Guide for Neuroscientists: Senate LHHS FY27 Outside Witness Testimony

Deadline: Close of business, Friday, May 22, 2026. Format: Four pages maximum, 8.5 x 11, single-sided, single-spaced, 1 inch margins, no cover page. Submit as Microsoft Word (not PDF) to lhhs@appro.senate.gov with subject line “FY27 OWT”. Include contact information (name, email, physical address, telephone) in the body of the email.

This guide is a companion to testimony I submitted for the FY 2027 cycle. The aim is to make it easy for neuroscience colleagues to produce their own testimony without starting from a blank page. Pick the sections and bullets that match your lab, your state, and your disease focus. Rewrite in your own voice. Volume of independent voices from working scientists is the goal, not uniformity. A BCI-specific appendix for neural interface colleagues is included at the end.

1. Why this matters and why you

Written testimony becomes part of the permanent Congressional Record. It provides evidentiary basis for Congressional oversight, litigation, and future policy. Appropriations staff read testimony when drafting report language. Individual PI voices carry weight that large-coalition letters (AAMC, FASEB, Research!America) do not, because appropriators need local, concrete, specific harms to justify bipartisan action to colleagues whose default is to cut discretionary spending. Your lab is the example.

Senate LHHS is chaired by Sen. Shelley Moore Capito (R-WV). Ranking member is Sen. Tammy Baldwin (D-WI). The FY 2026 bill they co-authored funded NIH at $48.7 billion, passed the Senate 71-29, and was enacted in March 2026. This is a bipartisan subcommittee with a demonstrated willingness to push back against administrative cuts. Arguments framed around medical progress, state economic impact, national security, and administrative decisions that undermine congressional intent are the vein that moves this specific committee.

2. The arguments that are working in this cycle

These arguments passed the bipartisan threshold for FY 2026 and remain active for FY 2027. Pick the one or two that fit your research program. Do not try to cover all of them, you will dilute the ask.

Argument 1: Top-line appropriation ($51.3 billion ask)

The Research!America ask for FY 2027 is $51.3 billion for NIH. United for Medical Research and the broader advocacy community are aligned on this number. Use it as the anchor. Adding your own number weakens the coalition signal.

Supporting data points you can cite

•       NIH FY 2025 budget: $47.49 billion. Requested FY 2027: $51.3 billion.

•       Every $1 invested in NIH research generated $2.57 in economic activity in FY 2025 (United for Medical Research, 2026 Annual Economic Impact Report, March 2026).

•       $94.15 billion in new economic activity and 390,863 jobs supported nationwide in FY 2025 (UMR 2026).

•       Over the past decade, NIH funding has generated $822 billion in economic activity and supported 3.7 million jobs (UMR 2026).

•       93% of Americans believe it is important for the President and Congress to assign a higher priority to medical progress (Research!America national survey, January 2026).

Argument 2: National security and the ecosystem advantage

China’s coordinated biomedical strategy has closed the gap in device engineering and overtaken U.S. regulatory speed. The remaining U.S. advantage is the depth of basic neuroscience and neurobiology built across hundreds of NIH-funded laboratories over two decades. That knowledge sets the performance ceiling for every device, therapy, and diagnostic in development in both countries. The genomics precedent is the correct frame: the Human Genome Project and GenBank were entirely open; China read every paper and trained scientists in U.S. laboratories; the U.S. won genomics anyway, because the competitive advantage accrued to whoever had the workforce, startup culture, and clinical infrastructure to act on the data. The knowledge was open. The ecosystem that could translate it was not replicable. This argument works because it is non-partisan and anchors to Capito’s national security orientation and to Baldwin’s Wisconsin biomedical manufacturing base.

Supporting data

•       The 10-15 year training cycle from graduate admission to independent investigator is the rate-limiting step no industrial policy can accelerate.

•       A training pipeline that breaks rather than pauses during funding collapses, because every Ph.D. student not admitted represents a 5-to-7-year gap that no subsequent appropriation can backfill.

•       Foreign-born scientists represent a substantial fraction of NIH-funded investigators and trainees. China and the EU are actively recruiting departing U.S. researchers.

Argument 3: Forward funding mechanically reduced the number of funded grants

This is the single most appropriations-actionable argument available. In FY 2025, NIH expanded its use of multi-year forward funding, obligating the full value of selected grants upfront rather than annually. The arithmetic consequence was 5,564 fewer grants than FY 2024, with R01 success rates falling to approximately 17%, the lowest in nearly thirty years. The House FY 2026 appropriations bill blocked further expansion of forward funding. The Senate should affirm and extend that protection in FY 2027.

Why appropriators can act on this

•       It is not a budget cut. Congress appropriated the money.

•       It is an administrative distribution choice that can be corrected in statutory language.

•       It is fiscally neutral. The fix does not require additional appropriations.

•       It frames the Subcommittee as defender of Congressional prerogative against executive branch mechanisms that divert appropriated funds from their intended purpose.

Supporting data

•       5,564 fewer grants awarded in FY 2025 compared to FY 2024 (UMR 2026).

•       R01 success rate dropped to approximately 17% in FY 2025, lowest in nearly 30 years (UMR 2026).

•       Multi-year funded awards increased from 5% of total grant funding in FY 2024 to 7% in FY 2025 (AAMC).

•       19 states and D.C. experienced at least a 10% decline in number of awards, North Dakota and West Virginia saw more than a 20% drop (UMR 2026).

•       NIH was 30% behind schedule in awarding research money in 2025 (Fierce Biotech, March 2026).

Argument 4: NOFO collapse and the end of directed funding

NIH published 756 Notices of Funding Opportunity in 2024, 120 in 2025, and 14 by mid-March 2026. At current pace, 2026 will post 98% fewer funding announcements than the 2012-2024 historical average. Of 336 forecasted NOFOs as of spring 2026, 205 are past their posting dates with no explanation. Targeted funding announcements are how NIH signals research priorities and protects understudied areas that cannot compete through generic parent mechanisms. Their disappearance represents a structural elimination of scientific stewardship. This argument is newer and less widely used by advocacy organizations, which makes it a distinguishing contribution from a working PI.

Supporting data

•       NOFO publication: 756 in 2024, 120 in 2025, 14 by mid-March 2026 (E. Ginexi, former NIH program officer, Chronicle of Higher Education, March 2026; confirmed by AAU).

•       2012-2024 historical range: 511 to 1,081 NOFOs per year. Even during the COVID-19 disruption in 2020, NIH posted over 650 NOFOs.

•       205 of 336 forecasted NOFOs past their posting dates as of spring 2026.

•       The two oldest overdue forecasts in the system are BRAIN Initiative announcements from NINDS, 581 days overdue.

•       Lorsch (NIH Deputy Director for Extramural Research) acknowledged the 16-step approval process now required for each NOFO.

Argument 5: Training pipeline damage compounds across cycles

Every Ph.D. student not admitted this year represents a five-to-seven-year gap in the biomedical workforce pipeline that no future appropriation can backfill. The 10-15 year training cycle from graduate admission to independent investigator means that a funding collapse produces generational, not episodic, losses. Every postdoc who leaves for industry or for China takes infrastructure knowledge, collaborative networks, and clinical connections with them, and they do not return. Early-career faculty, whose tenure clocks run independently of NIH funding timelines, are rationally reassessing their commitment to academic research. These are the arguments that moved UMass Amherst to cut its doctoral cohort from 997 to 712 in a single year.

Supporting data

•       58% of NIH-funded researchers report slower hiring of new lab members (national survey, 2025).

•       62% of junior tenure-track faculty report being very concerned that recent changes could derail tenure (national survey, 2025).

•       UMass Amherst reduced its doctoral cohort from 997 to 712 in a single year (2025).

•       China is the top destination for departing U.S. biomedical researchers, with reported active recruitment programs.

Argument 6: Procedural protections for funded grants

Between 2012 and January 2025, NIH terminated fewer than six grants midstream in thirteen years, a fact documented in a dissent by Justice Ketanji Brown Jackson. According to CRS analysis as of April 10, 2026, HHS had terminated 1,392 NIH grant awards totaling $539 million in unliquidated obligations. Many terminations were based on keyword screening rather than scientific review. The departure from established norms of funding stability creates an environment in which no PI can responsibly hire trainees, commit to longitudinal studies, or invest in infrastructure. Appropriators can address this through statutory language requiring scientific cause, written justification, and an appeals mechanism before midstream termination.

Supporting data

•       1,392 NIH grants terminated as of April 10, 2026, totaling $539 million in unliquidated obligations (CRS, April 2026).

•       Pre-2025 baseline: fewer than 6 midstream terminations in 13 years (2012-January 2025).

•       Just 35% of researchers whose grants were cut or delayed report full restoration of funding by end of 2025 (STAT survey).

•       U.S. Supreme Court in August 2025 allowed NIH to terminate $783 million in grants related to DEI-flagged content.

Argument 7: GAO has already found NIH in violation of the Impoundment Control Act

On August 5, 2025, the Government Accountability Office issued decision B-337203, finding that HHS and NIH violated the Impoundment Control Act of 1974 by withholding appropriated NIH funds from obligation and expenditure. The Act exists to protect Congress’s constitutional power of the purse from executive-branch impoundment. This argument is powerful precisely because it is not advocacy. It is a legal finding by the Congressional watchdog, and it can be cited as persuasive authority in ongoing litigation. If you include this argument, frame it as Congressional prerogative being vindicated by its own institution, not as partisan grievance.

Supporting data

•       GAO decision B-337203, issued August 5, 2025, explicit finding of ICA violation.

•       NIH obligated nearly $8 billion less between February and June 2025 compared to the same period in FY 2024 (GAO).

•       Lowest five-month obligation total since FY 2018 (GAO analysis of USAspending.gov data).

•       NIH published zero grant review meeting notices in the Federal Register between January 22 and March 3, 2025, which GAO determined halted NIH’s ability to review and award grants.

•       Over 1,800 NIH grants terminated between February and June 2025 (GAO).

Argument 8: Obligation slowdown and the $500M-at-risk

As of late April 2026, NIH is roughly $1 billion behind the pace of prior years in obligating new research grants (Mueller and Hwang, New York Times, April 22, 2026). The National Cancer Institute has obligated approximately $72 million in new competitive grants this fiscal year, compared with nearly $250 million by the same point in a typical year. One NIH institute is now projecting that up to $500 million in congressionally appropriated funds may go unspent because there are not enough remaining staff to process grants for legal and financial compliance. Reductions in force and early retirements hollowed out grants management. A computational text analysis tool scanning approximately 235 flagged terms is now vetting grants, with up to half of proposals flagged in some divisions. One biological science grant was held up a week because “sex” and “gender” were used interchangeably. This argument works because it is stewardship-focused and non-ideological.

Supporting data

•       $1 billion behind prior-year pace for new grant obligations as of late March 2026 (NYT).

•       Approximately 1,900 new and competitive grants awarded October through late March, less than half the usual number (NYT).

•       NCI: $72M obligated in new competitive grants vs. nearly $250M by same point in typical year.

•       $500M at risk of reverting to Treasury unspent in at least one NIH institute (NYT).

•       Up to 50% of grant proposals flagged in some divisions by the 235-term text analysis tool.

•       American scientists spend approximately 40% of their time on grant administration, a fraction climbing under current policy.

3. State-specific data points (use the one for your state)

Appropriators care about their states. A state-specific data point anchors your testimony in harm to the Subcommittee members’ constituents. Pull your own state’s figures from the United for Medical Research state toolkit at unitedformedicalresearch.org/nih-in-your-state/.

For Capito (West Virginia)

•       West Virginia experienced a greater than 20% year-over-year drop in number of NIH awards in FY 2025 (UMR 2026).

•       West Virginia University substance use disorder and infectious disease research funded by NIH (Judith Feinberg program, rural health integration).

•       Opioid crisis research and Alzheimer’s research are named Capito priorities in her FY 2026 markup statement.

For Baldwin (Wisconsin)

•       Wisconsin hosts the Morgridge Institute for Research, BloodCenter of Wisconsin, Aurora Health Care, Medical College of Wisconsin.

•       UW-Madison NIH funding anchor institution for Midwest biomedical training pipeline.

•       Baldwin has been a consistent defender of graduate student funding and NRSA training grants.

For your own state

•       NIH dollars awarded to your state in FY 2025 (from UMR toolkit).

•       Jobs supported and economic activity generated in your state (UMR).

•       Top disease burdens in your state (cancer incidence, heart disease mortality, Alzheimer’s prevalence, from CDC NCHS “Stats of the States”).

•       Specific grants terminated or frozen at your institution in 2025.

4. Modular draft sections you can adapt

Each section below is written in a voice generic enough to be customized. Replace bracketed placeholders with your own specifics. Rewrite to match your own sentence rhythm and discipline-specific language.

Opening block (required by Senate LHHS instructions)

[Your Name], Ph.D. / [Your Title], [Your Institution] / Prepared for the Subcommittee on Labor, Health and Human Services, Education, and Related Agencies / Testimony Concerning FY 2027 Appropriations for the National Institutes of Health

Opening paragraph (customize)

Chair Capito, Ranking Member Baldwin, and esteemed members of the Subcommittee, thank you for the opportunity to submit testimony on FY 2027 appropriations for the National Institutes of Health. I submit in my individual capacity as a federally funded principal investigator at [your institution], where I direct a laboratory [brief one-clause description of your research, with disease relevance].

The ask (customize the disease reference)

I respectfully ask the Subcommittee to appropriate $51.3 billion for NIH in FY 2027 and, of comparable importance, to bar NIH from using administrative mechanisms that convert the Subcommittee’s appropriated dollars into fewer funded research projects than the budget supports. For patients with [your disease area], whose treatments depend on a continuous pipeline of federally funded basic research, the stability of that pipeline is not separable from its dollar size.

Forward funding paragraph (use verbatim or paraphrase)

In FY 2025, Congress appropriated a higher NIH budget than in FY 2024. NIH spent that appropriation. Yet the agency issued 5,564 fewer grants than in the prior year, and the R01 success rate collapsed to approximately 17%, the lowest level in nearly thirty years. This occurred not because Congress cut funding but because NIH expanded its use of multi-year forward funding, obligating the full value of selected grants upfront rather than distributing dollars annually, which mechanically reduced the number of new awards the same appropriation could support. The mechanism severs the relationship between what Congress appropriates and what the research enterprise receives.

GAO impoundment paragraph (use verbatim or paraphrase)

On August 5, 2025, the Government Accountability Office issued decision B-337203 finding that HHS and NIH violated the Impoundment Control Act of 1974 by withholding appropriated NIH funds from obligation and expenditure. GAO documented that NIH obligated nearly $8 billion less between February and June 2025 compared to the same period in FY 2024, the lowest five-month obligation total since FY 2018. The Impoundment Control Act exists to protect Congress’s constitutional power of the purse from executive-branch impoundment. This is not an advocacy position. It is a legal finding by the Congressional watchdog.

NYT obligation slowdown paragraph (use verbatim or paraphrase)

The New York Times reported on April 22, 2026 that NIH is now roughly $1 billion behind the pace of prior years in awarding new medical research grants (Mueller and Hwang). From October through late March, NIH awarded approximately 1,900 new and competitive grants, fewer than half the pace of recent years. The National Cancer Institute has obligated roughly $72 million in new competitive grants this fiscal year against nearly $250 million by the same point in a typical year. One NIH institute projects that up to $500 million in congressionally appropriated funds may go unspent because not enough staff remain to process grants. Congress appropriated $47 billion for NIH in FY 2026. The money exists. The capacity to deliver it does not.

NOFO collapse paragraph (use verbatim or paraphrase)

NIH published 756 Notices of Funding Opportunity in 2024, 120 in 2025, and 14 by mid-March 2026. At current pace, 2026 will post 98% fewer funding announcements than the 2012-2024 historical average. Of 336 forecasted NOFOs as of spring 2026, 205 are past their posting dates with no explanation. Targeted funding announcements are how NIH signals research priorities to the community and protects understudied areas that cannot compete through generic parent mechanisms. Their disappearance is a structural elimination of scientific stewardship.

State-level evidence (customize for your state)

The FY 2025 disruption struck unevenly. [Your state] received [$X million] in NIH awards in FY 2025, supporting approximately [Y] jobs and [$Z] in economic activity. Investigators in [your state] experienced [state-specific award decline or terminated grants data]. [Your institution] received [$A million] in NIH funding in FY 2025, ranking [N] nationally.

Disease burden hook (customize, powerful for appropriators)

The people of [your state] experience [disease X] at a rate of [Y per 100,000], above the national average. The research programs most likely to produce the next generation of treatments for [disease X] are concentrated in [your institution] and peer institutions whose pipelines depend on continuous NIH support. When NIH success rates fall below 20%, the first projects cut are the foundational ones whose clinical payoffs are five to ten years away. That is the research pipeline that has produced every major therapeutic advance of the past three decades, and it is the pipeline most exposed to administrative volatility.

Training paragraph (customize with your lab or institution)

Graduate training programs cannot absorb episodic funding shocks without multi-year consequences. [Your institution, or another named institution] has [reduced / maintained with difficulty] its Ph.D. cohort this year in direct response to FY 2025 uncertainty. Every student not admitted represents a five-to-seven-year gap in the biomedical workforce pipeline that no subsequent appropriation can backfill. The 10-15 year training cycle from graduate admission to independent investigator means that a funding collapse produces generational, not episodic, losses. Junior faculty, whose tenure clocks run independently of NIH funding timelines, are rationally reassessing their commitment to academic research.

Closing block (customize)

The U.S. biomedical research enterprise is the product of seven decades of sustained, bipartisan Congressional investment. It has delivered approximately 93% public support for continued federal medical research, $2.57 in economic return per dollar invested, and the therapeutic foundation for treatments now standard in every state represented on this Subcommittee. The FY 2027 appropriation is an opportunity to restore predictability to a system whose strength has always depended on it. I am grateful for the Subcommittee’s longstanding commitment to medical research, and I thank Chair Capito and Ranking Member Baldwin for the bipartisan leadership that produced the FY 2026 Labor-HHS bill.

5. What to avoid

•       Do not submit more than four pages. The deadline is strict and over-length testimony will be rejected.

•       Do not use DEI-flagged vocabulary. NIH’s current environment and recent grant terminations make terms like “diversity,” “equity,” “inclusion,” “underrepresented,” and “disparities” (in workforce context) a liability. Translate equity arguments into capacity, stewardship, workforce sustainability, and merit-based language. “Health disparities” is acceptable when clearly referring to disease outcomes, not workforce composition.

•       Do not focus on the current administration as the main target. The Subcommittee is bipartisan and Capito is a Republican. Frame grievances as Congress-versus-administrative-mechanism, not as partisan critique.

•       Do not pile on every argument you have. One or two arguments with supporting quantitative data beat five arguments with thin support.

•       Do not use em dashes, semicolons, or colons except before a numbered list. Appropriations staff read for clarity, and simple punctuation helps.

•       Do not submit as an institutional representative if you have not coordinated with your institution’s government relations office. If uncertain, submit as an individual PI.

6. Submission checklist

•       Four pages maximum, 8.5 x 11, single-sided, single-spaced, 1 inch margins, no cover page.

•       Header on first page: your name, that the testimony is prepared for the Subcommittee on Labor, Health and Human Services, Education, and Related Agencies, and that it addresses NIH (and/or CDC, ARPA-H, BARDA, AHRQ if relevant).

•       Microsoft Word format only. No PDF. No fax. No mail.

•       Email to lhhs@appro.senate.gov with subject line “FY27 OWT”.

•       In the email body: your name, email, physical address, telephone number.

•       Deadline: close of business, Friday, May 22, 2026. Strictly enforced.

7. Useful sources for your own research

•       United for Medical Research, 2026 Annual Economic Impact Report (state-specific data): unitedformedicalresearch.org/annual-economic-report/

•       UMR state toolkits: unitedformedicalresearch.org/nih-in-your-state/

•       Elizabeth Ginexi Substack, NOFO analysis: elizabethginexi.substack.com (primary source for NOFO collapse data).

•       AAU analysis of NIH grantmaking slowdown (March 2026).

•       Mueller and Hwang, “Pace of N.I.H. Funding Slows Further in Trump’s Second Year,” New York Times, April 22, 2026 (obligation slowdown, NCI data, $500M-at-risk).

•       GAO decision B-337203, Department of Health and Human Services—National Institutes of Health—Application of Impoundment Control Act to Availability of Funds for Grants, August 5, 2025: gao.gov/products/b-337203.

•       Congressional Research Service, NIH Grants Policy Under the Second Trump Administration (IF13131): crsreports.congress.gov.

•       Research!America submitted testimony (FY 2027): model language and top-line ask alignment.

•       Senate LHHS formatting instructions: appropriations.senate.gov/subcommittees/labor-health-and-human-services-education-and-related-agencies.

•       Blue Ridge Institute for Medical Research rankings: brimr.org.

•       NIH RePORTER (grant-level data for your institution): reporter.nih.gov.

•       CDC National Center for Health Statistics “Stats of the States”: cdc.gov/nchs/pressroom/stats_of_the_states.htm.

Appendix A: Supplemental Arguments for Neural Interface Colleagues

This appendix is for colleagues whose research focuses on implantable brain-computer interfaces, deep brain stimulation, cortical prosthetics, vagus nerve stimulation, spinal cord stimulation, or the biology of the device-tissue interface. It supplements rather than replaces Section 2. The arguments below are distinctive to the neural interface field and carry particular weight given the China competition frame and the recent visibility of BCI in national security discussions.

A1. The biology-as-precondition argument

Every regulatory, reimbursement, and commercialization argument for neural interfaces assumes a device that performs durably in a human brain over years. The biology that determines whether it does (the foreign body response, oligodendrocyte and myelin dynamics, neurovascular disruption, signal origin biology, stimulation adaptation, cortical reorganization around the implant) remains poorly characterized and is the science most exposed to administrative defunding. Every BCI in clinical trials today was designed against biological understanding from the 1990s and early 2000s. Closing this biology gap is not separable from the field’s translational success.

Concrete device-class examples that illustrate the cost

•       A retinal prosthetic (Second Sight Argus II) achieved FDA approval and a $150,000 CMS reimbursement code, then went bankrupt because the device could not deliver durable outcomes. The CMS code was subsequently removed.

•       A responsive neurostimulator for epilepsy (NeuroPace RNS) remains financially precarious more than a decade after FDA approval, despite real clinical benefit.

•       In both cases, the regulatory and reimbursement infrastructure existed, but device performance did not sustain it.

Specific biology gaps worth naming in your testimony

•       Signal origin biology: what are we actually recording, and how do astrocyte calcium dynamics, pericyte-mediated hemodynamics, and oligodendrocyte precursor activity contribute to recorded signals?

•       Circuit-level stimulation selectivity: continuous intracortical microstimulation in human somatosensory cortex produces complete perceptual fading within one minute, because the biology of adaptation is not understood.

•       Chronic performance degradation: oligodendrocyte degeneration, myelin failure, and functional connectivity remodeling progressively degrade recording over months, through mechanisms no current materials strategy addresses.

•       Biocompatibility metrics the field and FDA rely on (impedance, single-unit yield, GFAP staining) capture none of these failure modes.

A2. The expired BRAIN Initiative Biology and Biophysics announcement

The BRAIN Initiative Biology and Biophysics of Neural Stimulation and Recording Technologies announcement (RFA-NS-20-006) was the one NIH mechanism specifically designed to fund the foundational biology of neural interfaces. It expired in 2022 and has not been renewed. The science it funded now falls between study section boundaries (too applied for basic panels, too basic for clinical panels) and is effectively unfundable through current mechanisms. The two oldest overdue forecasts in the entire NIH system, 581 days overdue as of spring 2026, are BRAIN Initiative announcements from NINDS. Restoration of this directed funding mechanism, or an equivalent, is a specific and tractable ask.

Language you can adapt for your testimony

I respectfully request that the Subcommittee direct NIH to restore the BRAIN Initiative Biology and Biophysics of Neural Stimulation and Recording Technologies announcement (RFA-NS-20-006, expired 2022), or an equivalent directed mechanism with program officer authority to manage across study section boundaries. This announcement was the one NIH instrument specifically designed to fund the foundational biology that determines whether implanted neural devices perform durably in human patients over years. Its expiration has left the science between disciplinary categories, and no existing mechanism fills the gap.

A3. The precompetitive consortium argument

The neural interface field faces structural gaps no single laboratory or company can fill: a shared biological framework for failure modes, a vertically integrated rodent-NHP-human translational pipeline, an open-access database linking stimulation parameters to neural and perceptual outcomes, an updated biocompatibility and safety evaluation framework validated with FDA/CDRH, shared preclinical infrastructure, a federated clinical performance database, standardized BCI outcome measures, common components and manufacturing specifications, and a training environment that sustains the interdisciplinary density the field depends on. Every neural interface company independently rebuilds this same foundational understanding at the cost of years and millions of dollars. A precompetitive consortium, analogous to SEMATECH for semiconductors or the NIH genomics data commons, would fix this without displacing private capital. This argument may be better directed at the BCI commission or the NIH Director rather than Senate LHHS, depending on your testimony focus. If you include it in LHHS testimony, frame it as the funding mechanism question, not the organizational structure question.

A4. The China competition frame, applied to neural interfaces

China’s seven-ministry BCI strategy, the Neuracle commercial approval, and its regulatory fast-track have closed the gap in device engineering and overtaken U.S. regulatory speed. China’s Brain Project funds brain mapping, primate models, and AI-neuroscience convergence, but its three stated pillars (cognitive neural circuits, brain disease, brain-inspired AI) do not target the specific neurobiology of chronic neural interfaces. The U.S. advantage is the depth of fundamental neuroscience built across two decades of NIH and BRAIN Initiative investment, and that advantage is what makes durable devices possible at year five in a human brain. Restricting publication would not protect this advantage. Maintaining the training pipeline would.

Language you can adapt

The U.S. competitive advantage in neural interfaces is not in device engineering, where China has closed the gap, or in regulatory speed, where China has overtaken the U.S., or in manufacturing scale, where China’s industrial policy creates structural advantages. The advantage is in the depth and breadth of the fundamental neuroscience and neurobiology of chronic neural interfaces, built across hundreds of U.S. laboratories over two decades of sustained NIH investment. That body of knowledge tells every device maker in both countries what they are designing against. Its erosion is a direct national-security consequence of the administrative mechanisms dismantling the ecosystem that produced it.

A5. Relevant citations for neural interface PIs

If you cite the biology gap, the following literature anchors the argument. Use what fits your focus and substitute your own publications where relevant.

•       Hughes et al., Brain Stimulation 2022 (perceptual fading with sustained ICMS).

•       Wellman et al., Biomaterials 2020 (cuprizone-induced demyelination degrades chronic recording independently of electrode integrity).

•       Chen, Cambi, Kozai, Biomaterials 2023 (clemastine-driven remyelination improves recording stability).

•       Gregory et al., Acta Biomaterialia 2023 (reduced dendritic branching and spine density around implanted arrays).

•       Kozai et al., Journal of Neural Engineering 2026 (cross-species ICMS review, biophysical modeling, 15 co-investigators spanning rodent to human).

Forward this guide to colleagues through LinkedIn, your lab website, or Twitter/X. Volume of independent voices by May 22 is the goal.

A6. A note on the BCI commission and independent voices

Neural interface arguments are also being made to the national-security BCI commission through a separate channel, with a more technical focus on the precompetitive consortium structure and the specific biology gaps listed above. If you submit to Senate LHHS using the arguments in this appendix, you are not duplicating a single institutional submission. You are adding an independent voice from a working laboratory to the permanent appropriations record, which is legally and politically distinct from the commission’s synthesis. The two pathways reinforce rather than substitute for each other. Appropriators do not read the commission synthesis when drafting FY 2027 report language. They read the public testimony record. Conversely, the commission can point to a record of independent PI testimony when making the case that the biology gap is a structural priority rather than the concern of a single laboratory. Submit based on what you actually know and believe, framed for Senate LHHS staff rather than for commission reviewers.

Takashi D. Y. Kozai, Ph.D.

Ernest E. Roth Professor of Bioengineering, University of Pittsburgh

Prepared for the Subcommittee on Labor, Health and Human Services, Education, and Related Agencies

Testimony Concerning FY 2027 Appropriations for the National Institutes of Health

Chair Capito, Ranking Member Baldwin, and esteemed members of the Subcommittee, thank you for the opportunity to submit testimony on FY 2027 appropriations for the National Institutes of Health. I submit in my individual capacity as a federally funded principal investigator at the University of Pittsburgh, where I direct a laboratory developing implantable brain-machine interfaces for patients with paralysis, sensory loss, and treatment-resistant neurological disease, and where I co-lead UP NExT, an institutional neural engineering initiative spanning 23 faculty across Engineering, Medicine, and Arts and Sciences. I respectfully ask the Subcommittee to appropriate $51.3 billion for NIH in FY 2027 and, of comparable importance, to take structural actions that protect the Subcommittee’s appropriated dollars from administrative mechanisms that have, in the judgment of the U.S. Government Accountability Office, already operated outside the law.

The GAO has already found NIH in violation of the Impoundment Control Act

On August 5, 2025, the Government Accountability Office issued decision B-337203, finding that the Department of Health and Human Services and NIH violated the Impoundment Control Act of 1974 by withholding appropriated NIH funds from obligation and expenditure. The GAO determined that NIH obligated nearly $8 billion less between February and June 2025 compared to the same period in FY 2024, the lowest five-month obligation total since FY 2018. The Act exists to protect Congress’s constitutional power of the purse from executive-branch impoundment, and the GAO’s institutional role is explicitly to defend Congressional prerogative. This is not an advocacy position. It is a legal finding by the Congressional watchdog.

The situation has deteriorated rather than improved since that finding. The New York Times reported on April 22, 2026 that NIH is now roughly $1 billion behind the pace of prior years in awarding new medical research grants. From October through late March, NIH awarded approximately 1,900 new and competitive grants, fewer than half the pace maintained in recent fiscal years. The National Cancer Institute has obligated roughly $72 million in new competitive grants this fiscal year, compared with nearly $250 million by the same point in a typical year. One NIH institute is now projecting that up to $500 million in congressionally appropriated funds may go unspent because there are not enough remaining staff to process grants for legal and financial compliance. Two mechanisms drive this slowdown: the reductions in force and early retirements of grants management personnel, and a computational text analysis tool scanning grants against approximately 235 flagged terms, with up to half of proposals flagged in some divisions. One biological science grant was held up for a week because the words “sex” and “gender” were used interchangeably. Congress appropriated $47 billion for NIH in FY 2026. The money exists. The capacity to deliver it does not.

The U.S. lead is in basic neuroscience, not in devices or manufacturing

China’s seven-ministry brain-computer interface strategy, the Neuracle commercial approval, and its regulatory fast-track have closed the gap in device engineering and overtaken U.S. regulatory speed. What China has not replicated, and cannot replicate on any short timeline, is the depth of fundamental neuroscience and neurobiology that tells every device maker in both countries what they are designing against, what they are recording from, and what they must solve to build a device that works at year five in a human brain. That body of knowledge was built across hundreds of U.S. laboratories over two decades of sustained NIH and BRAIN Initiative investment. It sets the performance ceiling for every brain-computer interface, deep brain stimulator, vagus nerve stimulator, gene therapy, and regenerative medicine therapy now in clinical trials or on the market.

The genomics precedent is the correct frame for appropriators. The Human Genome Project and GenBank were entirely open. China read every paper and trained scientists in U.S. laboratories. The U.S. won genomics anyway, because the competitive advantage accrued not to whoever read the data first but to whoever had the workforce, startup culture, and clinical infrastructure to act on it. The competitive moat is the person who can read the paper and know what to build next, which requires ten to fifteen years of interdisciplinary training that no industrial policy can accelerate. That is the asset at risk.

The mechanisms being dismantled have direct device-class consequences

NIH published 756 Notices of Funding Opportunity in 2024, 120 in 2025, and 14 by mid-March 2026 (E. Ginexi, former NIH program officer, Chronicle of Higher Education and Substack, March 2026, confirmed by AAU analysis of grants.gov data). At current pace, 2026 will post 98% fewer funding announcements than the 2012-2024 historical average. Of 336 forecasted NOFOs as of spring 2026, 205 are past their posting dates. The two oldest overdue forecasts in the entire NIH system are BRAIN Initiative announcements from NINDS, 581 days overdue. The BRAIN Initiative Biology and Biophysics of Neural Stimulation and Recording Technologies announcement (RFA-NS-20-006), the one NIH mechanism specifically designed to fund the foundational biology of neural interfaces, expired in 2022 and has not been renewed. Every BCI in clinical trials today was designed against biological understanding from the 1990s and early 2000s.

The consequence of closing this gap is visible in the device class. A retinal prosthetic that achieved FDA approval and a $150,000 CMS reimbursement code went bankrupt because the device could not deliver durable outcomes, and the CMS code was subsequently removed. A responsive neurostimulator for epilepsy remains financially precarious more than a decade after approval. Every reimbursement, regulatory, and investment dollar Congress directs toward this field assumes a device that performs durably. The biology that determines whether it does is the science being defunded first.

The forward-funding mechanism converts appropriated dollars into fewer awards

In FY 2025, NIH expanded its use of multi-year forward funding, obligating the full value of selected grants upfront rather than annually, which mechanically reduced the number of new awards the same appropriation could support. The agency issued 5,564 fewer grants than in FY 2024, and the R01 success rate collapsed to approximately 17%, the lowest in nearly thirty years (United for Medical Research 2026 Economic Impact Report, March 2026). West Virginia experienced a greater than 20% year-over-year reduction in the number of awards, and 19 states and the District of Columbia experienced at least a 10% decline. Pennsylvania sustained approximately $40 million in terminated or frozen NIH grants by late 2025. The University of Pittsburgh, where I work, received $669.7 million in NIH funding in FY 2025 and ranked seventh nationally, yet the number of projects supported declined year-over-year even as total dollars rose, exactly the pattern predicted by forward funding. The House FY 2026 appropriations bill blocked further expansion of forward funding, and I urge the Senate to affirm and extend that protection.

Grant terminations have departed from established norms

Between 2012 and January 2025, NIH terminated fewer than six grants midstream in thirteen years, a fact documented in a dissent by Justice Ketanji Brown Jackson. According to Congressional Research Service analysis of HHS data, as of April 10, 2026, HHS had terminated 1,392 NIH grant awards totaling $539 million in unliquidated obligations. The GAO decision documented that NIH terminated over 1,800 grants between February and June 2025. Many terminations were based on keyword screening rather than scientific review. Just 35% of researchers whose grants were cut or delayed report full restoration of funding by the end of 2025 (STAT survey). The departure from this established norm has created an environment in which no principal investigator can responsibly hire trainees, commit to longitudinal studies, or invest in infrastructure, which is incompatible with multi-year scientific work.

The training pipeline is breaking in real time, not hypothetically

The workforce that staffs every biomedical research laboratory, clinical program, and life-science company in the United States is the product of a 10-to-15-year training cycle from graduate admission through postdoctoral fellowship to independent investigator. This cycle cannot be restarted by a future appropriation. The damage being done right now is not a projection. It is already in the public record, across named institutions, in the first quarter of 2026.

The Perelman School of Medicine at the University of Pennsylvania directed its graduate groups to reduce Fall 2025 Ph.D. admissions by approximately 35%, cutting the entering class from a recent average of 307 students to 201 (Daily Pennsylvanian, February 2025). The University of North Carolina at Chapel Hill admitted 25% fewer biomedical graduate students, approximately 75 across the biomedical sciences. UMass Chan Medical School rescinded provisional Ph.D. offers to dozens of biomedical sciences applicants. The University of Pittsburgh, where I work, temporarily halted doctoral admissions in early 2025 while the institutional impact of proposed indirect cost cuts was assessed. West Virginia University limited admission to its health sciences doctoral programs, citing the same funding uncertainty. These are not temporary adjustments that can be absorbed by the system. Every Ph.D. student not admitted represents a five-to-seven-year gap in the workforce pipeline, compounded across dozens of institutions nationally. No subsequent appropriation can backfill a cohort that was never admitted.

The damage at the faculty career stage is comparable. R01 success rates for early-stage investigators fell from 29.8% to 18.5% over two years. In a STAT survey of nearly 1,000 NIH-funded researchers published March 2026, 13% reported losing researchers to institutions in other countries, 7% reported that postdocs or other staff had rejected job offers, and two-thirds are now advising trainees to consider nonacademic paths. Fifty-three percent are advising students to consider positions outside the United States. These are labs closing, programs contracting, and young scientists redirecting mid-career. An assistant professor at the University of Washington, profiled by STAT, reduced her mouse colony by half, approximately 200 fewer animals, to avoid laying off staff after a grant program she had applied to was rescinded. This pattern is replicated in laboratories across the country that do not make the news.

At NIH itself, the institutional pipeline has been hollowed out. The agency lost approximately 1,200 staff in early 2025 alone (~5% of its 20,000-person workforce), including probationary grants-management personnel, staff scientists, and tenure-track investigators in the intramural program (Science/AAAS). The NIH Postbac Program, which has recruited approximately 1,600 recent college graduates into biomedical research each year since the 1960s, was suspended. The postdoctoral program was suspended. These are the entry points into NIH-trained careers that produced the workforce now staffing U.S. biomedical industry, academic medicine, and federal scientific leadership. Rebuilding these pipelines, assuming they can be rebuilt, will take a decade.

The ecosystem consequences compound across every downstream sector

The consequences of a damaged biomedical training pipeline are not confined to academic laboratories. Every biotechnology startup, every pharmaceutical clinical development team, every medical device firm, every academic medical center, and every federal scientific agency draws from the same pool of NIH-trained Ph.D.s and postdoctoral fellows. United for Medical Research documents that NIH funding in FY 2025 supported 390,863 jobs and $94.15 billion in new economic activity, with $2.57 in economic return per dollar invested (UMR 2026). Over the past decade, NIH funding has generated $822 billion in economic activity and supported 3.7 million jobs. The jobs are not merely academic, they are the technician, engineer, clinical-research coordinator, biostatistician, and regulatory-science positions that U.S. biomedical industry depends on as well as laboratory supplies manufacturers and suppliers.. Every laboratory that closes or contracts subtracts from this base. Every cohort of Ph.D. students not admitted reduces the hiring pool available to domestic industry five to seven years from now.

The national-security consequences are immediate. Foreign-born scientists, who represent a substantial fraction of NIH-funded investigators and trainees, are being actively recruited by China, the European Union, Canada, Singapore, and other competitors whose biomedical investments are accelerating. Each departing early-career scientist takes with them training, infrastructure knowledge, and collaborative networks that took years and federal dollars to build. They do not return. The training pipeline that produced U.S. leadership in gene therapy, mRNA vaccines, cancer immunotherapy, and neural interfaces is the same pipeline being dismantled now. China’s seven-ministry brain-computer interface strategy does not need to match U.S. basic neuroscience depth if the U.S. disassembles its own advantage faster than China builds one.

What I request the Subcommittee to do

•       Appropriate $51.3 billion for NIH in FY 2027, consistent with the request of Research!America and the broader biomedical research community.

•       Require in statutory language that NIH obligate appropriated funds through traditional annual increments rather than multi-year forward funding, except in narrowly defined circumstances with advance Congressional notification, and that NIH restore directed funding mechanisms with program officer authority to manage across study section boundaries. Direct NIH to report quarterly to the Subcommittee on competing awards relative to FY 2024 baseline, success rates by career stage, NOFO posting pace, and obligation pace against fiscal-year deadlines.

•       Restore staffing to NIH’s grants management function sufficient to obligate the full appropriated budget, restore the NIH Postbac and intramural postdoctoral training programs, and prohibit the diversion of appropriated research dollars toward keyword-based screening processes that delay award processing. No portion of a $47-to-$51 billion appropriation should go unspent for lack of administrative capacity to deliver it.

•       Recognize foundational neuroscience of neural interfaces, and the analogous precompetitive biology underlying gene therapies, regenerative medicine, and sustained drug delivery, as national-security priorities. Restoration of the BRAIN Initiative Biology and Biophysics announcement (RFA-NS-20-006), or an equivalent directed mechanism, would close a specific, quantifiable gap in the pipeline that supports every neural interface company, clinical trial, and Defense-relevant program currently operating.

•       Establish procedural protections for funded grants requiring scientific cause, written justification, and an appeals mechanism before NIH may terminate an award midstream, and affirm timely obligation of appropriated funds as a non-negotiable operational standard. NIH’s FY 2026 funds were released by OMB on March 16, 2026, nearly six months into the fiscal year.

Closing

The strategic question for FY 2027 is not whether to fund NIH. The Subcommittee has answered that question in every bill for seven decades, and Chair Capito and Ranking Member Baldwin delivered the bipartisan FY 2026 Labor-HHS bill that enacted $48.7 billion for NIH in March 2026. The question is whether the appropriated dollars reach the laboratories that produce the next generation of therapies, or are absorbed by administrative mechanisms that the GAO has found unlawful, while a coordinated competitor is accelerating investment in exactly the fields where U.S. leadership is most fragile. Congress appropriated the money. The question before this Subcommittee is whether the executive branch will deliver the science the public paid for. I am grateful for the Subcommittee’s longstanding commitment to medical research, and I would welcome the opportunity to provide additional detail on any of the issues above.

Respectfully submitted, Takashi D. Y. Kozai, Ph.D., Ernest E. Roth Professor of Bioengineering, University of Pittsburgh. Contact: tk.kozai@pitt.edu | 412-383-9044 

Email Body for Senate LHHS FY27 Outside Witness Testimony Submission

Recipient: lhhs@appro.senate.gov 

Subject: FY27 OWT

Dear Chair Capito, Ranking Member Baldwin, and Subcommittee staff,

Attached please find my written testimony for the Fiscal Year 2027 Outside Witness hearing record on National Institutes of Health appropriations. Submitted in my individual capacity as a federally funded principal investigator at the University of Pittsburgh, the testimony respectfully requests $51.3 billion for NIH in FY 2027 and five structural actions that protect the Subcommittee's appropriated dollars from administrative mechanisms that the U.S. Government Accountability Office, in decision B-337203 of August 5, 2025, found to be in violation of the Impoundment Control Act of 1974.

The core argument is national security. China has closed the gap in biomedical device engineering and overtaken the United States in regulatory speed. The remaining U.S. advantage is the depth of fundamental biomedical science built across two decades of sustained NIH and BRAIN Initiative investment, and the 10-to-15-year training pipeline that produces investigators capable of translating it. That pipeline is being dismantled administratively while a coordinated competitor accelerates investment in exactly the fields where U.S. leadership is most fragile. The April 22, 2026 New York Times reporting documents that NIH is roughly $1 billion behind the prior-year pace in awarding new grants, with one institute projecting up to $500 million in appropriated funds may go unspent for lack of staff to process them. Congress appropriated the money. The testimony addresses whether the executive branch will deliver the science the public paid for.

Contact information:

Name: Takashi D. Y. Kozai, Ph.D. Title: Ernest E. Roth Professor of Bioengineering, University of Pittsburgh Email: tk.kozai@pitt.edu Telephone: 412-383-9044 Physical address: 208 Center for Bioengineering, 300 Technology Drive, Pittsburgh, PA 15219

Thank you for the Subcommittee's longstanding commitment to medical research, and for the bipartisan leadership that produced the FY 2026 Labor-HHS bill.

Respectfully,
Takashi D. Y. Kozai, Ph.D.