The NIH Data Book page reporting Research Project Grants (RPG) and Fellowship awards by race has been updated for FY2025. Completely unsurprisingly, despite the impression that may have been conveyed by the assault on “DEI” last year, white PIs continue to secure the most RPGs, accounting for 68.7% (69.4% in FY24) of the total. Asian PI’s account for another 25.4% (24.5% in FY24) , AI/AN for 0.3%, NH/OPI for 2.0% and “more than one race*” 2.0%.

Black PIs account for 3.5% (down from 3.7% in FY24) of RPGs.

Image

The current census reminds us that the US population is comprised of 74.8% white, 6.7% Asian and 13.7% Black individuals. You know, for the geographic diversity fans who are apparently going on acreage or whatever, it’s handy to have a frame of reference. Also the StayMadAbby types.

Anyway. There does not appear to be much change in the distribution of RPG PIs in FY2025.

Fellowships are another matter. As you can see from this first graph, white trainees get the most grants, accounting for on the order of 72%-79% of awards in the interval from FY2021 (78.9%) to FY2025 (72.3%). Asian trainees account for 15.3%-18.8% in the interval from FY2021 (16.2%) to FY2025 (18.8%).

Image

Black trainees account for 4.9%-5.4% in the interval from FY2021 (5.0%) to FY2024 (5.4%). Because of the scaling issues, and the tremendous differences in the number of awards to each racial category, it is hard to make out changes over time. So I’ve represented these awards in terms of the percent change from the prior Fiscal Year. This makes it easier to see trends. The number of white Fellowship trainees was pretty stable year over year from 2021 to 2024, with a 9% drop for FY2025. Asian and Black Fellowships were up 9% and 11% respectively in FY2024, with Asian Fellowships increasing by another 2% in FY2025.

The biggest change is for Black PI fellowships. There were only 59.5% as many as in the prior year**. Down 40.5%, compared with the 9% change for white Fellowship trainees.

The current NIH Director, Dr. Jayanta Bhattacharya has been attempting to walk a razor’s edge on the anti-DEI agenda that saw grants get de-funded last year and many notices of funding opportunity and other declarations of NIH’s Interest in Diversity rescinded. In this, his favored approach is to declare the awards that have been discontinued as “bad science” and to state that may of the NOFO are calling for similarly bad science. He’s keen on trying to claim that the issue is hypotheses that cannot be tested***. An example is in the Why Should I Trust You podcast from 2/26/2026 where he claims a hypothesis about structural racism does not have “falsifiability” (~33:10 of the podcast). He then goes on to ask how you can have a control group and insists therefore it is not a scientific question to ask if structural racism contributes to a health outcome.

This discussion became more acute, and with relevance to the data I’m showing here, when the podcast hosts brought up a specific case of a Black trainee who had been denied the continuation of his F99/K00 award into the K00 (postdoc) phase. Around 40:51 of the pod, Bhattacharya starts in saying that from the title of the project it “has a premise that in principle couldn’t have a control group“. This, to his mind, is what justified the cancellation of this F99/K00 training grant. Leaving aside, for the moment, that a peer review panel scored that grant well enough to fund in the first place and likely concerned itself with matters of control groups. Leaving aside, for the moment, that the PI of this award has been pretty vocal about the fact he indeed included a control group.

This is a TRAINING award. A Fellowship. These are not research grants. These include a big research component, yes, but it is not the primary focus. Part and parcel of postdoctoral training is that if there are any limitations or flaws in the way the postdoc thinks about a scientific problem, the PI (and any additional mentors) is there to help train them to think about science better!

As Dr. Bhattacharya rightfully admits, he’s not familiar with the proposal. He’s making assumptions. Just like he does for many, if not all, of the cancelled grants. And he has this way of backing down on podcast appearances or interviews when he is challenged about health disparities. He admits they exist. He says they are important to address. He felt the need to brag about NIH funded progress on sickle cell disease, and on women’s health issues, to Congress. He even more or less agreed with a Congress Rep about the need for a diverse research force (albeit a terse “yeh” en route to trying to pivot it back to more comfortable ground).

It is really hard to fit all this together to excuse the huge reduction in Fellowships awarded to Black trainees for FY2025. It really makes it look like the reason for grants not being awarded was because of the race of the trainee. It can’t be the topic, the Director cops to science on health interests of minority populations being important. It can’t be the science, these are fellowships and the entire point is to help the trainee become a better scientist.


h/t: a “helpful” suggestion on the blueski.

*I never really know what to do with the “more than one race” category. I don’t know jack squat about who tends to use that, why and what their phenotype / affiliation would be to you, I or any other observer. I’m tempted to just leave these numbers out but have included them.

**If you want to view FY2024 as some sort of aberration, Black fellowship awards were 66.3% of the FY2023 total while white Fellowship awards were 90.5% of the FY2023 total.

***Never mind that a ton of NIH funded activity is not directed at testing hypotheses, but has nothing to do with DEI trigger words.

In a new post tagged NIH Funding Blog, we see the first report on success rates, which for some reason has been absent from the Data Book up until now*, even as other FY2025 statistics have been included in updated charts. These are overall, NIH-wide numbers.

Image

RPG success was 13.0%. This is down from 18.5% in 2024 and 21.3% in 2023.

R01- equivalent success rate was 13.0%, down from 18.7% (FY24) and 21.6% (FY23).

Just for grins, I broke out Ye Old Cumulative Probability of Award calculator. To get yourself over the 80% probability** in FY2025 you needed to submit five more R01 proposals than you did in FY2023. Five more. In addition to the seven that were required to get you to that 82% probability** of success in FY2023. Also seven more than the ten proposals that would get you over 90%.

I wonder if Rep Steny Hoyer had this updated information at hand during the NIH oversight hearing yesterday in which he questioned Director Bhattacharya about paylines [YouTube: ~46:00 mark]. Hoyer also mentioned that 25-35% of proposals funded was the ideal for good science.

As you are grinding your teeth and tearing your hair about how terrible this all is to be down to a 13% success rate for R01s, and it is, I’ll take this opportunity to remind you that the Hoppe et al. 2019 paper reported that R01 application success rates for proposals considered in FY 2011–2015 were 17.7%, if the PI happened to be white and 10.7% if the PI happened to be Black. Maybe some of you who were muted, if even nonchalant, about the latter will finally get a gut level check on exactly what that meant. Or perhaps we will need five straight years of this? Or, given the relative disparity was identical from FY2000-FY2015 if we combine the Hoppe outcome with the success rates published in Ginther et al 2011.


*The NIH blog entry notes: The NIH Data Book … is also being updated with more FY 2025 grants and application data on a rolling basis.

**never any sort of guarantee. this is spherical cow statistics here.

NIH testifying to Congress on 03/17/2026 [YouTube Link]

20:18 – lauding NIH funded research for the latest highly promising HIV medications which have the potential for an eradication success. [dm- no comment about USAID being defunded, something that was previously fighting HIV globally]

20:30 – namechecks CRISPR therapy for sickle cell, mentions it is primarily African-American kids affected. [HIV/AIDS, Black kids… both right wing trigger points. Don’t know if he’s aware…..]

21:20ish – lauding advances such as “computers…convert electrical signals in the brain to audible speech”. Goes on to brag on tiny electrodes in the spine which can restore movement and touch. digital models of retinal cells. [dm- all of this stuff has arisen from the sort of invasive animal research that gets the AR nutters’ primary attention. He’s phrasing this* (“computers convert”, “digital models”) in a way that seems very misleading about how we got here. How is he supposed to square this kind of health advance with the new office of ARA nutjobbery and organoid bullonio?]

blah blah repurposing existing medications…..

22:40 “applying gold standard science” [dm: DRINK!]

22:48 “what can happen with sustained NIH investments in basic and clinical science”. [dm: and of course he says nothing here about the admin desire to cut the NIH budget by 40%.]

23:14 – Bhattacharya says that reproducibility involves independent teams looking at the same result (?) and finding the same answer. [dm: looking forward to NIH funding multiple groups more or less addressing the same questions, or funding subsequent grants to verify the outcomes of prior grants.]

23:40 – talking about empowering Program to make decisions on the portfolio. Claims everyone he talks to in Program is all for this. some random crack about funding directly useful science “rather than produce papers that sit on the shelf”. [dm: how many examples do we need of scientific work that sat on a shelf….until it turned out to be really super useful at X, Y or Z health advance?]

24:17 – geographic diversity, very earnest comment he wants to work with Congress [dm: bread meet butter.] to make this a reality because it is important for the scientific vigor of the country. [dm: mmmhmm. geographic diversity is necessary but literally no other aspect of diversity helps with vigor. I see. ]

Now for the Q&A:

26:50ish – some nonsense about how we need to do science that unites instead of divides in the context of organoids supposedly replacing fetal tissue models and approaches. [dm: Leaving aside the fact that science has moved on from embryonic stem cells to large extent for good reasons, wtf is this principle that NIH only fund research that “unites” people with different opinions? Heading back to his opening remarks on HIV therapy and potential eradication, if this principle had been used since the 80s HIV would still be a death sentence with no available treatments.]

27:30 addressing the geographic diversity following Rep Aderholt [Alabama’s 4th, Big recipient of military funding curiously not geographically diverse) question.

27:50- “so many great scientists across the country who have less chance of getting their work funded”. [dm: where’s the proof of this, Jay? where’s the painstaking analysis of success rates, institutional infrastructure support, PI accomplishment, etc?] 28:20 he starts in on IDC as the way to build facilities which attract the great researchers that bring in the grants…”catch 22″. we have to invest outside of the top 20 (institutions) says JB. we need more than just a few biohubs. Aderholt agrees. [dm: sure! let’s absolutely swap the relative investment in NIH versus, say, the military.]

30:50 flat out lies about public confidence in science dropping during COVID because it wasn’t delivering for people. [dm: Oh yeah? Apparently you didn’t work at one of the vaccine super centers in 2021. People were DELIGHTED about science delivering for them. Also see the dwindling need for overflow morgue trucks in NYC. The cause was something else, gee what could it have been…?]

32:10ish Rep DeLauro stitching back together the “trust” issue with the fact that public confidence should be high because of all the here and now therapies and advances Bhattacharya bragged about from the top.

33:02- responding to question about when NIH will get back to normal grant making. happy talk gaslighting about how great everything is going and how heroically the staff worked last year.

37:50 some dumbass Rep saying that Indirects should be set at the median because if some institutions can do it for the median than everyone can. [dm: what?] Proposes 30% because agricultural grants get that. brings up private foundations, gates. [dm: I’m sure the NIH director will educate him on how Gates foundation charges things to direct, right? and about how agriculture ain’t biomedical science, etc. right?

39:24 same dumbass Rep saying we “can’t” increase the NIH budget and that the trillions debt is the big problem in the self same week we are blowing billions after billions after billions on blowing up another country for no apparent gain. hilariously says if NIH takes away the IDC it will somehow permit lots of more grants to be funded, crocodile tears about the young investigators. when we know damn well this guy will support budget reductions from IDC cuts. 40:00 Rep pissonmyheadandtellmeitsraining still talking. now on MAHA and processed food.

43:00ish Rep Hoyer asking about acting IC directors. Bhattacharya “I have put in a process” to identify and hire permanent directors, promising hires as early as this month. blah blah scientific merit and leadership ability. nothing about what was wrong with the ones he fired or the prior processes. Hoyer asks about political interference, Bhattacharya stammer around, doesn’t really answer but does say “it’s too important (to get the right person) to leave to politics”. [dm: we’ll see about that. NIEHS ring any bells?] Hoyer asks about the massive staff cuts from last year. “was it 22% overpopulated?” Bhattacharya dodges, says they are hiring, distances self from DOGE firings.

46:00 Hoyer goes at the slow pace of NCI grants being issued this year.. Bhattacharya ‘that report is lagging’ according to NCI director…basically “it’s on track” and “success rate will be at least 10%”.

Hoyer: “what is the average payline at this point in time?”. [dm: aaahhhhahahha] JB “we fund 8-10% of grants”. Hoyer; “when I was first on the committee it was….25%-35% and all the scientists who came before this committee said that was about right” JB brings up some supposed PI who submitted 60 grants [Hoyer snaps back that he’s talking about all NIH, not buying the anecdote gaslighting] and AI proposals and the denominator. JB admits lots of great grants don’t get funded by says this has nothing to do with payline [dm: wtf?] and doesn’t really say what it does have to do with.

[dm thought, mid-hearing: Bhattacharya keeps saying he keeps discovering health successes that seem like “science fiction” to him. Yeah, no duh. science has done some amazing stuff that was previously the province of speculative fiction. it’s amazing. the system that he criticizing and trying to upheave is what brought us here. How is he so dismissive of this and so arrogantly sure he can change it for the better instead of killing the golden goose?]

1:00:00 Rep Bice asking about any “bottlenecks” in getting grants out the door. JB doesn’t see one now. calls it political noise and not “reality”. [dm: Jeremy Berg’s graphs from NIH’s own data belie this answer] more on geography- “not right to let great ideas wither on the vine just because they are in oklahoma instead of boston. more blah blah about infrastructure / facilities support. solution: competition for facility support separate from project funding. describes a portal where institutions compete to hire the researchers who have the grants, “like NCAA athletes”. [dm: this does not sound that workable Jay, we’re not talking 18-20 year old football players moving around the country]

1:05:20- Rep Frankel questions grant cancellations. then mentions vaccine compliance in other countries without mandate is because of health care systems where people actually get coverage. basically questioning JB’s dismissal of vaccine mandates. Frankel throws bomb “the problem is doctors in high places sowing doubt”.

1:08:50 Frankel expresses concerns about anti-DEI craziness, women’s health. JB asserts commitment to women’s health. [dm: Frankel let him off the hook and let him do his red herring gaslighting by not directly tying the grant cancellations to her concerns]

1:12:00ish Simpson brings up IDeA progam. JB admits it has been great. but says the “scale …smaller than it ought to be”. starts blah blahing about his allegedly competition friendly approach to funding.


*make no mistake. This is very careful and intentional. Jay may sometimes stumble around verbally when he’s trying to square his illogic with the obvious facts but he works very hard at his weasel language, particularly in rote or prepared remarks.

A substack blog entry from NIH Program Officer Elizabeth Ginexi outlines the sharp discontinuation in the issuance of Notices of Funding Opportunity (NOFO) at the NIH in 2025 and 2026. These NOFO (previously Funding Opportunity Announcements; FOA) are known primarily around these parts as Program Announcements (PA, PAR, PAS) and Requests for Applications (RFAs), with Notices of Special Interest (NOSI) swimming along like pilot fish. The substack does a good job of pointing out that the ICs of the NIH do not just throw out these calls for specific or focused types of research proposals willy nilly. They are developed by a serious scientific process and were approved for issuance by a reasonable oversight process within the ICs.

Image
created by Elizabeth Ginexi; published on substack

The substack also does a good job of pointing out that the reason for the sharp reduction in NOFO is the exertion of political control over what science gets funded.

That’s it, that’s the actual problem. Full Stop.

But of course, this being written by a long term program officer who can’t help but misrepresent why the NIH needs NOFO in the first place. And as per usual, I think that if we try to use something trivially falsifiable as our justification for our criticism of the regime’s attack on NIH, we hand them a trivial defense. And disrupt our momentum with any favorably-inclined listeners. Such as, oh I don’t know, members of Congress.

In the post, the author writes:

NOFOs exist because not all research needs are obvious to individual investigators. When a new pathogen emerges. When clinical trials reveal an unexpected side effect that needs investigation. When one population experiences a disease at higher rates than others but nobody knows why. When a promising scientific approach exists but no one is applying it to a specific problem. These are moments when waiting for unsolicited grant applications is not enough.

I’ll partially credit the “new pathogen” one. Sure, there are occasionally topics that spring up that are not sufficiently on any extramural scientist’s radar and are sufficiently urgent that Program needs to beat the bushes as hard as possible to djinn up a lot of proposals that otherwise would not emerge so quickly. These are rare. As we saw with COVID-19, there was already a ton known because of prior investigator-initiated research. Which greatly reduced the time to create a vaccine and therefore to reduce the carnage. But the rest of this… well, [needs citation] is the most polite way I can put it.

I’m sadly the veteran of well over two decades of following FOA/NOFO in my field. I’m sure it’s recall bias, but I do not remember a single one issued by the ICs of greatest interest to me where any of this was true. There are ALWAYS members of my field who are already ON IT. They have presented relevant studies at meetings. They have in many cases been submitting applications for funding on the topic to the NIH…sometimes for years. The problem is that they do not score well. So despite what appears to be a quite obvious need, grants are not funded. Eventually, in my experience, someone prods Program into action, directly or indirectly. Many someones perhaps. And somehow, someway…in some cases….a targeted funding opportunity emerges. Followed thereafter in the usual delayed course of grant award by the funding of some grants.

It isn’t just the specialized requests that fit into traditional RFAs or PAs which follow, rather than lead. I watched the whole Sex as a Biological Variable push come through…and I know for certain sure that many members of my field were indeed already specializing in examining SABV, and many were at least proposing to include SABV for years prior to this. It did not require that NIH Program tell them this was a good idea. There were many proposals submitted that the NIH chose not to fund.

I read over the Hoppe et al 2019 paper and followed all the NIH excuse making which surrounded that. I am mostly in pre-clinical research areas but I’m sure I’ve reviewed on panels which dealt with more than one project that touched on sub-population disease issues over the years. Same deal as SABV. I am certain, pending additional actual data from NIH, that there were many, many proposals in this area that simply were not funded. (This is what the Hoppe et al 2019 “topic analysis” tells us.)

Don’t even get me started about BRAINI.

The problem is rarely, if ever, because we dumb old scientists haven’t cottoned on to some pressing scientific area of inquiry that those genius Program Officers have discovered. More likely, this process is the other way ’round*. Which the blog more or less confesses to:

Writing a NOFO was one of my primary responsibilities as a program official. When my institute identified a gap, I would work with scientific experts to define the problem precisely, determine what kind of research was needed, and draft an announcement that would attract the right investigators.

Where “my institute identified a gap” requires some citation of evidence that this was not merely the accumulation of unfunded proposals, PI’s phone calls or scientific meeting conversations, or the reading of Discussion sections of papers and reviews. Where “scientific experts” generally means extramural scientists. Sometimes the very people who have been submitting proposals and not getting funded. But sometimes merely “the right investigators**”. Who are brought into a subfield topic by the lure of easy money, thereby supplanting funding to those poor PIs who have been fighting the good fight all along. There were definitely sex-differences researchers in my field who were pretty pissed off about the usual suspects now getting funded on stuff they had been proposing for years. It’s outside my usual circles but I guarantee the sort of scientific colonialism that gives established white PIs the grants to study health concerns in minority populations has some interesting stories about those who were not funded.

So. If I am so correct, why has NIH used their NOFO/FOA system? Simple. Because they are, as always, high on their own supply of pretending that the priority scores and percentiles that emerge from initial peer review are uncontaminated assessments of pure scientific merit. They refuse to understand the inevitable grinding circularity of review by the already-successful. They refuse to acknowledge that peer review is driven by subjective preferences of scientists. They refuse to admit that relative rankings driven primarily by three assigned reviewers would change with a different set of assignments even within the same study section.

They operate as true believers that peer review outcomes are highly reliable measures of scientific merit.

Which makes them reluctant to fully embrace their role in deciding….that’s right…priorities.

They go through all this hoop jumping to define the problem precisely and determine what kind of research is needed to get to a point where they dare to put the finger on the scale in convening a review panel which only reviews proposals on the topic. Which doesn’t guarantee that at least a few will get “fundable scores” but it sure does increase the odds. A lot. After that (or in cases where special emphasis panels are not used) they get to use the NOFO to stiffen their spine to argue for a grant to be funded at a lower priority ranking over another grant with a higher score.

This, as Glinda told Dorothy, was always in their power to do.

ICs did not need NOFO in the past and they do not need NOFO now to service a goal of funding less popular or new ideas. Which is why this should not be part of the argument.

All they have to do is pick up the grants of those who are already proposing those topics. If anything, the regime has further underlined the role of Program de-emphasized the role of peer review scores by 1) creating the ND-but-competitive outcome and 2) telling strict-payline ICs that mindless adherence to score is no longer allowed! This is, on the face of it, all the ICs need** to get around the lack of NOFO and to make reasonable decisions to pick up the proposals that are addressing new and underfunded issues.

The thing to keep an eye on here is whether they do it, or whether they default to their previous feigned helplessness that they are not allowed to pick priorities without a NOFO. The thing to keep an eye on is whether POs are trying their hardest to balance the portfolio, whether the new political review process interferes or whether they decide to comply in advance with whatever they fear from the political layers of review.

[ETA 03/19/2026: See substack blog post NIH, NOFOs, and the scientific community from Mark Histed for related reading]


*Look, I’m not complaining too much about this system in a personal way. I’ve benefited from targeted NOFO in terms of getting funded. At least once in a way that would appear to the DM derangement syndrome reader to be the sweetest of old-boys-club type of deals. I’ve also struggled for many submissions to get “new” topics funded only to see targeted NOFO come out belatedly and the grants go to seemingly everyone else but me. I tried to do the right thing on SABV as NIH started to ramp up their justification campaign…and got absolutely killed by the sex-diffs folks on study sections who were pissed off about NIH doing this general SABV thing instead of funding their proposals.

**One would be right to be a little concerned about the reference in the substack post to the “right” investigators, and the real reasons they do not simply pick up the first grants that come in proposing a topic at present. Combine this with some suspiciously specific language in some of those prior NOFO and some very tight windows between announcement and due date and, well, there is a certain smell here. Highlighting the backslapping choosing of favorites* that may go on with some NOSI is probably not a strong argument here either.

The NIH finally updated the data underlying good ol’ Report 302 of the Databook to include FY2025, which ended 09/30/2025. This allows us to examine the fate of R01-Equivalent applications by the percentile rank calculated from initial peer review. I like this Report because it permits us to get a good idea of the implicit payline across all of NIH or at each IC across recent Fiscal Years. It can give you some idea of how likely your grant proposal which scored at a given percentile was to fund or not fund at the IC in question. This can be important to understand why a grant application’s chances seem very different from past Fiscal Years. Those depictions of funding by percentile do not, however, give us the overall success rates, which are another good measure of changes over time, differences between ICs, differences between categories of investigator, etc, etc.

This recent update of Report 302 follows a blog entry from NIH which gave us some initial information on Funding Rate for FY2025. Funding Rate is a measure which reflects whether a PI who applied in a given FY received a grant award, regardless of how many proposals they submitted. This per-PI statistic, you will immediately realize, is designed to deflate the perception of the difficulty of grant funding to the extent anyone mistakes their Funding Rate statistic for the per-application Success Rate. I had to graph the data that they provided in a Table to grasp it properly and I came up with this depiction, which shows that the percent of applicants who were funded in FY2025 was way down across several categories of PI. I would view the difference from a baseline established FY2021-2023 and also note that the decline in per-applicant success rates started in FY2024, perhaps related to the NIH appropriation [PDF] increasing only 2.5% over the prior year compared with 5.2% and 5.5% increases over the prior year in FY2022 and FY2023. The FY2025 appropriation was 0.7% lower than in FY2024.

The Report 302 charts are not the best at depicting the proportion of grants that are funded or not funded at a given percentile rank because they give counts of each outcome within a bin. It is, however, pretty simple to copy the tables into you favorite data handler and calculate a funding probability at each percentile. For this, I’ve used the total of R01-equiv funded plus the not-awarded proposals, i.e., I’ve left off the R56 outcomes for now.

This makes it even easier to make out that the implicit payline (that score ~at which it and anything better is funded) was 4%ile in FY2025, 10%ile in FY2024 and 11%ile in FY2023. It also shows a ~20 percent chance of your proposal being funded ran out around the 15th percentile in FY2025, compared with 21st percentile the prior year. In FY2022 and FY2023, you had this robust (!) 20 percent chance all the way to a 26 percentile score.

For some reason, the RePORT pages which give the aggregate Success Rate stats have not yet been updated for FY2025. So we might be motivated to use these new data we have to try to get at the number for overall NIH-wide success rates on a per-proposal basis.

The report.nih.gov certificate has been out of date for the last couple of days so I can’t re-check directly. But I had some very different numbers which I quickly sketched out for “R01-Equivalent” from the Report 302 table than is typically reported for “R01-Equivalent” from the aggregate charts. By memory, NIH success rates overall have been running around 20% for several of the past several fiscal years. So what’s up?

This has to do with the fact that Report 302 only includes grant proposals with a score. All ND grants are not included in the denominator.

E.g., when I take the Report 302 table and calculate the percent R01-equiv awarded from the total of those plus the not-awarded, I come up with a success rate of 21.6% for FY2025, down from 31.1% (FY2024), 35.7% (FY2023) and 35.7% for FY2022. These numbers go up by roughly 1.2% in all prior FY, if I include R56 awards in the counts. In FY2025 including the R56 increased the overall success rate to 22%.

I managed to get back on to the NIH databook by ignoring the threat warnings and grabbed the R01-equiv application numbers. As remembered, the overall success rates were 22% in FY2022 and FY2023 and fell to 19% for FY2024. But this confirms that the total applications (including the R56) in Report 302 are only 44.5% of the NIH aggregate total applications in FY2022 and FY2023 and 43.6% in FY2024. So slightly over half of R01-equiv were triaged.

[ETA 011326: Jeremy Berg posted some related IC by IC success rate analysis in a Bluesky thread today.]

The FY2025 funding picture was disrupted significantly by the mandate for Multi-Year Funding of about half of the budget for extramural awards. This produced a mid-year declaration by NCI that their ~7-8 percentile payline was going to be 4 percentile, at least for the latter part of the year. Since about half of ICs do* not publish their paylines, we can only infer what has occurred with respect to funding/percentile relationships from the post-FY funding data. I don’t actually remember when the RePORT Data Book page which presents the funded / not funded grants by percentile rank is usually updated but I’ve been waiting, with bated breath, to compare. I think I was probably not alone in wondering if we were not going to see data transparency from the NIH under the new regime.

Image

The wait is over.

The graph for all NIH R01-eqv grants funded or not funded for FY2025 shows a couple of things. There are something above 300 grants funded for each of the best scoring percentile bins. There’s some slope of fundability descending from the first several percentile ranks, whereby a few grants are funded even way out in the 21-25%ile zone. And when defining the implicit payline as I always do, by “the percentile below which almost everything funds”, there IS still a relatively normal distribution. Sure, we’re looking at 4 percentile NIH-wide implicit payline. Which is very low. But it also looks a lot like prior years in which the probability of funding thereafter declines with distance from this implicit or inferred payline. Another way of saying this is that the probability of funding increases with proximity to that payline.

Bottom line: Peer review was still a major determinant of what got funded in FY2025.

Image

So how about that 4%ile virtual payline? How bad was this? For handy reference see the same chart for FY2024. It placed that implicit cut line around 10%ile. This is much, much worse in FY2025. The examination of individual ICs may not be all that illuminating, as I suspect it will mostly be confirmatory of the overall trend. Unsurprisingly, NCI as the largest IC and a “hard payline” IC looks very much like the aggregate.

That is, an inferred** payline of about 4%ile in FY2025 compared with 10%ile for FY2024.

You have to squint your eye and make a highly subjective qualitative call on “virtually everything funds” for the ICs that claim not to use (and certainly do not publish) a payline, due to smaller numbers and therefore more variability. NIDA was somewhere around 10-14%ile in FY2024 and this dropped to roughly 9%ile in FY2025. NIMH was at maybe 13-15%ile in FY2024 and…holy moly what a disruption in FY2025. Seriously go look. NIAAA is a small IC and the profile tends to get messier with fewer grants awarded by FY2024 looked like a 13%ile implicit payline and this sits around 5%ile for FY2025.

Image

One of the nose sniffingest strict payline institutes, NINDS, is more of the same. Payline of 8%ile in FY2025 and you were SOL past 16%ile no matter what your category or subject matter. This was down from 11%ile in FY2024, a handful of grants funded in the 17-23%ile bins and one 25%ile award. NIDDK’s implict payline was halved, from 13 to 6%ile.

Interestingly, NIAID appears to have held ~steady at 10-11%ile from FY2024 to FY2025. One might have predicted they would have paid a price for all the Fauci hate. I guess not.

Obviously there are many ways to look at the funding patterns for FY2025. This is but one view on how things are going, one that I happen to find interesting and useful for comparing practices between FYs and ICs. YMMV.


*”did not”, past tense. All will go on the “we don’t have a payline” train this year.

**Reminder that these are post-fiscal year actual data for the subset of R01-equivalent mechanisms. A “strict payline” IC rarely ever is, some of the stuff funded in the grey zone could be ESI policy, Program priority or a reflection that published paylines almost necessarily have to be conservative relative to what they eventually afford to fund.

I didn’t actually listen to the Why Should I Trust You podcast that closely, I was just skipping to try to find a particular segment. Skipping around today, I ran across something to elaborate a mystery discussed yesterday.

The question is: What does Director Bhattacharya mean when he says the NIH will prioritize cutting edge research as a substitution for ESI affirmative action funding policy?

16:40- Bhattacharya discusses a project he completed some years ago* looking at “how close to the cutting edge was NIH research“. So yeah, this should be interesting.

I figured out how old were the ideas in published research by NIH funded researchers. In the 1980s …NIH published research was at the bleeding edge…ideas that were 0, 1 or 2 years old. In 2000s, 20teens…the typical NIH funded research..working on ideas that were 7 or 8 years old.

So the main argument is that in ~1985 the NIH funded scientific endeavor had discovered and published less science than it had by ~2015. This seems obvious, since it was smaller, and had less time for accomplishment. The distinction spreads across a major factor, i.e., the commitment of Congress to double the NIH budget ending in FY2003, at the beginning of the critical comparison interval.

One thing that seems an interesting distinction is that he’s talking about ideas. Not techniques or approaches, which are often definitional of Innovation in the prior approach to grant review. Also, I might observe, the way we often define younger scientists as cutting edge, particularly when a department is hiring a new Assistant Professor. It is not uncommon that existing faculty prioritize a new hire that brings new techniques that are not currently expressed in the department.

This further questions his assertion later in the podcast that by selecting cutting edge ideas, the NIH will be selecting for younger investigators. If the game becomes one of only proposing studies that are based on ideas that have not been published until very recently, established labs are going to be very good at pivoting. If, as Bhattacharya asserts, the system rewards same-old, same-old research ideas at present, continually-funded investigators succeed because they play that game. This doesn’t in any way predict that they cannot adapt to a new “cutting edge” reality. At times I have pointed out that I have more ideas for a research grant by breakfast than I could ever write or work on if they were funded merely for the asking.

~17:35- “the NIH has become used to funding the same ideas over and over again because it is safe. I put in structures so that NIH goes back to funding research on the cutting edge

Here, one wonders what will happen with ideas that are not new in the published literature, but struggled to get funding from the NIH. He does not appear to be defining cutting edge by the dearth of NIH funded grants on the idea, but rather by the dearth of papers.

17:44 …researchers used to getting money…on ideas that are 7, 8 years old, are less likely to get that money than researcher that are proposing brand new ideas that are at the cutting edge

The examples will be plentiful. Of areas of investigation that took 7 or 8 years, maybe even more, to go from initial idea to the generation of enough papers and data that a subfield gains a good understanding of what has been proposed or discovered. It is like many a study section discussion of Innovation which eventually get around to asking – what is the point of novel assays or approaches if they do not become integrated into a body of research which springs from that innovative original work? This was even addressed by a 1996 working group which tried to oppose the elevation of Innovation as a major review criterion. Go read.

This is also where NIH talking out of the other side of its mouth on rigor and replication will enter the chat. How do we get to a full understanding of how replicable a finding may be if the clock for potential funding stops ticking at year 3 after initial publication? Let’s bench race a scenario. Imagine a PI gets a grant funded on some cutting edge idea. They probably have published the idea to a limited extent in developing the successful grant application, or maybe at the worst they get something out in Year 3 that is substantive. Now, if other peers see that as valuable, they get to work prepping their grant proposal, including a bunch of preliminary data generation. Those grants are going in, at best, in Year 4 and 5 of the original award. If lucky, considered for funding at 3-5 years after the original idea is published. By Bhattacharya’s measures, we’re already past the 0, 1 or 2 year limit for a cutting edge proposal.

In this highly probable timeline, replication is not possible. Not with NIH funding anyway.

~18:20- the problem of critics not having a handle on the numbers or being clear about what they refer to raises it’s ugly head. It allows the Director to go off on correcting the error on total funded grants while the interviewers stammers around and fails to nail him down on the fact that fewer grants are being awarded, across most relevant categories.

20:10 “I look at the narrative…seems ignorant about how grant funding decisions are made

This, IMNSHO is where we go wrong with hyperbolic characterizations and with not being clear in our understanding of what we’re discussing, to the extent we can glean real information about grant award. It lets the regime sound like they know more and the critics do not. Of course the podcast host is absolutely right that fewer grants were funded for the same money because of multi-year funding. But this gets obscured because Bhattacharya can dodge around about how many new versus total grants have been funded and not talk about how many fewer new grants have been awarded.

21:20 Bhattacharya is explaining how they cut Diversity supplements. Hilariously admits that the parent grants are of high merit and all that, but says the supplements “are an add on at the end of the year for DEI purposes“.

Ok first, the “purposes” is doing a lot of work here. Yes, the administrative supplements for enhancing diversity are typically awarded to support a trainee who qualifies under the previously stated NIH’s Interest in Diversity. But the science that they are working on is the already funded science which Bhattacharya concedes, en bloc, are good projects. These generally support salary, not a lot of additional research. These do not fund different research, they put another person on the already approved research Aims! As outlined in this NOFO:

The funding mechanism being used to support this program, administrative supplements, can be used to cover cost increases that are associated with achieving certain new research objectives, as long as the research objectives are within the original scope of the peer reviewed and approved project, or the cost increases are for unanticipated expenses within the original scope of the project.

So to the extent it is “DEI research” that Bhattacharya is trying to say is meritless or, as he puts it, “politicized science, not actual science” [~22:43]. Canceling administrative supplements isn’t a mechanism for addressing that. Nothing about the supplement is allowed to stray outside of the original scope of the peer reviewed and approved project. It could not be any clearer that the reason administrative supplements were being canceled is because of the identity of the trainee it was supporting.

As bit of a sidebar the “end of the year” which he is careful to mention, as if it is damning, is a structural requirement of the NIH that could be simply waived away. They could administer these by awarding them right away once approved. (IME, the minimum timeline for deciding on administrative supplements is on the order of 6 weeks not the 9 months we are familiar with for competing awards.) I’m sure the applicants and the trainees would greatly appreciate a more rapid timeline. It is their choice to structure these as being awarded at the next non-competing start date.

~23:00 Bhattachary blaming the “previous leadership of the NIH” for politicized science. It seems relevant to observe that the administrative supplement program dates back at least to FY2005, and was continued under the first Trump administration, as was the statement of NIH’s Interest in Diversity.

26:54- Referring to the suspension of all grants to Columbia University, he references some past sanctions of Bob Jones University and says “I don’t think taxpayer dollars should be used to support institutions that are racist.” He seems quite willing to determine when a racism has happened in this case. Yet he’s pretty pugnacious about questioning whether any inattention to a health disparity topic or bias in funding to those topics might be due to racism. Maybe they should have asked him if the continued Ginther Gap and topic-associated disparity identified in Hoppe et al meant NIH was racist and all appropriations to it should be halted?


*He appears to be referencing this paper: Packalen, M., Bhattacharya, J. NIH funding and the pursuit of edge science, Proc Natl Acad Sci U S A. 2020 Jun 2;117(22):12011-12016. doi: 10.1073/pnas.1910160117. Epub 2020 May 19.

Well, this is fascinating.

Director Jayanta Bhattacharya went on the Why Should I Trust You podcast in an episode dated Feb 26, 2026 [h/t].

Around 59:00 he is answering the podcast host after she pressed him about his priority for training early career scientists and data showing in 2025 there were 20% fewer early stage investigators funded. The Director says that the approach has to be “organic” and says the way NIH has been addressing the problem of early career researchers is in error:

Partly it makes sense, like training programs. but also partly by affirmative action. right, so you have a lower score allowed to be funded just because they are early career. …That’s never going to work.

He then goes on to say the proper strategy is to commit to funding “cutting edge ideas” because “typically, it tends to be younger researchers with those ideas“.

To remind anyone who is unaware, the approach of using heavy handed top-down quota based affirmative action to fund the proposals submitted by Early Stage Investigators (no major NIH funding, no more than 10 years past terminal degree) kicked into gear in 2008 after a startup period from 2007 onward. The essence is that the Institutes and Centers of the NIH use various strategies in their funding to make sure that ESI applications are funded at approximately the same success rate (this is the quota part) as the applications of experienced investigators. This has meant the publication of relaxed paylines for consideration for funding at the ICs that publish paylines. For those that do not publish paylines, they use language which amounts to the same thing.

This has remained policy until at least 2024, possible through 2025 as well. As Bhattacharya put it, applications with a lower score were funded just because the PI was early career.

Which means, of course, that applications with higher merit as assessed by peer review were not funded, just because the PI was experienced.

Check out this page at the writedit blog for links to policy pages on funding priority for some guidance to how this works, if you are unfamiliar with the process. At present the NCI page essentially says the ESI payline will be 10%ile where the payline for established investigators will be 4%ile, as an example of the published payline type of IC. The NIAAA page for FY2024 says “NIAAA will follow the NIH Policy and work towards supporting new investigators on R01 equivalent awards at success rates equivalent to that of established investigators submitting new R01 equivalent applications” as an example of the more weasel-word “we don’t have a payline” type of IC.

This focus on “cutting edge” proposals sure sounds a lot like we are returning to a prioritization of the Innovation criterion, just as the NIH has done a major re-design of peer review to de-emphasize Innovation.

Now, we all know that years, indeed decades, of the NIH trying to force reviewers to prioritize Innovation did not work. Their own data showed us that voted priority scores correlated best with the Approach criterion scores, next best with Significance and only third best with Innovation. Perhaps they should have listened more to the 1996 working group recommendation that Innovation should not be raised to the status of a major review criterion?

“Creativity/innovation is currently included in the Significance…Creativity/innovation, in and of itself, is not necessarily a hallmark of scientific merit…By definition it is a relatively rare trait”

Now, we don’t know how the Director plans to make grant selections more based on “cutting edge” research. Possibly he will force Program to ignore the outcome of peer review entirely. I doubt that. I believe we will continue to see ICs which set a virtual payline below which almost everything funds and pick up grants above that in a probability that has a rough correlation with score. The only real question is who will be deciding on grants which address new priorities and by what process are those established.

A new page (blogpost?) on the NIH site, purportedly authored by four IC Directors and nine Acting Directors, is supposed to clarify said funding processes. It sounds like business as usual to me. Importantly, the IC Directors retain final funding authority. This page insists they remain committed to “maintaining objective peer-review which is central for NIH to uphold our values of transparency, impartiality, and fairness“. It assures us of “Internally documenting each decision for applications that will be funded, including its alignment with IC and agency priorities“.

Internally. Don’t we all presume they did this before? Wasn’t there some sort of internal documented process for the grey-zone pickups above the implicit payline? I mean, I’ve written more than one rebuttal to study section critique that was requested of me in the context of a PO potentially proposing a grant for funding outside of the implicit payline zone.

Only one of the thirteen individual IC Director blurbs mentions “cutting edge“, only three mention Early career investigators and only two mention “early-stage” research. Seven of the blurbs do mention innovation but without any particular highlighting or connecting to early stage investigators- frankly these all sound like IC priority boilerplate we’ve read for decades.

This communication does not robustly support the continuation of ESI affirmative action procedures.

I don’t know that we need to take a podcast statement from Bhattacharya as settled policy. He says a lot of stuff on podcasts, only some of which is likely actionable or will be actioned. But it is well worth keeping an eye on any signals that ESI preferences will be scaled back.

It will be interesting to see if there is a way to accomplish the support of early career applicants for major research funding by selecting for supposed “cutting-edge” research. I am skeptical. I am willing to believe that on average the ESI population proposes more cutting edge (to the extent we can define that) work. Established investigators do as well, after all, where do you think the ESI population learned to conduct that cutting edge research? So in terms of all of the cutting edge proposals, there are probably as many from established as from ESI investigators.

So we’re back to…peer review? Trying to select cutting edge from non-cutting edge? Well, we know damn well that the current peer review process is not going to be helpful. Way back in 2007 departing NIH Director Zerhouni was quoted:

Told about the quotas, study sections began “punishing the young investigators with bad scores,” says Zerhouni. That is, a previous slight gap in review scores for new grant applications from first-time and seasoned investigators widened in 2007 and 2008, [then NIGMS Director Jeremy] Berg says.

Now just think about this. NIH has continued the affirmative action policy for eighteen years. During which they have made clear in their policy statements that they have had to continue special percentile breaks to achieve the funding quota. This shows that despite this highly robust declaration eighteen years ago that they really, really, really (no, really) wanted peer review to treat the applications of ESI fairly, peer review refuses to do it. To this very day.

Even, perhaps mind bogglingly, as those ESI who benefited from the policy have advanced in their careers and are currently a huge, probably dominant, presence on study section.

The scientopia.org domain has been unstable, and mostly unavailable, for years now. The bottom line is that there was never sufficient revenue to pay the hosting and registration fees. So I have returned to good old freebie WordPress.com. For the most part I’ve managed to port over my content, although links that refer back to the scientopia site (or in some cases ScienceBlogs) are not updated*.

Unfortunately, not everyone who blogged at Scientopia has bothered to set their content up in a new location. I find myself most frequently wishing to refer to a particular post that appeared on the DataHound blog on May 15, 2014 titled Mind the Gap, this link is via the Internet Archive Wayback Machine. This interest arose this week as I was contemplating the data NIH posted showing that approximately 41% of PIs are “At-Risk”. The At-Risk designation is defined here at this NIAID page:

an at-risk investigator has had substantial, independent NIH funding as a PI and, unless successful in securing a substantial research grant award in the current fiscal year, will have no substantial research grant funding in the following fiscal year. A “substantial” research grant is defined by its inverse, i.e., an NIH research grant that is not on the List of Smaller Grants and Awards that Maintain ESI Status. Whether the PI is currently funded by a substantial research award does not matter.

You can see from an expired PAR that includes “At-Risk” in the title that at least some NIH ICs use this in a manner similar to the New Investigator designation. Presumably, as a priority factor for funding. Unfortunately, most of the NIH websites have been stripped of material referring to this population and their rationale for having this designation. And of course it is impossible to know, beyond searching RePORTER for grants funded under specific NOFO that mention At-Risk investigators, to what extent NIH has taken steps to prioritize funding for such individuals.

In the original post from 2014, DataHound set out to answer a related question** of interest to NIH funded careers:

Image

If a given investigator has a year with no reported funding, what is the likelihood that they will show funding again in a subsequent year? For investigators who were funded in FY2006, but not FY2007, there are 6 possible years (FY2008-FY2013) for them to be re-funded. For gaps in years after FY2007, there are fewer years of “follow-up” available. The results are shown below:

You can see that this is the question of what happens if At-Risk investigators are not saved by obtaining funding. What happens if a PI does not have NIH funding for a year, two years, or more? What is the probability they will regain funding?

Depending on FY, something on the order of 21-30% of them regained funding within two years. An asymptote of about 40% was reached after the unfunded gap reached 4 years. I don’t know if NIH was looking at a similar type of analysis when they came up with the At-Risk designation, hopefully something like this fed into their policy making. As I said in that last blog post, there is something of a Scylla/Charybdis problem in having a policy of funding more and more ESI applicants while also trying to save the At-Risk population. There has to be some firm understanding of PI exit rates to reach a rational balance of goals. If the historical asymptote of 40% refunded after losing all major awards is a good target, would it be more advantageous to reach that asymptote after only one or two years of uninterrupted funding? Or to use this in combination with the At-Risk designation to target ~40% of the 41% for saving?


*You can often take a look at the URL and do a little search here on the WP site to find it, if interested.

**His analysis was based on R-mechanism*** funded PIs from Fiscal Years 2007-2013.

***If this analysis included any R-mech, no matter the size, this differs a bit from the At-Risk definition that excludes minor funding such as R03 and R21.

I will be honest, I am somewhat amazed we are getting any funding data for FY2025 out of NIH. The databook has not yet been updated, so we’re not out of the woods yet. Still, there’s a new page up at the NIH, posted 2/10/2026. It appears designed to address a specific complaint / “priority” expressed by the current NIH Director Jayanta Bhattacharya about the support for early career investigators. They even cite one of his podcasts.

As you know, Dear Reader, the NIH has wrung its hands over the fate of younger/newer applicants since approximately forever and has created numerous approaches to try to improve various statistical markers of this. For now, this history continues to be detailed here and touches on important waystations from R23 NIRA in 1977, to R29 FIRST in 1986, the cessation of R29 / checkbox era in 1997/1998 and the creation of ESI status and the heavy handed top down quota based affirmative action award system that we have at present in 2008. Yes, there was a period from 1998 to 2008 when younger/newer investigators were left to the wolves. Yes, this included the post-2003 ensuckification of success rates that attended the end of the doubling and flatlining of NIH appropriations. Guess who this screwed over?

Image

Getting back to the point, NIH has posted a table of the number of applicants and awardees from FY2021 to FY2025. I think this is the first good look we’ve had at anything close to comprehensive success stats from FY2025. In this case it is their currently favored* per-applicant statistic. That is, the number of PIs who submitted at least one application for an R01-Equivalent grant and the number of awardees. Disaggregated, in this case, by ESI qualifying, New Investigator (not ESI), Established and At-Risk status. The latter is potentially important because we’re most used to seeing them grouped with Established and so this will subtract out some of the usual pool.

First off, why oh why is NIH so bad at this? They list the numbers by PI status /stage within each FY and it is impossible to make any comparative sense of it. I had to graph the percentages for easy reference and viewing, this is what NIH has been calling the “funding rate”. Again, this is a per-applicant stat, not the per-application success rate statistic. What we can see first is that the Multi-Year funding of awards in FY2025 put a hurt on funding rate with ESI PIs funded at 18.9% and Established PIs funded at 19.6% compared with 26.1% and 27.3% in the prior FY2024 with the same overall NIH budget. And even that was a hit compared with FY2023, i.e., 29.8% for ESI PIs and 31.9% for Established PIs.

Still, you would think that NIH would want to make the case that ESI PIs did not suffer any disproportional hit clearer. Maybe? Of course the webpage (blogpost?) does say “we continue making considerable efforts to enhance support for ESIs pursuing NIH support” and lists five bulletpoints. Perhaps they do not wish to draw attention to the fact that there is nothing about this funding rate statistic from 2021 to 2025 that suggests that their efforts, quote-unquote, are doing anything in particular. The databook chart on funding rate for First-Time (including non-ESI NI, I assume) versus Established (including At-Risk, I assume) shows no particular difference from FY1998 to FY2024.

Image

Okay, for completeness sake, what about these other categories of PI? Well, the At-Risk PIs (who have had a major award from the NIH but will not have any such support in the next FY if they are not funded in the present FY) are perhaps obviously less successful, but do not appear to have taken any disproportional hit in FY2025. There might even be some small evidence that this group has received a slight bit of help in FY2024-2025 compared with FY2021-2023. Maybe I have just been noticing but it does seem like NIH got more robust about At-Risk PIs circa FY2022-2023, including specialized funding opportunities.

Most striking are the New Investigators who have never had major NIH support and do not qualify for ESI because they are more than 10 years past their terminal degree. This included 7,639 applicant PIs in FY2025, as compared with 6,065 ESI PIs, and the totals are roughly similar across FY2021-FY2025. So this is not some accident of small numbers. We can speculate wildly about “who” this NI population involves. Is it those folks who are on the same Assistant Professor timeline as the ESIs but they just happen to have had the ball bounce against them, years-wise, and are juuuuust over the limit? Are they folks who are deep into their careers, previously funded by some other source and are now getting around to trying the NIH? People who left the Intramural Research Program, Janelia Farms, or some foreign version of such non-competitive funding environments and are struggling to adapt to the bare knuckled fight of NIH extramural support careers?

Or are they unappreciated geniuses, who propose work now and again to the NIH, but can’t break through. You know, because of review cartels, intellectual group think, coastal elite hegemony, their maverick science views, DEI policies which overlook them, etc, etc, etc. I.e., another group of people who will find a sympathetic ear in the Office of the Director of NIH. Stay tuned…

Image

One final coda, because it drew attention on the socials. One person expressed themselves as surprised that so many applicants fall into the “At Risk” category. That is, PIs with prior success at winning R01 awards from the NIH who, if they are not successful in a given Fiscal Year will not have any major NIH award in the following FY. Including both those who have no funding as PI and those who may be in their final year of an award.

It is common. It has been 41% of applicants from FY2021 to FY2025, save for 40% in FY2024. This does not surprise me at all. I’ve written various things that touch on this, including how natural human sympathies and lack thereof can create an amplified boom-and-bust pattern in grant success. I’ve talked about the traffic holding pattern of grant revisions on more than one occasion. I have observed on many an occasion on this blog that in my experience the days of being able to “submit a grant when you need one**” are long gone. The days of regularly renewing one core grant just for doing a workmanlike job of producing are likewise in the rearview mirror.

The future under the new regime is likely dark for At-Risk investigators. There is an inescapable problem if we are going to keep driving new entrants with ESI policies with an unchanging, or decreasing, appropriation for the NIH. We have to have PIs who exit if we are to have new PIs enter. It’s simple math. I don’t like it. I’ve qualified as an At-Risk applicant and I’ve submitted at least one NIH application to a NOFO that was open to me because of this status. I didn’t get an award that way, as it happens, but I fully expect I may be At-Risk again before I am ready to retire from this business. That is the way the ball bounces.

But something has to give. We cannot prioritize the 41% At-Risk while at the same time stuffing the chute with more ESI. So the best case scenario is that talking heads talk, but the NIH just sticks to muddling along as usual. Quietly “saving our long time investigators” and servicing ESI more overtly, ending up with the same uncomfortable balance Program has struck since 2008.


*Beyond the scope of today, but the NIH adopted this to conceal/minimize the continued Ginther Gap

**I had one colleague observe as much to me as late as 2015ish. Sure, there has to be one or two people who have high assurance their every application would be funded. This sure as heck isn’t most people, and hasn’t been for my entire career.

A recent notice (NOT-OD-26-040) informs us that the NIH will no longer require advanced permission from Institutes or Centers for conference grants. For reference to this type of award, if you are not familiar with them, check RePORTER for R13 or U13 mechanisms. Although the NOT reaffirms to us that “NIH’s support of conferences is contingent on the interest and priorities of the individual Institute,Center, or Office (ICO)“, it also claims this is about reducing administrative burden.

As part of ongoing efforts to reduce administrative burden on the applicant community, NIH will remove the prior approval requirement for submission of conference grant applications under the R13 and U13 activity.

This follows a prior NOT (NOT-OD-26-019) that removed the requirement for prior approval of any grants with requests for $500k or more in direct costs. (For some reason the original policy still appears on a NIH online help page.) The new policy says that Letters of Intent will no longer be part of the application process for any purpose, despite re-stating that this was to assist Program with their burden.

NIH has occasionally requested LOIs within Section IV of the Notice of Funding Opportunity (NOFO) to help Institute, Center, and Office (ICO) staff estimate the potential peer review workload and recruit reviewers.

It then states that

Given NIH’s centralization of peer review processes to improve efficiency and strengthen integrity through the Center for Scientific Review (CSR), the LOI is no longer serving the same purpose to estimate ICO workload. To further increase efficiency and minimize applicant burden, NIH is removing the LOI from the application process.

So this part at least is about minimizing applicant burden. Sounds good, right? Also, the NOT informs us that:

Effective immediately, NIH will no longer require applicants requesting $500,000 or more in direct costs (excluding consortium F&A costs) in any one budget period to contact the funding Institute or Center (IC) before application submission. In line with this change, applicants are no longer required to include a cover letter identifying the Program Official contact which notes that the IC has agreed to accept assignment of the application.

In this case it doesn’t directly reference burden on the applicants.

One thing that is consistent about these moves is that it removes Program staff from a gate-keeping position. Previously, Program Officers could reject requests to approve a conference support application, a big budget R01 or applications for targeted funding opportunities that required a LOI approval.

No doubt some of you, Dear Reader, cheer this change. No longer are you subject to the whims of some Program Officer that hates you or has some buddies they need to take care of first, right? Why should they get to head off your chance to have peer review decide if this is an appropriate expenditure of NIH funds? Why should that long running Gordon Conference that bores you get to keep getting funded while your idea for a new and fresh scientific conference cannot gain support before even being allowed to try?

Why should your R01 proposal be subject to an entirely arbitrary $499,999 direct cost limit, especially when this has not been changed in decades and is something on the order of half of the spending power it was when your more-senior colleagues first got their big R01s funded?

I get it. I get your frustrations. I have had them myself, particularly where it comes to the less formal and workaday interactions where Program staff express themselves less than enthusiastic about my latest ideas, even if they do not have a formal way to gatekeep. As you know, I have occasionally observed that sometimes you have to just submit the grant even if the seemingly relevant Program officer isn’t supportive. I have noted how a good score from a study section has a way of countering programmatic reluctance. Gatekeeping can be a bad thing.

Gatekeeping can, however, also be a good thing.

With this new policy, the devil will most assuredly be in the details.

We are in a time in which it is very clear that NIH funding of science projects will be used, in part, to prosecute a political agenda (e.g., trying to tag Pete Buttegieg with lasting consequences of the Norfolk Southern train derailment). It is clear NIH funding will be used to prosecute the alleged health agendas of political people. There will be an explicit effort to award funds to some projects designed more to prove some theory associated with the regime (e.g., herd immunity, environmental causes of autism, weird ideas on healthy eating) than to illuminate facts. The regime is signalling quite overtly that they plan to award NIH funding preferentially to Red states and institutions that kowtow to their attacks on various things, regardless of peer review merit.

A smart apparatchik might understand that until they replace every serious person from Program, there is a risk that grants to support, say, a conference of anti-vaxxers, or mega R01s designed to prove the Tylenol theory of autism, or the health benefits of the inverted food pyramid would simply be disallowed. They might fear that proposals from traditionally lesser-funded States or institutions might not be automatically accepted if they are deemed lesser in merit or priority.

So one simple solution is to remove the gatekeeper function from the Program staff.

Keep your ears peeled. I bet we are also going to see a sea change in the informal discussions with Program. I bet POs are going to be less assertive about discouraging (certain kinds of) proposals, particularly from specific regions of the country.

Yes, the answer is yes.

It is extremely painful to have your grant proposal just miss the cut for funding on one version and then to have the revised version end up way out of the race or Not Discussed. This has come up with regularity in the online discussions of NIH grant review. It starts, at root, with the issuance of grant review comments to the PI in the summary statement along with the opportunity to revise (amend) the grant proposal and re-submit it for another round of review. It is accelerated by the fact that reviewers of a revised version of a grant have access to the summary statement of the prior version.

It just makes sense, to the uninitiated, that in a Just World a grant which is revised in light of prior criticisms of peers should be scored no worse, and probably better, than the original version. Right?

But the answer is also no. No, because the NIH has been trying for my entire time in this business to break peer reviewers of their impulses. To get peers to review revised grants without reference to how the prior version scored in a prior study section.

I started writing NIH grants when the rule was that one could only amend the proposal twice (i.e. to the A2 version), after which it had to be submitted as a “new” proposal. This followed an era in which A6 and A7 amended versions sometimes were funded. It was also during an era in which the impact of the NIH budget doubling was forcing a grant holding pattern. In which seemingly one’s proposal was only going to get taken seriously on the A1 or A2 version. (Oh, and btw, this was an era in which there was no ESI designation or funding policy. No R29 FIRST award set-aside for newbies either. Yeah.)

Image
Funded grants by revision status across Fiscal Years I think I got this from Jan 2011 CSR Peer Review Notes which has apparently been disappeared from the internet.

I lived through the NIH’s decrease of the A2 limit to A1, their attempt to ban resubmitting essentially the same proposal as a “new” grant and the subsequent (and current) backdown. A charitable view might say the NIH was trying to restore a sort of “fish or cut bait” stance of reviewers on original submissions, in an attempt to help speed funding to scientists who had the best ideas. A less charitable view might be that NIH was just trying to juke their stats on time to funding from the original submission of an idea.

I have been trained on many study sections that we are not to somehow benchmark the review of an amended (revised) proposal to the score / percentile / outcome of the review of the prior proposal. We are not supposed to indicate that we had reviewed the prior version of any such proposal. Any hint of benchmarking to a prior score often leads to SRO correction, and possible muttering from other reviewers as well.

When I was first invited to study section, the “Review Criteria Format Sheet” listed a series of headers which started with Significance. The second header was for Response to Previous Review (for revised applications). This resonated with the discussion, to my memory, in which the quality of the response to the review of the prior version was a primary point of comment. Eventually they buried the review template box for commenting on the quality of the resubmission to Additional Review Criteria down below Biohazards.

All of this was required because many of the people who are doing peer review of NIH grants are constitutionally and professionally likely to be instructors and explainers who literally cannot overcome their prepotent desire to help the applicant do better next time. It is why we entered the long path into this business in the first place. It is part of our professional workaday behavior to help people improve their academic work product. In short, it is who we are. Relatedly, NIH started inserting a box for Additional Comments to Applicant way at the bottom of the scoring template under Additional Review Considerations which emphasized reviewers “should not consider them in providing an overall impact/priority score“. This was supposed to be a sort of pressure-relief valve.

We are at another transition point in which the pain for “just-missed” scores and the corresponding outrage over the scores of revised proposals getting worse is ramping up. The 2025 assault on the NIH included a multi-year funding plan, now continued into FY2026 because Congress failed to pare this back in the recent appropriations bill, which inevitably reduces the number of new grants that are funded. This means more “just-missed” applications, particularly from the historical perspective of what scores should have funded. More revised proposals coming back in for review. The CSR is between what is supposed to be only two rounds of review with enhanced triage procedures- about 70% of proposals will not be Discussed compared with the prior 50%. Time will tell if CSR decides to continue this, me I suspect they will. That means not only scores going backwards, but probably many revised proposals that will be ND after being scored the first time.

Cue more outrage.

I don’t know what the best path forward should be. As I repeatedly note, peer reviewers at the NIH are driven first and foremost by a sort of diffuse “fund / don’t fund” binary and a lot of what is said in the summary statement is more a justification of this position than it is a sober quantitative addition of strengths and weaknesses.

I’ve had the pleasure of two low single digit percentile scores in my career. These were on revised proposals, an A2 scored at 2%ile and an A1 at 1.6%ile. The A2 followed a 21%ile A1, back in a time and at an IC where such a score was a strong “maybe” for exception pay. The A1 followed a 19%ile, ditto. I still assert that it would be very hard to show where my 2%ile and 1.6%ile proposals were objectively far superior to the prior versions. These were likely the top scores in the study sections for that round but there is no way in hell these were “perfect” proposals. Nor that they were objectively superior to a whole host of my other grant proposals over the years that got worse scores, from middlin’ within-payline (~8-9%ile at a certain time), to reach/stretch percentiles (hey, I’ve had pickups in those ranges) to NDs. Point being, the excellence of those scores reflects a set of reviewers saying “jeez, fund this thing already” to program and NOT them saying “this is objectively such an exquisitely crafted grant proposal that we cannot help but give it a fine score“.

Friends, we were already down to paylines (and inferred paylines) of 7-10%ile. You can check the funding data for the last few pre-chaos fiscal years for yourself. NCI said theirs was going to drop to the 4%ile range in FY2025 due to the multi-year funding requirement.

We will undoubtedly have an immense traffic holding pattern of previously reviewed grants stacking up.

Study sections simply cannot give them all within-payline scores to reward them for improving upon an already excellent “fund this!” proposal.

Way, way back I was introduced to the communication concept of Lying to Children in the context of the friction between scientists and science journalism. My take on that was related to the recreational use of drugs and messaging about the likely harms of such drugs, where my default stance is on the side of the facts as best we know them.

The idea of Lying to Children actually has its own wikipedia page:

A lie-to-children is a simplified, and often technically incorrect, explanation of technical or complex subjects employed as a teaching method. It is usually not done with an intent to deceive, but instead seek to ‘meet the child/pupil/student where they are’, in order to facilitate initial comprehension, which they build upon over time as the learner’s intellectual capacity expands.

Another way to look at this is to view it in terms of outcome goals. So, in the case of drug misuse disorders, drug dependence, etc there is a public health goal of trying to reduce the demonstrated harms to self, family, community. Does it matter if we subvert the most accurate depiction of the known knowns in service of convincing people not to start using, to use less, to seek help with cessation, etc?

I’ve previously touched on the frightening possibility that perception is everything in changing drug use epidemiology. I say “frightening” because it suggests that the real risks, the subject of my professional life, are somewhat tangential. I touch on our most fundamental lies-to-children in that post as well. Namely that “Drugs are bad”, meaning that if you try recreational drugs, even just a little you are going to be hooked into a spiral of drug dependency and despair. The Nancy Reagan “Just Say No” version of the “truth” about drug use.

I still do not have great answers to that. No matter how much time and effort I spend on trying to determine things related to “the real risks”.

The assault on NIH funded science that is now in a second year has a certain resonance with the Lies-to-Children friction between strict accuracy and outcome goals. Or perhaps it is between the complexity of reality and the outcome goals. I have observed more than once to people that the traditional petty complaints of scientists about one or other aspect of the NIH system have come back to bite us on the ass. Strict accuracy and/or complexity has a way of serving many masters, some of whom are frankly malign.

The IDC / overhead issue had, shall we say, a certain resonance with working academic scientists. People who should know better have contributed a lot of support to the notion that overhead inevitably means “waste”. We seem to have at least temporarily dodged a nasty bullet pointed at these very real costs of doing science, thanks mostly to academic institutions convincing key Republican Congress Critters it would be bad for their districts.

Complaints about supposed power cartels on study sections that hold back true innovators is being deployed at every turn by the representatives of the regime, including the current NIH Director. These sentiments are fueling the appointment of high level IC staff without the input of extramural scientists under the rationale that anyone with credibility in the present system is automatically suspect as a hide bound, anti-innovation, suppressor of new truths. A champion of group think. These sentiments are behind the decimation of NIH ICs’ Advisory Councils and intramural Boards of Scientific Counselors. These sentiments are, presumably, behind recent warnings that serving as an ad hoc reviewer on a study section is a problem for future appointment to a panel for a term of service.

Gripes about supposed ZIP code bias in grant award are traditional, and it is absolutely the case that some areas of the country receive a lot more NIH funding than do others. This reality is fueling both demands for geographical affirmative action from Congress Critters [see ~1:05 of this testimony of NIH Director Bhattacharya before Congress*] and proposals for block grants awarded to States.

The impact of other lies to children about the way NIH has worked are less clear. It is traditional, and has a lot of resonance politically, to wring hands about the future of science. To highlight decreased support for graduate students and postdocs. To insist we are about to lose all new Assistant Professors and therefore we need to double down on ESI support. Will this have positive or negative impact for our goals? Bhattacharya has himself mentioned the plight of younger scientists. It is, of course, unclear which early career scientists he will plan to support- perhaps this is all part and parcel of the agenda to support maverick scientists. AKA, COVID deniers, anti-vaxxers and those keen to “prove” political ideas favored by the current regime such as the causes of autism, the new food pyramid, healthy living instead of medicine, etc.

Diversity, Equity, Inclusion are other thorny issues. After all, this is the reddest of red meats for the present regime in their attacks on the NIH; this was Day One business. As you know, Dear Reader, I’ve been comparatively muted on these issues and in particular the Ginther Gap over this past year. Part of this was that there is no sense in trying to make headway on this issue right now, given the stance of the regime. Part of this was that the backlash against tepid, halting, foot-dragging NIH fixes was already happening prior to the election of the current regime. Some of it was that even I have limits to banging my head against a brick wall.

Most of this is that the various lies we tell to children on this issue, including about the NIH’s responses to the Ginther and Hoppe publications, have come around to harm the goal. In some cases, however, I fear that communicating the complicated story isn’t much better. There does not seem to be any level of truth that will not ultimately serve the wrong agenda when it comes to the racial bias in NIH grant funding. Any forthright effort to redress the bias, whereby Black PI’s applications were at a significant disadvantage, was made into grist for the assertion that any Black PI that was funded lacked merit.

This gained support from, you guessed it, voices inside the house. All of NIH’s excuse making was victim blaming. The cherry picking of results to show their glass was half full. The pipeline strategies which said the subjects of the reported bias were the real problem, meritless, and new Black PIs were needed in the um…far off future. The NIH’s protracted refusal to be forthright about how systemic features of review lead to circular, grinding conservatism floated the regime’s agenda. Their refusal to back down from their ridiculous assertions that peer review outcomes reflect Platonic merit with high fidelity (laughably down to the 5-percentile level of resolution) likewise fueled the attack.

In such an environment, is there any point to doubling down on sober factual analysis? Of pointing out that when 20% of a funding disparity is “explained” this leaves 80% of it unexplained? Of showing how many grants with white PIs were funded at percentiles far below that of any funded Black PI application? Is there any point in showing how NIH efforts on DEI fall far, far short of redressing the bias that was reported and replicated?

Is there any point in getting down to brass tacks about outcomes, regardless of any fine talk from Collins or Lauer and regardless of character testimonials from their biggest fans?

The policy currency, in Congress and without, appears to be the simple anecdote. Research trials on cancer patients in which cessation of NIH funding can be trivially tied to at least one patient dying. This is not the time for complicated discussion of whether a causal arrow can be clearly drawn from halted grant funding to one patient outcome, or how long it will take for experimental cancer trials to move toward medical success in a broad population.

The simple version of reality seems to be the strategy. The Lies-to-Children version.

As is the throwing of sure-lose agendas under the bus. For now. Because we pinky swear we will get back to that juuuuust as soon as we turn the corner on the current regime’s chaos. On the real problems. Because we need to move forward. On what is really important.

For some.


*Note that the exchange with the Indiana Senator finishes with the assertion that a lack of geographic diversity of NIH grant funding leads to “group think” by Bhattacharya [see 1:07 of the hearing]. Which, of course, the regime finds to be the only aspect of homogeneity of funding that leads to such invariant thinking. Despite the fact that a large number of the scientists currently employed in lesser-funded geographic regions received the bulk of their scientific training in the “group think” geographic regions [hat tip].

NOT-OD-26-029 informs us that as of May 25, 2026 (applications slated for Council in the January 2027 rounds) there will no longer be non-standard application dates for grant proposals that involve HIV/AIDS research. This has been a constant during my career. As the notice indicates, the special due dates / deadlines were established in 1988 [PDF], i.e., long before I started paying attention to NIH policies.

The HIV/AIDS related proposals were due late in the cycle, i.e., May 7* for Cycle I, Sept 7 for Cycle II and Jan 7** for Cycle III at present. The corresponding dates for new R01s are Feb 5, June 5 and October 5. So you can see how this compresses the review cycle for SROs and sure enough, NIH is begging overwork as the motivator:

HIV/AIDS applications must be identified, segregated and their status validated in manual processes outside of the normal referral stream. In 1988 NIH had to apply these processes in the context of reviewing about 32,000 grant applications. In 2025 NIH will receive over 102,000 applications and managing the additional administrative burden of the non-standard deadlines is no longer practical.

Fair enough, although I’m pretty sure they are giving us all-NIH application numbers and not HIV/AIDS numbers here. I’d have to dig a bit in the databook to figure out the number requiring “manual processes”.

I don’t see any obvious reason to be suspicious of this move in the sense of it diminishing investment in HIV/AIDS related research. We will have to stay tuned to see the degree to which these proposals are reviewed in the same, or similar, study sections as in the past. I’m haven’t paid much attention to these reviews for a long time and I don’t have a good sense of how many are reviewed by IC in-house study sections (now terminated or moved to CSR), by Special Emphasis Panels or by regular old CSR standing sections. This will, of course, be all-critical to PIs who conduct HIV/AIDS related research.

As usual, NIH can’t help including a bit of gaslighting. They say “The advantages of the non-standard dates to AIDS applicants were indistinct“, which is of course ridiculous. For those investigators who pursue both HIV/AIDS and non-HIV/AIDS research having an extra deadline for submitting a proposal facilitates keeping the hopper full. Sure, you can wave your hands about how deadlines aren’t anything special for someone who plans ahead but…come on.


*May 7 would appear to be the final HIV/AIDS deadline.

**Yes, having a due date in early January after the winter holiday season was annoying. Speaking as someone who has submitted at least one HIV/AIDS related NIH grant proposal in the past.

As you know, Dear Reader, the current and following rounds of NIH study section meetings have adopted temporary* new triage/streamlining rules.

Previously, approximately half of proposals are Not Discussed at the meeting, based on the average of preliminary overall impact scores from the assigned reviewers.

This has now been extended such that approximately 70% will Not be Discussed.

Along with this change, NIH has instituted a new “Not Discussed, but Competitive” designation for the ~middle third, ie the top half of the ND.

The first set of scores are being returned to PIs now and I’ve heard some chatter that all of the ND proposals are first marked in eRA Commons as ND. The “Competitive” designation has been showing up a day to many days later.

This has led to some folks asking if I know what is going on.

From what I can tell, the SRO dashboard or whatever still only allows them to enter ND. The SRO then has to send their list of “Competitive” NDs to some sort of eRA commons person who can make the adjustments.

So calm down, everyone, there is not some sort of post-meeting re-juggling of the ND list.


*I remain skeptical that this will be temporary. Cutting meetings down to one day will be seen as a win. Alternately, getting through a larger set of apps assigned to one section will permit the reduction in total number of sections will be seen as a win. And the “competitive” ND is supposed to allow greater Program flexibility to pick up grants. This dovetails with the recent diktat about strict-payline ICs no longer doinn in that.

Design a site like this with WordPress.com
Get started