manufacturing facilities, and did not follow its own manufacturing protocols ([25]).
2. **N3 probe design flaw**: The N3 probe exhibited false positive reactivity because two molecules in the probe frequently bound to each other in the absence of any virus, triggering a false fluorescent signal. This problem worsened with the age of the kits, which is why early evaluation runs using fresh materials did not catch it.

The combined effect was that the only authorized COVID test in the United States produced unreliable results at precisely the moment when early detection and containment were most critical. The N3 component was eventually dropped from the panel entirely, and replacement kits were not available until late February. A 2023 HHS Office of Inspector General audit concluded that the failure resulted from "internal control weaknesses" in CDC's development processes and laboratory quality procedures, including inadequate separation of duties, insufficient documentation, and lack of independent expert review prior to distribution ([26]). The OIG further warned that without corrective action, CDC risked "damaging public trust, which could undermine its ability to accomplish its mission."

The three- to four-week delay during February 2020 — a period when community transmission was establishing itself across the United States — is widely regarded as having materially worsened the trajectory of the U.S. outbreak. And the irony from the open-access perspective is acute: here was a government-developed, openly disclosed test, free of commercial interests, that failed not because of any inherent problem with the "open" model but because of basic manufacturing quality failures at the agency itself.

### Appendix A and the extraction-free protocol restriction
One of the most consequential — and least discussed — features of the CDC EUA IFU is **Appendix A: "Heat Treatment Alternative to Extraction."** In response to the global extraction reagent shortage, CDC validated a simple heat treatment protocol (95°C for 1 minute) using the Quantabio UltraPlex 1-Step ToughMix (4X) as a specimen processing alternative that bypassed nucleic acid extraction entirely. CDC's own data showed comparable detection performance between the extraction-based and heat-treatment methods.

This was a significant technical achievement: a direct, extraction-free protocol validated against the flagship government test, requiring only a thermal cycler and basic consumables — no complex extraction instruments or scarce extraction kits.

**But Appendix A carried a critical restriction: "This procedure is only for use by public health laboratories."**

An initial restriction to public health labs was arguably reasonable — CDC had validated the heat treatment method with a limited number of clinical specimens, acknowledged a potential reduction in sensitivity, and stated it should only be used "when a jurisdiction determines that the testing need is great enough to justify the risk of a potential loss of sensitivity." Starting with public health labs (which have closer institutional ties to CDC and more experience with method validation under emergency conditions) while further qualifying the protocol was defensible as a cautious first step.

But the restriction should have been temporary — a matter of weeks, not permanent. Once the method was further qualified (and CDC's own performance data showed comparable detection), the restriction should have been lifted to allow any authorized CLIA high-complexity lab to use the extraction-free protocol, especially given that the entire rationale for the method was a reagent shortage affecting *all* labs, not just public health labs.

Instead, the restriction appears never to have been lifted — the FDA-hosted March 7, 2023 revision of the CDC IFU still contains the identical "only for use by public health laboratories" language ([23]). The practical effect was that even as extraction reagent shortages crippled testing capacity at clinical laboratories across the country, the validated extraction-free alternative remained restricted to the narrower tier of public health laboratories. CLIA high-complexity labs — which constituted the bulk of the authorized testing network and clinical lab infrastructure in the United States — could not use it.

This represents a compounding narrowing of the testing funnel:
1. FDA EUAs generally restricted authorized COVID testing to CLIA high-complexity clinical laboratories, excluding the vast landscape of potential testing settings (schools, workplaces, community health organizations).
2. The CDC test was the most broadly available authorized assay within that clinical lab universe.
3. But the extraction-free simplification — the very innovation that could have helped labs most affected by supply shortages — was restricted to an even narrower subset: public health labs only.

Meanwhile, extraction-free protocols became one of the most important areas of innovation during the pandemic: SalivaDirect's extraction-free, multi-platform approach was a central reason for its scalability and supply-chain resilience, and many other EUA-authorized tests adopted extraction-free or direct-sample protocols. The contrast is stark — an academic lab (Yale) built an open, extraction-free platform that scaled to over 200 labs and 6.5 million tests, while the government's own flagship test validated the same general approach but locked it away from the labs that needed it most.

### The broader context: testing as a public health problem, not just a clinical one
The Appendix A restriction is symptomatic of a deeper structural problem that the pandemic exposed. COVID-19 testing was framed and regulated as a **clinical diagnostic activity** — individual patients, healthcare provider orders, CLIA-certified laboratories, FDA-authorized test systems. But the actual need was for **public health screening at population scale**: schools, workplaces, airports, community settings, congregate living facilities.

This mismatch between the regulatory architecture (designed for clinical medicine) and the operational need (public health surveillance and screening) meant that:
- Testing capacity was bottlenecked through the clinical laboratory system, which could not scale fast enough.
- Innovations that could have democratized testing — simpler protocols, extraction-free methods, LAMP-based assays, direct sample testing — faced regulatory barriers calibrated to clinical diagnostic standards rather than public health screening realities.
- Even when validated simplifications existed (as with the CDC's own heat treatment protocol), institutional caution and regulatory framing restricted their use rather than expanding it.

The pandemic demonstrated a need for a **distinct regulatory regime for non-medical testing and screening** — one that could authorize rapid deployment of validated, simplified protocols to a much broader range of settings and operators, without requiring each to operate as a CLIA high-complexity laboratory. The open EUA concept, as pioneered by SalivaDirect and aspired to by FloodLAMP, was an attempt to build something like this within the existing framework. But the CDC Appendix A example shows that even when the technical barrier was solved, the institutional and regulatory barriers remained.

### The regulatory asymmetry: why only the government can create a "universal" test
The CDC test reveals a striking structural asymmetry in the U.S. diagnostic regulatory system. The CDC — a government agency — was ultimately able to create a test that any CLIA high-complexity laboratory in the country could run, simply by obtaining the reagents. No stewardship model. No lab-by-lab designation. (Notably, the *earliest* version of the CDC EUA did limit testing to "qualified laboratories designated by CDC," but later reissued authorization letters broadened this to all CLIA high-complexity labs.) The EUA authorized the test for use by all qualifying labs, the reagents were distributed through the International Reagent Resource and commercial channels, and the protocol was fully public. CDC also provided a distinct form of openness through broad **Right of Reference** (RoR) to its performance data, which other EUA applicants could leverage when seeking their own authorizations — making the CDC assay function as a kind of regulatory "commons" ([4]). In effect, the CDC model was a **public reference protocol with a constrained authorized bill-of-materials**: fully disclosed, broadly available, but sourcing was gated to CDC-approved lots and channels. And even this model was available only to a government agency.

No other entity — academic, nonprofit, or commercial — has access to this pathway. When Yale developed SalivaDirect with comparable openness (fully disclosed protocol, publicly available primer sequences, commodity reagents, multiple validated instrument/reagent combinations), it could not simply publish and distribute. FDA required a stewardship and designation model: Yale (later SalivaDirect, Inc.) had to hold the EUA, designate each participating lab, maintain oversight, manage amendments and bridging studies, and bear the ongoing governance burden. This is the "open access" model described earlier in this report — a genuine innovation in regulatory design, but operationally heavy and dependent on a single steward's capacity and funding.

FloodLAMP pursued the same open-protocol EUA approach and unable not only to obtain an EUA authorization, but to get significant engagement from the FDA despite many efforts (see regulatory/correspondence), including presenting at the Reagan-Udall Foundation for the FDA (see archive file: `various/fl-presentations/FDA Reagan Udall Foundation - FloodLAMP Presentation (4-21-2022).md`). Beyond SalivaDirect, no other organization appears to have successfully replicated the open-protocol/open-access EUA pattern during COVID.

The asymmetry raises a fundamental question: **What justifies the assumption that only a government agency can produce a test reliable enough for unsupervised use by CLIA high-complexity laboratories?**

These are, by definition, laboratories that have demonstrated competence to perform the most complex diagnostic testing. Under CLIA, they are inspected, proficiency-tested, and staffed with qualified personnel. They have the technical capability to validate and run sophisticated molecular assays — and, critically, they have long done exactly this through laboratory-developed tests (LDTs). Granted, only a minority of CLIA labs have the capability to _develop_ a new LDT, but all can _run_ already developed tests, provided the test doesn't require special equipment or isn't unusually complex. Legally, a CLIA high-complexity lab can independently develop, validate, and clinically deploy its own molecular diagnostic test without any FDA premarket authorization, relying on its own scientific judgment and CLIA-regulated quality systems.

This makes the asymmetry even more paradoxical: FDA's regulatory framework simultaneously (a) trusts CLIA high-complexity labs to develop their own tests from scratch as LDTs, but (b) does not trust them to implement someone else's fully disclosed, validated, open protocol without a steward standing between them and the authorization. A lab that could design, validate, and deploy its own RT-PCR test targeting the same viral sequences apparently could not be trusted to follow a published protocol from an academic group — unless that group held an EUA and formally "designated" the lab.

The tension deepened in 2024 when FDA finalized a rule attempting to extend premarket review authority over LDTs — effectively seeking to bring the same CLIA high-complexity labs that had been independently developing and running their own tests under FDA's device-regulatory apparatus. This was a direct challenge to the autonomy that CLIA labs had exercised for decades. A federal court struck down the rule, and FDA subsequently moved to rescind it ([13], [14], see `regulatory/ldts/_AI_FDA 2024 LDT Rule - Status and Legal History.md`). The court's decision effectively affirmed that these labs possess the competence and regulatory standing to operate independently — which only sharpens the question of why non-governmental open protocols cannot be deployed to them without the overhead of a stewardship/designation wrapper.

The most plausible explanation is institutional, not technical: the FDA's product-centric regulatory architecture requires that *someone* be legally accountable for each authorized test system. For a commercial kit, the manufacturer is accountable. For the CDC test, the CDC is accountable. For an open protocol from an academic lab, there is no manufacturer — so FDA requires a steward to fill the accountability role. This is a reasonable structural concern, but the pandemic demonstrated that it can function as a bottleneck that prevents good, validated, open science from reaching the labs that need it. A regulatory framework that recognized the competence of CLIA high-complexity labs — the same competence that underlies the entire LDT system — could have enabled a lighter-weight pathway for deploying open protocols at scale, without requiring each one to build a bespoke stewardship infrastructure.


## Comparative table: openness characteristics of COVID-19 EUA diagnostic tests

### Original table (from FloodLAMP presentation)
slide 12 from `various/fl-presentations/BARDA Market Research - FloodLAMP Presentation (2022-03-07).md`
The following table compares openness characteristics across six categories of COVID-19 diagnostic tests, from a typical proprietary IVD EUA through several open or partially-open protocols, including both FloodLAMP submissions.

| | Typical IVD EUA | CDC EUA | SalivaDirect™ | SHIELD | FloodLAMP EasyPCR™ | FloodLAMP QuickColor™ |
|---|---|---|---|---|---|---|
| Disclosure of all chemicals and reagents? | No | Yes | Yes | Yes | Yes | Yes |
| Chemical and reagents available from multiple vendors? | No | Yes | Yes | No | Yes | Yes |
| Disclosure of primer sequences? | No | Yes (std for PCR) | Yes | No (ProptryThermo) | Yes | Yes |
| Primers commercially available from multiple vendors? | No | Yes | Yes (CDC Primers) | No (ProptryThermo) | Yes (CDC SD Primers) | Yes (Available but not launched) |
| Supply chain robust? | No | No/Maybe | Yes | No/Maybe | Yes | Yes |
| EUA Sponsor Organization Type | For Profit Company | Govt | Academic Not for Profit | Academic Not/For Profit ? | Public Benefit Corp | Public Benefit Corp |
| Designation of CLIA labs | Kit Sales | N/A (open RoR) | Impact & Expansion | Impact & Expansion | Impact & Expansion | Impact & Expansion |

### Final consolidated table
Below is the final revised version of the table, incorporating review from two independent AI analyses of the underlying EUA documents and public sources. Where both analyses agreed on a change, the change was adopted. Where they differed, a decision was made with rationale noted below.

| | Typical IVD EUA | CDC EUA | SalivaDirect™ | SHIELD | FloodLAMP EasyPCR™ | FloodLAMP QuickColor™ |
|---|---|---|---|---|---|---|
| Disclosure of all chemicals and reagents? | No | Yes | Yes | Partial (kit-level) | Yes | Yes |
| Chemical and reagents available from multiple vendors? | No | Yes* (by Dec 2020) | Yes | No | Yes | Yes |
| Disclosure of primer sequences? | No | Yes (published sequences) | Yes | No (proprietary ThermoFisher) | Yes | Yes |
| Primers commercially available from multiple vendors? | No | Yes, constrained (IRR + CDC-accepted lots) | Yes (CDC primers; multi-supplier validated) | No (proprietary ThermoFisher) | Yes (CDC/SD primers) | Yes |
| Supply chain robust? | No | No/Maybe | Yes | No/Maybe | Yes | Yes |
| EUA Sponsor Organization Type | For Profit Company | Govt | Academic Not for Profit | Academic / public university | Public Benefit Corp | Public Benefit Corp |
| Designation of CLIA labs | Kit Sales | No designation step; open to all CLIA HC labs | Impact & Expansion (stewarded designation) | Impact & Expansion (designated labs) | Impact & Expansion | Impact & Expansion |

\* CDC multi-vendor support is constrained: primer/probe and control materials must be from IRR or specific CDC-posted acceptable commercial lots — not "any vendor." Extraction methods have broader but still enumerated multi-vendor support.

### Annotations on changes from original table (with decision rationale)

**Changes where both reviews agreed (adopted):**

1. **SHIELD — "Disclosure of all chemicals and reagents"** changed from "Yes" to **"Partial (kit-level)"**: The covidSHIELD protocol specified which products to purchase (ThermoFisher TaqPath COVID-19 RT-PCR Kit), but the underlying primer/probe/reagent formulations within that kit were proprietary. This is a meaningful distinction from truly open protocols like CDC or SalivaDirect where the reagent compositions are fully disclosed. *(One review changed this; the other agreed in its narrative that SHIELD was "less open" due to proprietary Thermo components but didn't change the cell.)*
2. **CDC EUA — "Disclosure of primer sequences"** annotation changed from "std for PCR" to **"published sequences"**: The N1, N2, and RP sequences were newly designed for SARS-CoV-2 and publicly published; they were not pre-existing "standard" PCR primers. Both reviews agreed on this point.
3. **SHIELD — "Disclosure of primer sequences" and "Primers commercially available"** annotation expanded from "ProptryThermo" to **"proprietary ThermoFisher"** for clarity. Both reviews made this same change.

**Changes where reviews differed (decision with rationale):**

4. **CDC EUA — "Chemicals and reagents available from multiple vendors"**: One review used "Yes (by Dec 2020)"; the other used "Yes*" with a footnote about constrained sourcing. **Decision**: Merged both — **"Yes* (by Dec 2020)"** — because both qualifications are valid. The multi-vendor extraction support was not present at initial authorization (Feb 2020) and was added through subsequent IFU revisions. And even after expansion, primer/probe sourcing remained gated to IRR-distributed or CDC-posted acceptable lots, which is meaningfully more constrained than SalivaDirect's open sourcing from any vendor.
5. **CDC EUA — "Primers commercially available from multiple vendors"**: One review used "Yes (IRR + commercial)"; the other used "Limited (IRR + CDC-accepted lots only)." **Decision**: Adopted **"Yes, constrained (IRR + CDC-accepted lots)"** as a middle ground. The CDC IFU explicitly states that "only material distributed through the CDC International Reagent Resource and specific lots of material posted to the CDC website are acceptable for use with this assay under CDC's Emergency Use Authorization." Labs cannot synthesize the publicly known sequences and use them under the EUA. This is meaningfully more constrained than a simple "Yes" but "Limited" alone understates that multiple sources do exist — the constraint is on *which lots from which vendors*, not on the number of sources.
6. **SHIELD — "EUA Sponsor Organization Type"**: One review kept "Academic Not/For Profit ?" (original); the other changed to "Academic / public university." **Decision**: Adopted **"Academic / public university"** — the EUA sponsor was the University of Illinois, which is unambiguously a public university. The "?" about for-profit status related to Shield T3 LLC (an operational entity), but the authorization holder was the university itself.
7. **CDC EUA — "Designation of CLIA labs"**: One review used "N/A (open IRR distribution)"; the other used a verbose multi-clause description. **Decision**: Adopted **"No designation step; open to all CLIA HC labs"** as a concise synthesis. The CDC did not use a stewarded designation model like SalivaDirect; any qualifying CLIA high-complexity lab could obtain and run the test. (Note: the *earliest* CDC EUA letter did limit testing to "qualified laboratories designated by CDC" before later revisions broadened access.) The original table's "RoR" annotation referred to Right of Reference — CDC's provision of performance data to other EUA applicants — which is a distinct concept from lab designation and is better discussed in narrative context than in this cell.
8. **SalivaDirect / SHIELD — "Designation of CLIA labs"**: One review kept the original "Impact & Expansion" wording; the other used more technically descriptive labels ("Stewarded designation by EUA holder" / "Designated labs under EUA holder"). **Decision**: Kept the original **"Impact & Expansion"** framing (reflecting the user's characterization of purpose/intent) but added parenthetical mechanism descriptions for clarity.
9. **FloodLAMP QuickColor — "Primers commercially available"**: One review simplified from "Available but not launched" to "Yes"; the other kept "Yes (available but not launched)." **Decision**: Adopted **"Yes"** — primer sequences were disclosed and available for custom synthesis. Product launch status is a separate issue from primer availability and is better addressed in narrative context.


## Appendices

### Appendix A — Key terms (glossary)
| Term | Practical meaning in this context |
|---|---|
| **EUA** | Emergency Use Authorization under FD&C Act §564 for medical countermeasures during declared emergencies |
| **IVD** | In vitro diagnostic test (device) |
| **LDT** | Laboratory-developed test; historically under FDA enforcement discretion, primarily regulated via CLIA for lab operations |
| **Open protocol** | Protocol disclosed in enough detail for independent implementation |
| **Open access (regulatory)** | Multiple independent entities can legally operate under a shared authorization without each filing a full application |
| **Steward** | The entity that holds the authorization and manages changes, onboarding, and quality/labeling obligations |

### Appendix B — “Openness” checklists for future projects
**Open protocol checklist**
- [ ] Complete workflow: pre-analytical → analytical → reporting  
- [ ] Bill of materials with part numbers and acceptable alternatives  
- [ ] Instrument-specific parameters  
- [ ] QC/controls guidance (positive/negative, internal controls)  
- [ ] Versioning and change log

**Open access checklist**
- [ ] A legal mechanism for multi-entity execution (designation / licensing / umbrella)  
- [ ] Onboarding, verification, and audit plan  
- [ ] Governance for amendments (who tests, who approves, how to bridge)  
- [ ] Funding model for ongoing stewardship  
- [ ] Post-market reporting and complaint handling plan

---

## References

> URLs are provided for convenience. Many FDA sources are PDFs; some links may change over time if FDA reorganizes pages.

### Primary documents (FDA, authorization artifacts)
[1] FDA press release (Aug 15, 2020): “FDA Issues Emergency Use Authorization to Yale School of Public Health for SalivaDirect…”  
https://www.fda.gov/news-events/press-announcements/coronavirus-covid-19-update-fda-issues-emergency-use-authorization-yale-school-public-health

[2] FDA: SalivaDirect — Letter of Authorization (PDF; updated; held by SalivaDirect, Inc.)  
https://www.fda.gov/media/141194/download

[3] FDA: SalivaDirect — EUA Summary (PDF)  
https://www.fda.gov/media/141192/download

[4] FDA: In Vitro Diagnostics EUAs (index page; includes tables, templates, umbrella EUAs)  
https://www.fda.gov/medical-devices/covid-19-emergency-use-authorizations-medical-devices/in-vitro-diagnostics-euas

[5] FDA: Molecular Diagnostic Tests for SARS‑CoV‑2 (table; includes note on designated labs)  
https://www.fda.gov/medical-devices/covid-19-emergency-use-authorizations-medical-devices/in-vitro-diagnostics-euas-molecular-diagnostic-tests-sars-cov-2

[6] FDA: Deliverable 15 — Emergency Use Authorization Assessment: Final Report (PDF)  
https://www.fda.gov/media/152992/download

### SalivaDirect scientific & retrospective literature
[7] Wyllie et al. / SalivaDirect team (2024): “Scalable solutions for global health: the SalivaDirect model” (PMC)  
https://pmc.ncbi.nlm.nih.gov/articles/PMC11554656/

[8] Vogels et al. (2020): SalivaDirect protocol paper (PMC)  
https://pmc.ncbi.nlm.nih.gov/articles/PMC7418705/

### Open-source diagnostics policy / models
[9] Monaco & Ware (2022): “Structural Challenges in Deployment of an Open-Source Diagnostic Product”  
https://theoryandpractice.citizenscienceassociation.org/articles/10.5334/cstp.530

[10] Emperador et al. (2020): “An open-source molecular diagnostic platform approach for outbreak and epidemic preparedness” (PMC)  
https://pmc.ncbi.nlm.nih.gov/articles/PMC7564747/

### Policy context: LDT oversight and related developments
[11] American Hospital Association (Aug 21, 2020): HHS announcement on FDA premarket review of LDTs  
https://www.aha.org/special-bulletin/2020-08-21-special-bulletin-trump-administration-blocks-fda-requiring-labs-submit-coronavirus-tests-review

[12] RAPS (Nov 15, 2021): FDA revises COVID test approach after HHS reverses LDT policy  
https://www.raps.org/news-and-articles/news-articles/2021/11/fda-revises-covid-test-approach-after-hhs-reverses

[13] FDA: “Laboratory Developed Tests” (page reflecting 2024–2025 developments and updates)  
https://www.fda.gov/medical-devices/vitro-diagnostics/laboratory-developed-tests

[14] Reuters (Sep 18, 2025): FDA to roll back LDT oversight rule following court decision  
https://www.reuters.com/legal/litigation/us-fda-roll-back-lab-developed-test-oversight-rule-following-court-decision-2025-09-18/

### Representative press coverage focusing on “open source” aspect
[15] ContagionLive (Aug 15, 2020): “FDA Grants Emergency COVID-19 Authorization to Yale's Open Source Method…”  
https://www.contagionlive.com/view/fda-grants-emergency-covid19-authorization-yale-open-source-method-saliva-testing

[16] STAT (Aug 15, 2020): Yale’s saliva test / open approach coverage  
https://www.statnews.com/2020/08/15/yales-saliva-test-gets-emergency-authorization-a-step-toward-cheaper-more-rapid-covid-19-testing/

[17] TIME (Oct 13, 2020): “This 3‑Dollar COVID‑19 Test Could Do More…”  
https://time.com/5899765/saliva-covid-19-testing/

[18] LabMedica (Aug 17, 2020): Open‑source saliva PCR test coverage  
https://www.labmedica.com/covid-19/articles/294784114/new-open-source-saliva-based-covid-19-test-can-be-deployed-in-clinical-laboratories-across-us.html

[19] SelectScience (interview, ~2022): “Saliva-based open-source tests for COVID-19…”  
https://www.selectscience.net/article/saliva-based-open-source-tests-for-covid-19-how-saliva-sampling-improves-compliance-saves-costs-and-relieves-the-strained-supply-chain

### Additional related sources used for specific points
[22] COVID‑19 Diagnostics Evidence Accelerator (May 19, 2022): Meeting 46 Summary (PDF) — includes FloodLAMP statement re: open-source protocol EUA submissions  
https://evidenceaccelerator.org/sites/default/files/2022-05/COVID%2019%20Diagnostics%20Evidence%20Accelerator%20Meeting%2046%20Summary_Final.pdf

### CDC test (EUA IFU and related)
[23] CDC: "CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel — Instructions for Use"  
Catalog # 2019-nCoVEUA-01. Revision 06 dated Dec 1, 2020; later revisions through at least March 7, 2023 retain the Appendix A public-health-lab-only restriction. Includes Appendix A (Heat Treatment Alternative to Extraction, restricted to public health laboratories), Appendix B (Pooled Specimen Preparation), Appendix C (Implementation and Monitoring of Pooled Specimen Testing).  
IFU PDF (FDA-hosted): https://www.fda.gov/media/134922/download  
EUA Letter of Authorization: https://www.fda.gov/media/134919/download  
EUA index page: https://www.fda.gov/medical-devices/covid-19-emergency-use-authorizations-medical-devices/in-vitro-diagnostics-euas

### CDC initial test kit failure
[24] Kondor et al. (2021): "CDC 2019 Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel: Root Cause Analysis" (PLOS ONE)  
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0260487

[25] CNN (June 20, 2020): "Early CDC test kits were delayed because of contamination issues, HHS report affirms"  
https://www.cnn.com/2020/06/20/politics/cdc-test-kits-contamination/index.html

[26] HHS Office of Inspector General (Oct 24, 2023): "CDC's Internal Control Weaknesses Led to Its Initial COVID-19 Test Kit Failure, but CDC Ultimately Created a Working Test Kit" (Report A-04-20-02027)  
https://oig.hhs.gov/reports/all/2023/cdcs-internal-control-weaknesses-led-to-its-initial-covid-19-test-kit-failure-but-cdc-ultimately-created-a-working-test-kit/

### Attached documents provided by the requestor (local)
[20] “Anne Wyllie Nomination for FDA Foundation 2022 Innovations in Regulatory Science Awards” (June 3, 2022) — provided as attached markdown  
[21] “FDA Issues EUA for SalivaDirect” (FDA press release copy, Aug 15, 2020) — provided as attached markdown