Who Are We?


As part of our effort to characterize the community that is supported by NIH, we looked at the degrees held by NIH-supported principal investigators (PIs) and whether the composition of that population has changed over the years. As you can see below, ~70% of PIs on NIH research project grants hold a PhD, and this proportion has remained steady for the past 25 years. Most of the remaining PIs have either MD or MD/PhD degrees, with a small percentage unknown or holding other advanced degrees (for example, DVM, DDS, and DMD). The percentage of MDs has declined over the years from ~20% to around ~17%, while the percentage of PIs with a MD/PhD degree has increased from ~8.5% to almost 11% in the same timeframe. 

graph of NIH-supported PIs by degree type The total number of PIs has increased in this timeframe, and the range of NIH training and career development awards has evolved (see details of specific programs). Notably, we established several new career development awards for clinicians pursuing patient-oriented research in the late 1990s. It seems, however, that the system is remarkably stable. It will be interesting to see if these proportions remain stable over the course of the next few years as science continues to advance and translational research receives emphasis through the National Center for Translational Sciences (NCATS). Meanwhile, we’ll be taking a look at these data and others in our quest to model the future of the biomedical workforce. We’re going to be busy!


  1. The degree mix of NIH PIs is more interesting than microMike suggests or Dr. Rock has shown in this further effort to characterize the NIH community. A little more history would be illuminating. In the mid-1980s, I was trying to help my boss (a member of an NIH Committee asked to advise on the lack of clinical investigators as first noted by Wyngaarden) identify possible reasons for the situation that went beyond the usual suspects of salary differentials, medical faculty emphases, etc. At that time and for many previous years, the NIH Data Book reported the degree mix of NIH program staff and leadership. It had shifted to a PhD dominance in the late 1970s, if memory serves. My boss found that distinctly uninteresting as a possible contributing factor to the decline in the number of MD investigators funded by NIH, but I have never forgotten it. I believe, still, that it is an important consideration. Whether causal or not I leave to others. I suggest, however, that if Dr. Rock were to extend the analysis back 25 years and compare the degree mix of NIH program staff with the degree mix of NIH PIs for the entire period, 1961 through 2009, the rise of the PhD as PI would exceed the decline of the MD as PI. The MD/Phd PIs are a special case that likely will track their own curves as NIH staff and NIH PIs.

    History is replete with examples of minor events leading to major changes. No one appears to have noted during NIH’s “Golden Years” the possibility that shifting the degree mix of NIH program staff might have a similar effect on the degree mix of NIH PIs. Now we call them “tipping points.”

  2. Whoever these MDs and MD/PhDs are, they are smart as they can always make their living in the clinic. Whoever still wants to become a PhD is crazy in my opinion, unless that person prefers to be an eternal postdoc. I quite frankly would prefer to see statistics on how many PIs had and will have to close shop due not not getting funded. When I look around me, there seems to be a growing number. NIH is still pretending nothing big has changed and everything is just peachy.

  3. While we are wondering who we as investigators are….
    What percentage of the PhDs are RN-PhDs and how has that representation changed over the same period of time?
    I’ll leave it to Others to speculate what the omission means.

  4. I have 2 comments –
    Firstly, it appears that the number of grantees with “other” degrees has flatlined and is negligible, while the number with “unknown” degrees has declined – as a DVM/PhD with R01 funding I find this very troubling, as NIH is clearly (though obviously not intentionally) driving professionals with a different expertise set out of the game. Diversity in research approaches is essential, but is becoming increasingly incompatible with success. Teaching at a vet school, I find it almost impossible to get the students interested in research careers, and given these data, it is small wonder.
    Secondly, I would like to see these data conflated and/or correlated to other metrics broken down by degree type:
    Time from award of terminal degree to first R01
    Number of R01s per PI
    Success rate of renewal and/or subsequent R01 applications.
    Without some kind of long-term outcome measures like these, the data are meaningless – to award grants to PIs that are ultimately unsuccessful as scientists, or who go back to more lucrative clinical careers is a waste of money.
    And I agree with the last comment – why would anyone in their right mind go to grad school any more?

  5. I agree with nearly all of the points suggested by the individuals above. I have an additional point I would like to put forth.

    I particularly agree with Ian Davis and micromike that these data are not as useful as could be. A simple fact is that PhD’s are specifically trained to do the science whereas MD’s are specifically trained to practice clinical medicine. It’s great when MD’s can get some training in science and research and persue research funding, but that’s not what they were initially trained to do. Thus, it’s not a big shocker that the vast majority of funding goes to individuals who hold a PhD. It would be interesting to see some additional metrics regarding the outcome of the funding (e.g. time from terminal degree, # of R01’s per PI, success rate of renewal). Also, as important, it would interesting to know some of the data regarding who is submitting for funding vs. who is getting funding. I do not know what those numbers are and have not looked for them but additional information could be gained by knowing who is applying, who is getting the funding, and who is keeping the funding (renewal). These isolated data could be much more valuable.

    My last comment is in response to Maveric. If I knew then what I know now…I would have become an auto mechanic…but no worries, according to NIH everything is going great.

  6. I found this article somewhat misleading -while the authors list the fractions of PI’s by terminal degree, they do not provide key information on the dollars awarded per grant to clinicians vs basic scientists. My own observation as a member of several study sections over the years is that the average dollar amount awarded to clinical (read MD!) grants vastly exceeds that for any other type of grant (read PhD!). Most medical centers have developed an addiction for the overheads on large NIH funded clinical grants – upper administrators like nothing better than a “multi-center” clinical trial whose hub is based in their center. In fact, even though there is one PI on such grants, there may be a dozen or more MD’s whose salaries are partly covered by such grants, so the total number of MD’s actually receiving NIH funds may actually exceed the total number of PhD’s funded by NIH. I would like to propose crowdsourcing this question: How many PhD’s vs MD’s receive ANY funding from NIH at your medical center?

  7. Why not you also publish how many PhD vs. MDs are the Directors of Centers, working in leading positions within CDC and other NIH establishments? Merely publishing statistics regarding grants by MD/PhD are not sufficient. Because there way more PhDs than MDs. and way more PhD-PIs applicants. A better statistics would be PhD-PIs divided by # of PhDs applied for a grant vs. MD-PIs divided by # of MD applies. I would highly encourage you to revise your post and say that absolute number by no mean provides the full picture as opposed to rates and proportion. Absolute number of deaths in CA cannot be compared with the absolute number of deaths in Vermont! without know the underlying population. Would you also provide some statistics who many of even those PhD guys were right/left hand of the healthcare industry! or basic science. Historically NIH is very biased toward clinical sciences vs. prevention vs. research on the social and distant causes of disease.
    NIH would love to spend millions on finding a cure for a disease once it is developed but wont spend a $1000 to identify effective ways of prevention. If it would not be to the public health efforts of the late 18th and 19th century we would be still looking for new antibiotics for cholera! and it would on the top list of NIH!

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

Leave a Reply to Shams Rahman, MD, PhD Cancel reply

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