The Regulatory Framework

How federal, state, and institutional rules create the system that requires beagles

Overview Soft Law IACUC GLP State Enforcement Gaps

Accountability Gaps in the Regulatory System

The United States has a multi-layered system governing animals in research — federal statutes, agency regulations, institutional committees, voluntary accreditation, and state laws. On paper it looks comprehensive. In practice, each layer contains structural gaps that compound across the system, creating zones of near-zero accountability for the animals inside them.

~95%

Lab animals excluded from AWA

Rats, mice, and birds bred for research

Source: AWA definition, 7 U.S.C. §2132(g)

80%

Dog breeders with violations

USDA OIG audit, 2025

Source: USDA OIG Report 33601-0001-22

~120

USDA inspectors nationwide

Covering 12,000+ licensed facilities

Source: USDA APHIS staffing estimates

AWA limits on research procedures

Only housing and handling regulated

Source: 7 U.S.C. §2143(a)(6)(A)

USDA Enforcement: Underfunded, Understaffed, Under-Penalized

The Animal and Plant Health Inspection Service (APHIS) within USDA is the primary federal enforcer of the Animal Welfare Act. Three Inspector General audits over 15 years have documented the same structural failures — each time recommending improvements, each time finding that prior recommendations were not fully implemented.

Inspector Shortages

Roughly 120 USDA Animal Care inspectors cover more than 12,000 licensed facilities across the country — breeders, dealers, exhibitors, carriers, and research facilities. Inspections are supposed to be annual for research facilities, but a Congressional Research Service report documents litigation alleging USDA failed to meet annual inspection requirements for AAALAC-accredited research facilities since 2019. Risk-based scheduling means compliant facilities get fewer visits, but “compliant” is determined by prior inspections from the same stretched workforce.

Penalty Weakness

The 2025 USDA OIG audit explicitly clarified that official warnings are not penalties and not enforcement actions — they are “regulatory correspondence intended to promote compliance.” Yet warnings are the most common federal response to documented violations. Facilities with 30, 40, even 50 citations over a decade may receive nothing more than a warning letter. The Animal Welfare Institute found that AWA enforcement has deteriorated further following a Supreme Court decision affecting agencies' civil penalty authority, with fines already small and routinely reduced through settlement.

The Education-Over-Enforcement Model

The 2010 OIG audit documented APHIS's systematic preference for “education and cooperation” over deterrence, finding that violators repeatedly violated the law after re-inspection. Inspectors failed to consistently cite repeat and direct violations. Penalty calculations yielded minimal deterrent fines. Fifteen years later, the 2025 OIG audit found 80% of visited dog breeders still had at least one noncompliance, including repeat items.

Key Finding

When enforcement does happen, it clusters around extreme, well-documented breakdowns where multiple federal interests align. The Envigo prosecution — 4,000+ beagles surrendered, the largest AWA fine in history — only occurred after years of documented mistreatment that USDA officials allegedly failed to act on. Reuters reported a federal probe scrutinizing why senior USDA animal welfare officials took limited action despite repeated documentation.

The AWA's Structural Blind Spot: Housing Without Science

The Animal Welfare Act contains a provision that most people outside the regulatory system do not know about. Section 2143(a)(6)(A) of Title 7 explicitly prohibits the federal government from prescribing standards for the “design, outlines, or guidelines of actual research or experimentation.”

What the AWA Regulates

• Minimum cage sizes and enclosure standards

• Temperature, ventilation, and lighting

• Feeding and watering schedules

• Veterinary care for illness and injury

• Exercise requirements for dogs

• Record-keeping and annual reporting

• Facility registration and licensing

What the AWA Cannot Touch

• Whether a study should use animals at all

• Whether alternatives exist that could replace animal use

• The scientific merit or necessity of a protocol

• Pain endpoints — how much suffering is permitted

• How many animals are used in a study

• Whether results justify the harm inflicted

• Post-study outcomes (euthanasia vs. adoption)

Data Gap

This means USDA can cite a facility for a cage that is two inches too small, but has zero authority to question whether the experiment in that cage should be happening at all. A beagle can be subjected to a year-long toxicity test that causes organ failure, vomiting, and seizures — and as long as the cage meets size requirements and a veterinarian signs off, the AWA is satisfied.

IACUC Independence Failures

Institutional Animal Care and Use Committees (IACUCs) are positioned as the primary ethical gatekeepers for animal research. Required under both the AWA and PHS Policy, they review protocols, conduct facility inspections, and are supposed to ensure that pain and distress are minimized. The structural problem is that IACUCs operate inside the institutions they oversee.

Institutional Capture

IACUC members are appointed by institutional officials who also depend on research grants for revenue. The committee includes at least one “unaffiliated” member, but that person is still selected by the institution. Scholarly critique describes “mission lapse” dynamics where committees rely on trust rather than robust, independent scientific review — especially where mandates are ambiguous about what constitutes adequate justification for animal use.

Self-Reporting to OLAW

The Office of Laboratory Animal Welfare (OLAW) oversees PHS-funded research, but its model depends on institutions reporting their own noncompliance. OLAW evaluates reports and determines follow-up, but the system inherently depends on the institution surfacing problems. A GAO report on NIH oversight of animal use described an oversight sequence that depends substantially on internal reporting and referral mechanisms.

No Public Accountability

IACUC protocol reviews are not public. Minutes are not published. Dissenting opinions are not disclosed. Pain categories are self-reported by the institution. The most consequential decisions about animal welfare — which experiments are approved, what pain levels are authorized, whether alternatives were genuinely considered — happen behind closed doors with no external audit.

AAALAC: Voluntary Accreditation, Not Enforcement

The Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) is frequently cited by research institutions as evidence of high standards. Facilities point to AAALAC accreditation the way hotels display star ratings. But AAALAC is a private, voluntary accreditation body — not a regulator.

Voluntary

No facility is required to seek AAALAC accreditation

Industry-Funded

Accredited facilities pay fees; peer reviewers come from similar institutions

No Penalty Power

AAALAC can revoke accreditation but cannot fine, shut down, or prosecute

Methodology Caveat

A Congressional Research Service report documents litigation alleging that USDA failed to meet annual inspection requirements specifically for AAALAC-accredited research facilities since 2019 — suggesting that AAALAC status may have functioned as a reason for less government oversight, not more. Rise for Animals' multi-year analyses show that AAALAC-accredited facilities still accumulate serious AWA citations.

The Rats, Mice, and Birds Exclusion

The single largest gap in the Animal Welfare Act is definitional. In 2002, a farm bill amendment formally excluded rats of the genus Rattus, mice of the genus Mus, and birds bred for use in research from the AWA's definition of “animal.” This exclusion means approximately 95% of all animals used in U.S. laboratories have no federal legal protection under the AWA.

What exclusion means in practice

• No USDA registration required for facilities using only excluded species

• No federal inspections of housing, care, or handling

• No required annual reports on numbers used or pain categories

• No minimum cage sizes, enrichment, or veterinary care standards

• No USDA enforcement authority over violations

• No public transparency about how many are used or what happens to them

• PHS Policy applies only if the facility receives federal funding

• Private companies using only rats/mice operate in a near-total regulatory vacuum

Why This Matters

This exclusion is directly relevant to the beagle question. When advocates push to reduce dog use in research, the industry can shift testing to rats and mice — species with even fewer protections. The exclusion creates a perverse incentive: replacing dogs with species that have zero federal oversight can be framed as “progress” while the total volume of animal suffering may not decrease at all.

State-Level Enforcement Vacuum

State anti-cruelty laws are often assumed to provide a backstop when federal enforcement fails. In practice, research laboratories exist in a legal safe harbor that most state cruelty statutes cannot reach.

Explicit Research Exemptions

Many state anti-cruelty statutes explicitly exempt research conducted at licensed or registered facilities. The conduct that would be a felony if done to a pet dog in a backyard is legally permitted when done to a beagle in a laboratory — as long as the facility holds a USDA license.

Judicial Implied Exemptions

Even where statutes do not explicitly exempt research, courts may hold that state cruelty law is inapplicable to federally regulated research. The Maryland high court's decision in Taub v. State is widely cited as a template: state cruelty law was held inapplicable to a federally governed research context, substantially weakening criminal accountability as a backstop.

Only Two States Run Their Own Programs

Massachusetts requires research institutions to be licensed to use dogs or cats in research, with unusually explicit experiment-design principles (use alternatives when appropriate, avoid unnecessary suffering, terminate if continuation may cause unnecessary injury). New York operates a Laboratory Animal Welfare Program requiring certificates for institutions using living animals. These two states are the exception. The other 48 rely almost entirely on the federal floor.

Bottom-Ranked States: Federal Floor Only

The Animal Legal Defense Fund's 2025 rankings list the bottom five states for animal protection as North Dakota, Alabama, Idaho, Kentucky, and Mississippi. None are among the 16 “Beagle Freedom” adoption-law states. In several, shelter-transfer governance is entirely municipal, creating highly uneven local outcomes. Oklahoma still has a statute directing delivery of shelter animals for scientific purposes on demand. These states rely entirely on USDA — the same agency whose own Inspector General has documented repeated enforcement failures.

International Gaps: A Global System With No Global Enforcer

Animal testing requirements are not just American. The regulatory framework is a global system with American, European, and Japanese components that reinforce each other through harmonized guidelines — but enforcement remains purely national.

The Harmonization Trap

• ICH M3(R2) harmonizes nonclinical safety study requirements across FDA, EMA, and PMDA

• OECD Test Guideline 409 — the guideline that names beagles — is adopted by 38 member countries

• A company developing a drug for the US, EU, and Japanese markets must satisfy all three regulators

• Changing one country's requirements without changing the others creates competitive disadvantage, not reform

Import and Transport Gaps

• CDC governs dog imports under 42 CFR §71.51, focused on rabies risk, not welfare

• APHIS veterinary services rules address disease certifications, not transport conditions

• Air transport welfare relies on IATA Live Animals Regulations — an industry standard, not law

• No single, unified federal transport-welfare regulator exists outside the AWA framework

• Dogs bred in countries with weaker protections can be imported into the US research pipeline

Data Gap

The EU's Directive 2010/63 is more comprehensive than US law in several ways — it covers all vertebrates and cephalopods, requires severity classifications, and mandates retrospective review. But even the EU cannot enforce its standards on studies conducted in the US, China, or India for products sold in European markets. The global regulatory system creates demand for animal testing that no single country's reforms can eliminate.

What Reform Would Look Like

Understanding the gaps reveals what meaningful reform would require. Incremental fixes to any single layer are insufficient because the failures are structural and compound across layers.

1. Close the AWA Definition Gap

Include rats, mice, and birds in the AWA's definition of “animal.” This single change would bring ~95% of lab animals under federal minimum standards, require USDA registration, mandate annual reporting, and create enforcement authority where none currently exists. It would also prevent the shell game of replacing dogs with species that have zero federal oversight.

2. Remove the Research Procedure Firewall

Amend Section 2143(a)(6)(A) to allow federal standards on research design where significant animal suffering is involved. This does not mean bureaucrats designing experiments — it means establishing enforceable floors on pain endpoints, mandatory use of alternatives when validated methods exist, and maximum suffering thresholds that cannot be overridden by institutional committees.

3. Fund Enforcement

Increase USDA Animal Care inspector staffing to levels where annual inspections are actually achievable. Restore meaningful civil penalty authority. Establish mandatory escalation from warnings to penalties after repeat violations. End the practice where facilities with dozens of citations receive nothing beyond regulatory correspondence.

4. Make IACUCs Independent

Require external IACUC members to constitute a majority, not a token seat. Mandate public disclosure of protocol approvals, pain category justifications, and alternatives considered. Create an external audit function — either through OLAW with real inspection authority or through a new independent body — so that ethical review does not depend on institutions policing themselves.

5. Eliminate State Cruelty Exemptions

Repeal research exemptions in state anti-cruelty statutes so that criminal law can function as a genuine backstop. Expand state-level research licensing programs beyond Massachusetts and New York. Require state-level inspection authority that operates independently of USDA.

6. Accelerate International Harmonization of Alternatives

Push for coordinated updates to ICH and OECD guidelines that accept validated non-animal methods as primary evidence, not just supplementary data. The FDA Modernization Act 2.0 removed the animal-testing mandate domestically; the same principle needs to be embedded in the international guidelines that drive global practice. Without this, US reforms simply shift testing overseas.

Why This Matters

None of these reforms are technically impossible. Massachusetts already licenses research facilities at the state level. The EU already covers all vertebrates. Independent ethics review already exists in human clinical trials. The barriers are political and economic, not structural. The system works exactly as it was designed to work — the question is whether we accept that design.

Signs of Movement

Despite the depth of these gaps, several developments suggest the regulatory landscape is shifting — though whether the pace of change matches the scale of the problem remains an open question.

Federal Agency Action

• FDA (April 2025): Announced plan to reduce/refine/replace animal testing using New Approach Methodologies

• FDA (March 2026): Draft guidance on NAMs in drug development with validation framework

• NIH (July 2025): Will no longer issue funding exclusively for animal models

• EPA: Continues framing NAMs as pathway to reduce vertebrate animal testing

• USDA (Feb 2026): Announced coordinated crackdown on chronic dog welfare violators with DOJ, DHS, and HHS

Legislative Momentum

• FDA Modernization Act 3.0 (S.355): Advanced in late 2025, replacing “animal testing” with “nonclinical testing” in regulations

• Violet's Law (H.R. 3246): Targets adoption/non-laboratory placement for federally owned research animals

• 16 states now have Beagle Freedom adoption laws, up from zero a decade ago

• Virginia (2023): First state to require public posting of USDA inspection/incident documents

• Maryland (2025): Proposed SB0535 to license animal testing facilities and create a state inspector role

Related Deep Dives

Animal Welfare Act7 tabs

7-tab deep dive: coverage gaps, pain categories, licensing, housing, enforcement, timeline, US vs EU

FDA Modernization Act4 tabs

The 1938 mandate, Act 2.0/3.0, two-species rule, why change is slow

Beagle Freedom Laws5 tabs

17 states, interactive map, state-by-state detail, after rescue

OECD Test Guideline 409

The single most important guideline — the one that names beagles

EU Directive 2010/63

The most comprehensive lab animal framework globally

Data & Transparency Gaps

What isn't tracked and why it matters

Sources

1. Animal Welfare Act, 7 U.S.C. §2131 et seq.; USDA APHIS regulations at 9 CFR.

2. AWA §2143(a)(6)(A) — prohibition on federal regulation of research design.

3. AWA definition exclusion: rats, mice, and birds bred for research, 7 U.S.C. §2132(g).

4. USDA OIG audit of dog breeder inspections (2025). Report 33601-0001-22. 80% noncompliance finding.

5. USDA OIG audit of problematic dealers (2010). Report 33002-4-SF. “Education and cooperation” preference.

6. USDA OIG audit of dog breeder oversight (2021). Data reliability/complaint-response findings.

7. CRS IF13002. AAALAC-accredited facility inspection shortfalls since 2019.

8. Animal Welfare Institute. AWA enforcement deterioration post-SCOTUS civil penalty ruling.

9. PHS Policy on Humane Care and Use of Laboratory Animals. OLAW, NIH.

10. Health Research Extension Act of 1985, 42 U.S.C. §289d. IACUC requirements.

11. GAO report on NIH oversight of animal use in foreign research. GAO-820-818631.

12. Scholarly critique of IACUC “mission lapse.” Bioethics journal analysis (2022).

13. Taub v. State (Maryland). State cruelty law held inapplicable to federally governed research.

14. Massachusetts M.G.L. ch. 140 §174D; 105 CMR 910.000. State research licensing.

15. New York Laboratory Animal Welfare Program. Wadsworth Center, NYSDOH.

16. ALDF 2025 State Rankings. Oregon #1, North Dakota #50.

17. DOJ, United States v. Envigo RMS, LLC. 4,000+ beagles; largest AWA fine.

18. Reuters (2023). Federal probe of Envigo scrutinizes top USDA officials' inaction.

19. ASPCA enforcement reporting, FY2022-2023. Warning letters as “regulatory communication.”

20. Rise for Animals. AAALAC violations analysis; multi-year inspection data.

21. FDA press announcement, April 2025. NAMs roadmap for reducing animal testing.

22. FDA draft guidance, March 2026. General considerations for NAMs in drug development.

23. NIH (July 2025). Funding to prioritize human-based research approaches.

24. FDA Modernization Act 3.0 (S.355), advanced late 2025.

25. Violet's Law (H.R. 3246). Federal research animal retirement/adoption.

26. Virginia §3.2-6593.1. Post-research adoption and transparency requirements.

27. USDA coordinated enforcement announcement, February 2026.

28. Data Liberation Project. APHIS inspection-report dataset, 2014-2026.

29. Bailing Out Benji. Quarterly USDA violations compilations.