Looking at Recent Data on R21 and R01-equivalent Grants


One topic of frequent interest to NIH leadership is how R01-equivalent awards compare to other research grant awards. The R01 is the standard mainstay of NIH’s research portfolio, and the oldest grant mechanism in use by NIH. As those familiar with the blog and RePORT know, we usually look at R01s in conjunction with other awards providing similar support analogous to an R01, which includes R37s or MERIT program awards. Of the R01-equivalent pool however, R01s make up the overwhelming bulk of these grants so while we call them R01-equivalents for accuracy-in-reporting reasons, it is highly appropriate to consider R01-equivalent data as representative of R01 trends.

Over the past years we’ve been looking at trends in R01-equivalents compared to trends in awards through the R21 activity code. R21s are exploratory/developmental grants which encourage developmental research, without requiring preliminary data and support early and conceptual stages of project development. After the R01s, R21s are the second most-awarded NIH grant, and R21s receive the second largest percentage of all NIH research project grant dollars. Let’s look at some data on R01-equivalents versus R21s from the NIH Data Book on RePORT:

FY2012 FY2013 FY2014
Number of R01-equivalent grant applications: 29,627 28,044 27,502
Number of R01-equivalent awards: 5,437 4,902 5,163
Success rates for R01-equivalent applications: 18% 17% 19%
Total amount of funding that went to competing R01-equivalent awards: $2,335,391,619 $2,072,911,352 $2,290,939,281
Total amount of NIH funding that went to R01-equivalents (competing and non-competing): $11,021,860,936 $10,174,867,296 $10,359,458,392
Number of R21 grant applications: 13,743 13,229 14,331
Number of R21 awards: 1,932 1,771 2,013
Success rates for R21 applications: 14% 13% 14%
Total amount of funding that went to competing R21 awards: $422,632,904 $382,713,272 $433,814,063
Total amount of NIH funding that went to R21s (competing and non-competing): $774,963,587 $763,384,905 $807,267,070

In FY2014, the number of R21 applications and awards continues to grow and increased to its highest ever since 1998, unlike the pattern for R01-equivalents. While there is no coordinated initiative to encourage scientists to apply to the R21 program, it can be useful for early-stage project development, when the scope of the project is more suited to a shorter, more exploratory award that does not require preliminary data.

We also looked at what proportion of Research Project Grants were R21s in fiscal years 2012 through 2014, and see that this remained fairly stable, at 21.4, 21.3, and 21.8% for FYs ’12, ’13, and ’14 respectively.

The data presented above may help you in deciding which mechanism of support you should consider when applying to NIH. One urban myth has been that R21s have a higher success rate than R01 which is actually contrary to what the data show. Regardless of success rate, however, it is always good to try to match your application with program objectives and let your aims drive the activity code, so to speak. As always, reaching out to an NIH program director before you submit your application is your best bet if you need additional advice on which avenue of support you should seek.

Keep your eyes on the blog as we will continue presenting info and data on the R21 and other NIH programs.


  1. Comparison of R21 success rates with NIH-wide R01 success rates is inappropriate, since many institutes have stopped supporting R21s. How does the success rate compare to R01s when you focus only on those institutes that support the R21?

    1. The ICs that do not participate in the R21 parent announcement support R21s through separate PAs or PARs, so most ICs are represented in the published data.

  2. It seems entirely backward that R21s are less numerous. Wouldn’t it be smarter to do a lot more short term exploring and developing before throwing a full R01 at a question? It would be cheaper and more efficient in a lot of cases. Maybe what you should do is award a lot more R21s and give bonus points to R01 apps that follow a related R21.

    1. Not necessarily. Short-term grants bring instability to a research program – most postdocs are longer than 2 years, and grad students typically require 4 years of support. I think having the long-term security of a 5 year grant is critical to the health and well-being of an individual PI’s research program.

    2. The R21 vs R01 discussion is interesting. Perhaps a middle solution would help. A 3-year R01, for example. There are some new mechanisms that are using this approach, which is refreshing.

  3. Most RO-1s are reduced to 4 years. R21, although with lower budget seems attractive to those who have ideas without much preliminary data support. The high risk/high “pay” aspect seems to be lost in stringent review by reviewers who forget the logic behind R21 applications or the requirements. Many reviewers also seem to think that the R21 money comes at the expense of R01 and are reluctant to give R21s a different look.
    I also am at a loss to understand why some institutes have abandoned investigator initiated R21 applications.
    A separate panel to look at only R-21 applications might bring more novel and exciting ideas to the table.

    1. The assumptions that R21s do not need preliminary data is pure eye wash because those without prelim data rarely get funded. Which means that the really risky “high payoff” grants are not getting R21 support. R21 has become a smaller R01 in which applications with lesser prelim data than R01 are getting funded.

  4. I would suggest abandoning competing applications. The ideas are usually several years old and anything new could be easily submitted as new application. Besides, after 4 to 5 years, it might be worth rethinking of the grant potential and bring fresh insights.
    You also avoid the harsh judgement on productivity, etc!

    1. Productivity is a factor even for a new application. Of course, the investigator’s job description (i.e. how much time they have to devote to teaching, clinical or administrative duties) is not factored in to this equation. It would be tricky to do this since standards vary so much.
      What is more objective is the important ratio of productivity versus prior funding obtained. This is a measurable and is almost never factored in. For example, say you are comparing the productivity of someone who has had two or more R01 equivalent grants in the 5-7 years prior to submitting their proposal versus someone who has been scraping along on small foundation grants to get enough data for a viable R01. The productivity of the better funded investigator should be a lot higher (I mean a lot). Most reviewers I have seen, however, do not grade productivity on a funding curve. Certainly, reviewers are not provided any metrics on investigator productivity/funding and few have the time to weigh these factors. One would think that program officers would have plenty of time to generate such data in the many months between grant submission and review. Such an endeavor could lead to fairer review of proposals. It would also expose some people who are always quick to submit the next grant and seldom follow through on the ones they have obtained.

    2. Yes but I have spent 20 years working ona treatment and no have to take certain steps to take a fundamental discovery to translation – those steps cannot be avoided and are arguably the very thing NIH is seeking. A failure to fund those essential steps in treatment development leaves me having to bang my head against a wall untill NIH and congress get real again. The failure of study sections to score honestly by NIHs own definintion of what a score of 1-3 means because of the NIH SROs command to spread scores is just toxic – a proposal worth a 1, 2 or 3 by NIHs own criterion should not be scored 4-9 just to spread scores. All in all the scoring is a shambles and the reviewers know it. I am seeing more and more reviewers growing impatient with doing this. If you want applications listed in rank order instead then just say so. Bottom line? – be honest about scores and be honest what the actual rate is for funding of grants under review – stop the nonsense with mixing in non competitive funding to hide reality. That’s just political nonsense and no one is fooled by it.

  5. This table is scary: It looks like the NIH gives most of its $ to non-competing grants…Any comments?

    1. Non-competing awards represent the funding for the years beyond the first year (which is competing) on any multi-year grant. When the NIH commits to a project, it commits for the entire period of its funding, hence the non-competing years (2 up through 5, depending on the mechanism), but usually pays it one year at a time.

    2. The”non-competing” grants are the yearly award cycles for grants that were awarded. Unless you show poor progress or something else happens, then you will be awarded your next year of funding under both the R01 and R21 mechanisms. They are just showing what their payout is for those grants versus “new” awards.

  6. Pay lines (the percentile needed to get funded) have remained at historically low levels for many years now. Therefore, fewer applicants have the sustained funding necessary to generate enough data to provide sufficient proof of principle to justify 4-5 years of support in a full R01 application. Therefore, more applicants opt for the R21, which has a lesser requirement for preliminary data or published productivity. This dramatically drives up the number of R21 applications, which to a lesser extent drives up the total number of R21 grants awarded, but it also drives down R21 success rates.

  7. The real payline for R01s at individual institutions is nowhere NEAR 19% or even 17%. In the past few years, grant proposals with percentile ranks of 9-11% have been above the payline and gone unfunded. One program officer shared with me that hit rates for R01s at her Institution were more like 5%. These numbers are terribly misleading. Would that it were true that paylines are back up to almost 20%!!

    1. Success rates count the original and -A1 and higher submissions as the same grant. Also, sometimes institutes pick grants above the payline for funding, depending on funds available in a program, or priorities of the institute. This makes the success rate metric about double the average payline.

    2. I agree. The chances of an R01 grant being funded are far lower than 19%. What is the source of these numbers? If it were purely unsolicited R01’s funded/unsolicited R01’s submitted, that number would be significantly less than 10%.

  8. The keep using the same accounting gimmick to arrive a “rate of success” of 19%. Really?. The real rate of success is the funding pay line for each Institute. The problem is: that figure does not look good. NIH has become a big joke, in which the “connected” ones succeed.

  9. The % success for R21s vs. R01s is meaningless unless categorized by institute. I just got an e-mail from NIAID clearly stating that their payline for R01s is MUCH tighter than for R21s.
    In contrast, NIGMS has a better (and more flexible) payline for R01s but doesn’t grant R21s at all.

  10. It looks like only a third of the whole NIH budget is going towards R01/R21? Any explanation on this?

    1. Yes, I’d like to know why this is too. Also, I’d like to know why the NIH, in many cases, seems to completely disregard the prior grant productivity/history of grant applicants. I’ve seen many NIH funded scientists who published zero or 1 paper after 4 or 5 years of funding. Established PIs are getting around the system after low productivity on a 5 or 4 year grant by just applying for other NIH grants. NIH/reviewers are not looking at the past funding history and related success/productivity of the applicants. Millions upon millions are being wasted just because of cronyism and lack of accountability….. which could therein help solve a lot of the funding crises.

  11. This data is bogus. An NIH PO shared with me that the success rates are calculated as

    # awards/# grants. Which sounds good, but NIH considers noncompeting years as separate awards. So an R21, when funded, counts as an award the first year and an award in the second year. So we can assume that about half of the “awards” for a given year are the second year of the previous R21 grant. So cut that success rate in half. For R01s, divide NIH’s fake success rate by 4 (average duration). Looking pretty grim, right?

    NSF does the same thing, but the grant counts as 1 award only. Instead they count every undergraduate supplement (where they give you a small budget for more UG students) as another award. So they can fund 1 grant, give the PI a couple of grand for 3 students, and magically they just funded 4 awards!

    I think if NIH/NSF actually published how many grants are submitted and how many of those same grants got funded, it’s very very low. # of awards is wall street accounting at its best.

  12. I am a big fan of R21s, but think they should be increased to 3 years and at a minimum of $175K/yr. If R01s were reduced to 4 years, there would be plenty of cash for this with lots of $ to spare.

    1. Yes!!!! Totally agree…. too many scientists with 4 or 5 years of funding are sitting back and producing nothing with no accountability. 3-year R21s would be a good compromise. Come on NIH, academic science doesn’t have much more time for major changes to occur. Something needs to be done and quickly. Step 1… accountability of all grantees and no more cronyism…. it’s killing basic and translational science.

  13. What is the rationale for having 81% or R01 grant applications being rejected and wasted effort? Because there isn’t enough money given to the NIH by Congress with a $30 Billion budget allocation annually ? Really? If we as a nation spend $3 Trillion annually on healthcare doesnt that make the NIH funding just a miserly 1% in R&D investment? What does the NIH culture (ie all of us) consider it normal that the extraordinary amount of work going into the preparation by 81% of skilled researchers and wasted effort each round can be simply discarded? What are the percentage of grant applications that are worthwhile and should be carried out to improve US healthcare. Second question… what is the proportion of R01 grants that are awarded to basic science projects as opposed to clinical research projects that could prevent expensive complications that eat up most of the $3 Trillion in healthcare spending? I heard >90% goes to basic science? True or false?

  14. One big issue is how much of the R01 & R21 funding is squandered in institutional overheads (IO). A list of IO % + institution should be made public. A 2nd issue is what % of R01 & other NIH grants is paid out as salary to the PI. Since some institutes pay nearly 0% salary, it’s very likely that an entire R01 goes to pay the PI. Coincidentally, these institutions have up to 99% overhead, so it’s 2 R01’s to pay the salary of the PI!

Before submitting your comment, please review our blog comment policies.

Leave a Reply

Your email address will not be published. Required fields are marked *