Julie Locher opened NIH RePORTER and started counting. What she found was not a hypothetical policy concern — it was a concrete workforce accounting. 573 active K23 awards, nearly 46 percent of all active K23s in her analysis, were funded through Clinical Trial Required announcements. Those awards, under NIH Program Update 3, are now on a countdown to extinction. The funding opportunity notices that created them will be expired by July 13, 2027.
NIH’s official rationale is almost disarming in its simplicity: K awards exist for career development, and running a clinical trial exceeds that purpose. Under Program Update 3, K awardees who want to lead a trial must now find separate institutional, philanthropic, or grant funding to do it. What reads on paper as a tidy jurisdictional clarification is, in operational terms, a requirement that early-career clinician-scientists either work at a well-resourced institution with a well-funded mentor, or abandon the aspiration of leading a randomized trial during the years when that skill is actually formed. The policy does not eliminate clinical trial training. It just prices most candidates out of it.
A “Gold Standard” With No Path to Practice
Theodore “Jack” Iwashyna, a Johns Hopkins ICU physician and research mentor, captured the core absurdity in a single post heading: “When their definition of ‘gold standard RCT’ is ‘no RCT at all.'” His point cuts to the methodological contradiction embedded in the policy. NIH is an institution that requires randomized controlled trial evidence to inform clinical practice, yet it has just removed one of the few structured mechanisms by which clinicians learn to design and execute RCTs. You cannot mandate the gold standard while defunding the training that produces the people who can apply it.
NIH’s own Program Update 3 language acknowledges this tension, noting that “NIH recognizes the importance of supporting the development of scientists who can lead rigorous clinical trials” and promising “additional strategies focused on this need.” A replacement mechanism is, per the update, forthcoming. Jenna Norton, a Program Director with expertise in social determinants and data standards, puts the problem plainly: “Making changes before a new plan is in place is disruptive.” That is a diplomatic description of what is, in workforce terms, a hard stop imposed on a pipeline that takes years to refill.
The counterargument from NIH leadership, specifically the contention attributed to former Office of Extramural Research director Mike Lauer that clinician-scientists can develop trial expertise by participating in “big trials” led by senior investigators, deserves direct scrutiny. Norton addresses it point by point: the large trials that could serve this function were already insufficient in number before 2025, they rarely engage with implementation science, and the Notices of Funding Opportunities that would spawn new ones have been curtailed in the current environment. Removing K awards does not redirect traffic to an alternative route. It removes the on-ramp while the highway itself is under construction.
The Equity Fracture Hidden in the Mechanism
The distributional consequences of Program Update 3 are where the policy shifts from administratively clumsy to structurally damaging. Locher’s RePORTER analysis does not just show aggregate numbers — it shows concentration of impact. The K23, the award specifically designed for clinically trained investigators pursuing patient-oriented research, bore nearly half of all Clinical Trial Required designations. Physicians building the methodological foundations of their research careers were disproportionately relying on this mechanism. Eliminating it does not affect all K award categories equally: it lands hardest on the pathway designed for the investigators most likely to eventually lead clinical trials that answer clinical questions.
Locher identifies the fault line with precision. Investigators at highly resourced institutions, or those whose mentors already lead funded trials with available slots, have a workaround. They can absorb the new requirement. Investigators at smaller or less-endowed institutions, or those pursuing research questions in areas where large funded trials are uncommon, face a structural barrier that has nothing to do with their scientific capability. Per Locher’s analysis, access to clinical trial training under Program Update 3 becomes a function of zip code and mentor funding status rather than investigator merit or research importance. That is a training policy that encodes inequality into the workforce it is supposedly developing.
Implementation science sits at the center of this equity problem. Norton flags it as the subfield most acutely exposed: the discipline concerned with translating evidence into practice, with understanding why effective interventions fail to reach patients who need them, is exactly the kind of work that does not generate large industry-sponsored trials. Pharmaceutical sponsors do not fund implementation research. The NIH K award was one of the few mechanisms that did. With Program Update 3, investigators asking how to get proven treatments to underserved populations lose their primary training vehicle precisely because that training required the independence to run a trial of their own design.
A Pipeline That Breaks Quietly, Then All at Once
Norton points to an uncomfortable institutional dynamic behind the policy. The push to eliminate K award clinical trial designations predates the current administration — it originated with NIH Director of the National Institute of General Medical Sciences Jon Lorsch, who has since been promoted under Jay Bhattacharya’s leadership. The idea did not emerge from a systematic workforce analysis. It emerged from a conceptual argument about what “career development” means. That argument was made by a scientist whose domain expertise, per Norton, does not extend to clinical trial methodology or the workforce conditions of clinician-scientists.
The consequences will not be visible immediately. The 573 active K23 awards in Locher’s analysis represent investigators already in the pipeline. Their work continues. What disappears is the cohort behind them — the clinicians now in fellowship or early faculty positions who would have used a K23 Clinical Trial Required award to design their first feasibility study, recruit their first patients, and develop the operational fluency that makes a future R01 or U01 application credible. That cohort will not announce its absence. It will simply not materialize. And in ten years, when trial leadership capacity is thin and implementation science has fewer investigators than the field requires, the causal chain back to Program Update 3 will be diffuse enough to dispute.
NIH’s promised replacement mechanism may yet address these concerns. But a workforce pipeline runs on certainty. Clinicians choose research careers based on the funding environment they can project several years forward. A policy that expires existing NOFOs by July 2027 while offering only the assurance that alternatives are “actively being worked on” does not create a bridge. It creates a gap and asks early-career investigators to trust that something will appear on the other side before they fall. For an institution that rightly demands evidence before changing clinical practice, NIH has implemented a workforce restructuring without publishing its study design, its endpoints, or its plan for monitoring harm.
References
- Julie Locher, LinkedIn post on K23 RePORTER analysis, 2025
- Jenna Norton, LinkedIn post on K award elimination and implementation science, 2025
- Theodore “Jack” Iwashyna, LinkedIn post on NIH Program Update 3, 2025
Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.
