K99-R00 Evaluation: IC distribution

Jul 22 2014 Published by under Uncategorized

The K99-R00 program is an NIH-wide program but, as is typical at NIH, each institute and center has considerable flexibility about the details about how the program is administrated. For example, for the two cohorts of K99 awardees that I have been examining, the number of K99 awardees ranges from 1 for the National Institute (then National Center) for Minority Health and Health Disparities (NIMHD, MD) to 54 for the National Cancer Institute (NCI, CA). The number was not simply proportional to budget size. For example, the number of K99 awards from the National Institute of Allergy and Infectious Diseases (NIAID, AI) was 13, smaller than five other institutes despite the fact that NIAID has the second largest budget at NIH. Moreover, as one would expect by chance, the fraction of women and men among K99 awardees varies from IC to IC. This may be relevant to understanding the disparity between women and men in the probability of transitioning from an R00 award to an R01 or similar award.

These data are summarized graphically below:

IC distribution


The size of each circle is proportional to the number of R00 awardees from each IC.

These data may be relevant to understanding the gender disparity. For example, both NIAID and the National Institute for Neurological Diseases and Stroke (NINDS, NS) have percentages of men among R00 awardees that are slightly higher than the NIH average but all of the women from these institutes who have received R00 awards have gone on to obtain R01 or equivalent (DP2) funding through the present.

Understanding the origins of the gender disparity between women and men going from a K99 award to R01 or equivalent funding is important for determining what policy adjustments should be considered.

11 responses so far

  • lurker says:

    Thank you, Datahound (Jeremy) for taking a hard look at the K99/R00 program. As a R00'er myself, I take tiny comfort I'm not the sole loser having still not yet gotten an R01, I'm just part of the other 50%-60% where an R01 is still out of reach 2-3 years out from the end of the R00. The 3yrs of the R00 is too short a time to reach R01 level productivity, yet in my situation, the study sections are repeatedly dinging for not productive enough to merit an R01. Perhaps the bias in determining what is productive enough for an R00 ESI versus non-R00 to merit an R01 could contribute to the gender disparity as well.

    No need for any pity for me, I know the landscape is rough for everyone. I'm grateful that at least I got the small R00 bump, which is the only reason I'm an AssProf now. Longevity will be an issue though.

    Jeremy, you said in earlier posts that the R00 program is a huge success because the 40-50% of R00's do get R01's, which I take is you being half-glass-full when comparing this to the 10% success rates for everyone else. Is it really a success?

    If the extramural program is like a patient in sepsis, 50% of the time can keep good food down, the other 50% pukes it out, I'm not so inclined to claim success and health with this one aspect of the program since other metrics are so sick right now.

    What if we got rid of the K99/R00 and move everyone to a MIRA-like program? Even the playing field with no more >$1million BSD grantees (not counting their HHMI support too)?

    • datahound says:

      lurker: The reason I consider the K99-R00 program to be a success (at least by some measures) related to expectations and comparisons. NIH has not had many programs that transition students or postdocs into academic positions with ~90% success and then show ~50% (or perhaps more over time) transition to R01 funding. This is particularly impressive in these challenging times.

      I agree with you that the extramural program is quite troubled at present and there are many aspects that need to be addressed. The most fundamental problem is that there are too many people competing for funds available for individual R01-like grants. This imbalance could be addressed by increasing the overall budget (not likely in the present political environment), transferring funds from larger programs to R01s (after evaluations to determine to see if these larger programs are truly meeting their goals), and/or distributing funds more evenly among investigators (as you suggest). As you probably know, I have been an advocate for putting processes in place to make sure that well-funded investigators are producing at levels consistent with their level of support (from all sources).

      I think the new NIGMS MIRA concept has considerable potential (depending, as always, on the details of how it is implemented). A major challenge is that this is an NIGMS-specific idea and other institutes and centers likely have different perspectives.

      At any rate, focus on and enjoy your science and best of luck winning additional support.

    • DrugMonkey says:

      Careful not to confuse paylines, success rates and per-investigator funding rates.

      Oh and just so we're clear, this new MIRA nonsense is for the elite few. This is a major screwup if you ask me. (Which they don't)

  • lurker says:

    What "problem" would more MERITs solve? As I see it, only the BSDs are getting MERITs right now anyway, not ESIs nor the proletariat labs, which are getting the short end of the NIH funding stick right now. So more MERITs to make the lives of BSDs like you, DM, more cushy? If it means more time for your blog posting and twitter feed (which I love, btw), well okay then...

    I see DM's point that Pioneers and Innovators tend to already be directed at the elite (or damn lucky) few and do not help the too-many-mouths problem, just makes NIH Extramural look like they are doing "something". MIRA might also be another boondoggle, but what I think there is promise if the program sets funding ranges from a minimum of $150K to maybe a max of $750K, which dials up and down depending on productivity and not say tied to conjuring up discreet projects.

    If MIRA is done right, it could promote more chances for noobs like me to enter and stay viable, allow buffering of short dry spells in more established labs, and cap BSDs from mismanaging the boatloads of dollars they get just because their work makes it into the GlamourPubs.

    • datahound says:

      From my perspective, there are multiple problems being conflated here. The major goal of MIRA (as I understand it) is to better align how science generally works (research program-based as opposed to project-based) and also to deal with the fact that many investigators need multiple grants (due to the fact that grant sizes have not keep up with inflation) to maintain even a modest laboratory in many settings and fields). The implementation as proposed will require eligible investigators to trade higher levels of funding for more stability in support. Whether this will actually happen, of course, remains to be seen. Again, as proposed, MIRA would also potentially be available for new investigators, potentially providing them a path to a more stable start to funding early in their careers. Of course, all of this depends on how the peer review system deals with this mechanism.

      MIRA is unlikely to address substantially the too-many-mouths/no-enough-money challenge unless other ICs buy into it and shift funds from large programs to investigator-initiation R01 scale research (or Congress loves the idea and provides more money).

      Note that the NIGMS Feedback loop post on MIRA specifically states "It’s important to note that MIRAs are not intended to be a method for supporting only a perceived elite group of investigators or promoting only high-risk, high-potential-reward research." I know that MIRA-like program is something that the NIGMS Director has been discussing and pitching for years as a way to better capture how laboratories operate in his experience and is not just a reaction to the NIH-wide discussions about "HHMI-like" funding.

      • drugmonkey says:

        I hope that you are right about FC's intentions. I am only going by what is reported or blogged or twitted about his thoughts.

        We won't know how much greater stability and an increase in the basic unit of funding (to match what the full modular could buy when it was put in place?) will reduce churning until we try it. But my guess is that it will have a big impact IF it can be applied to large numbers of PIs. It will do nothing if it is only applied to a small number of PIs.

        2-3 R37s per PO would be a lower limit to even budge the needle from where I sit.

      • drugmonkey says:

        Crap, I misread you. NIGMS always marches to a different drummer no?

        • datahound says:

          Yes, NIGMS does tend to try new ideas out. I think FC is much more sympathetic to the heavy hitters who are now struggling for the first time. Jon Lorsch is much more sympathetic to the established but more modest researchers with relatively small labs and lots of personal involvement who want to work on a problem (his work is on the mechanisms of eukaryotic translation initiation and its control). He feels it is artificial to break such programs up into projects when the pieces are stronger when they fit together.

          NIGMS is proposing a pilot with NIGMS investigators with 2 or more R01s (interestingly 80% male by my estimation) and with ESIs. Take a look at the RFI or Feedback Loop post and comment away.

  • […] including the transitions to R00 and R01 grants, gender disparity in R01 transition probabilities, differences between NIH institutes and centers, and gender differences between R0o […]

Leave a Reply