Eating the Seed Corn in the Health Research World, Continued

I have written before about the extraordinary pressures on young scientists and how their low chance of securing research funding is driving many of them to quit. As our corps of NIH-supported researchers gets older and older, we become less prepared to meet the health science challenges that will arise 10 or 20 years from now.

NIH has just released a startling 46 second video that conveys the change in the age distribution of NIH supported principal investigators from 1980-2010.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

34 thoughts on “Eating the Seed Corn in the Health Research World, Continued”

  1. I have essentially given up. I was well funded by NIH in my previous position, where I advised 3 PhD students and one MD-PhD as an Assistant Professor and published enough in good journals to get promoted. But the money was not green enough for certain administrators and that was that. I have a good job now primarily teaching medical students, but the collapse, and that is what it is, of NIH and NSF support for all but the chosen few is destroying the research community at the root. At an international meeting in December I heard the new Director of NIGMS answer a question about funding an award with a 10th percentile rank (90th in the usual understanding of the term) over an award with a 7th percentile score (93rd), the reason being that the lower scoring application was funded to "save" that particular applicant's research program. Hah. I have served on many review panels. The top third in any pool of applications can be recognized, but rank order within that group is spurious. It depends on nothing other than whether the primary reviewers "liked" the application. Number 10 and Number 7 have equal merit. The next third is recognizable, too, and consists of applications that should get funded after revision. The bottom third is hopeless and will remain so, probably. If you want to divide the pool into fourths, OK. That still means 50% of applications will get funded, which is as it should be. The crunch, to which the current administration is oblivious, has also led to the pernicious practice of giving program directors the authority to pick the "winners" in advance. This CANNOT be done. Oh, and my former institution? Hanging on by its fingernails with a half-empty new hospital building that was supposed to be filled with the help of "indirect costs." Alas.

    1. Either you and I have met, or there are a lot of people like you (and me), and a lot of institutions hanging on by fingernails with half-empty hospital buildings, or "collaborative research buildings" that were supposed to be filled with soft-money scientists to help with indirect costs.

  2. That' frightening. The (perhaps) good news is that the NIH recognizes the problem. That recognition may prompt action to move the distribution in the other direction.

    1. Yeah, seeing "NIH" and "prompt action" in the same comment throws me off a bit. I'm guessing you are not a biomedical scientist.
      So much of what NIH has done, and has failed to do, has created this problem.
      Shirley Tilghman, among others, called attention to this problem in 1997! She tried to sound the alarm, and made a direct link between NIH decisions and indecisions, and these negative outcomes. She and her commission made some recommendations to remediate these issues, They were ignored. Then in 2011, the Biomedical Workforce Working Group that she lead raised the issue again, and again made some recommendations. And again, NIH responded with "Yeah, sorry, no can do. But here's some other stuff we can do that might not solve this problem, but at least it is something, so its not like we aren't doing something."

  3. The funding cuts are bad for researchers, especially young ones.. But why is it a bad thing for the age pyramid to mature towards something resembling a steady state? The 1980 distribution was not sustainable, short of medical research growing until it ate up the whole economy, or institutionalised triage as in the military (up or out).

    1. How is the 2010 distribution more sustainable? Can researchers in these fields come and go at their will? If not, the triage is already here!

    2. It takes about 8-10 years to turn a college grad into a mature scientist fully capable of training others. 12 years on the outside. I don't see any "eating seed corn" problems–50-year-old scientists have plenty of career left in which to train the newbies. It's more like "lost generation."

      That being said, I dropped out of science 25 years ago, and I'm glad I did. Even back then, things were way too tough on the young 'uns. And they were much better back then than they are today. I can still remember enough to talk shop with my former colleagues (fun!), but I try to avoid talking about funding and similar horrors. It's just too grim.

      The public policy problem, I think, isn't so much in basic research as it is in applied research and engineering. It's hard to recruit top talent to work for The Suits if said talent is valued far less than The Suits. Hence, the lust for cheap H1-Bs in Corporate America.

      1. "It takes about 8-10 years to turn a college grad into a mature scientist fully capable of training others. 12 years on the outside. "

        And it's not even that; it's when the Ph.D. becomes capable of doing good research, which I'm guessing is far sooner than they become a mature scientist capable of training others.

      2. "It's hard to recruit top talent …"

        They don't really want 'top' talent; they want cheap labor who can perform the basic tasks.

      3. Unless scientists learn much faster than actuaries, this seems implausible.

        A good actuary who is leading a team and doing work that isn't reviewed by another actuary usually has at least 20 years experience–and that's not arbitrary; the difference between 5 years and 10 is very noticeable, but the difference in efficiency (how much of the time, when he says "we ought to look at x" does it pay off) continues to increase until 20 years at least.

        1. Maybe scientists learn different things than actuaries?

          It used to be uncontroversial that scientists have their best thoughts on their 20s and 30s. Making them wait until they’re 50 before they can pick a project on their own would have been regarded as madness.

    3. I think the problem is that, like it or not, paradigm breaking thinkers are young. You can get good, valuable work from geezers like me, but rare is the geezer who can leave a lifetime of work moving up through a system (and soaking in its mores) behind and really think new thoughts and approach problems in a creative way.

      1. Yes. From these graphs, if you graduate in your late 20's with a Ph.D., you'll be working in other people's labs on *their* projects for 12 years – if you are fortunate. And it'll be a series of one- or two-year temp 'post doc' positions; I imagine that a decade of that will crush any independent spirits.

      2. I'm not sure I agree with your geezer point. Younger scientists do have an edge, but it might be social rather than intellectual. Older scientists are tempted by geheimrat status: administration or adulation both tend to dull the wit. The grey-haired person in the torn blue jeans who leaves a field of world expertise because s/he is bored with it is, I believe, just as likely to break paradigms as any kid in their 20's.

        At least that was my experience back in the 1980's. The better older scientists were the least likely to wear ties.

  4. It would be interesting as well to see the change in age distribution of peer reviewers over time.

  5. I wonder how much of this is tied, at least in academia, to the end of mandatory retirement, the reduction in hiring full-time, tenure-track faculty, and the general aging of the professoriate.

  6. Thanks for posting this. This is an amazing change. It looks like a new Ph.D. (age ~late 20's?) will have to wait *at least* a decade in general, to get a grant. I'm guessing that by then any originality will have been worked out of them by over a decade of temporary post-doc positions in others' labs.

  7. Well, apparently I cannot post as Dennis any longer. It's me anyway.

    To Mike Maltz, the peer reviewer distribution would look about like the R01 distribution, perhaps slightly shifted to the right. To Donald Coffin, the end of mandatory retirement is (in part) causal for the aging of the professoriate and the reduction in tenure-track hiring. It's not the whole story: University administrators have been scavenging tenure-track lines and replacing them with adjuncts for at least two decades now. That has also led to the reduction in new hires.

    So, the end of mandatory retirement is partly causal for the rightward shift in the age distribution, but it's not the whole story either. Another part of the cause is the huge ramp up in the number of scientists chasing NIH awards. It used to be the semi-private domain of med school scientists. Now, Universities with no medical affiliation at all (but big chem/biochem/molbio faculties) are chasing that money too. The overall rate of awards has plummeted as a result. Universities are having to set aside indirect funds not just as startup money for new faculty but also as retool money for mid-career faculty who have hit too many dry NIH/NSF holes with their grant applications.

    Back in the day when the award rates were 30-40%, it was fairly easy to say, "Oh, well. Bad luck, we'll rework this one and resubmit it," and believe that the thing would fly next time. When it's under 5%, and the project officer tells you, "You have a well-crafted proposal with a lot of promise. We could only fund one, though," well — that's a lot harder to accept and say, let's revise it and try again. When your primary mission is teaching (as mine is), it's easier just to go back to the classroom. It might not be as good for my career, but it's more satisfying.

    1. Here’s my take. As an undergrad in the early 70s at a solid but not spectacular state school, I had professors teaching my intro chem and physics courses. Teaching counted something for them to (continue to) advance. Some of those folks got tenured w/o getting a big research grant. But that NEVER happens now.

      These days at large research universities, you get hired with start up money and don’t have to teach for a while so you can get a running start on the indirect-costs-goldmine for your institution.

      No grant funds by tenure time? Bye. The message is real clear: teaching counts for little. It’s all about revenue generation for the dept and college and university.

      The attitude that nothing counts except research drives a lot today. That administrators think this is understandable; that faculty do is a poke in the eye for their majors.

      I finally gave up and joined the private sector.

      1. "No grant funds by tenure time? Bye."
        I actually wouldn't object to that standard for a research university. Unfortunately, it's more like "No second R01 by tenure time? If you leave quietly we won't sue you. Also, if you hadn't spent so much time on outstanding teaching and exemplary service, perhaps you'd have that second R01 by now. So don't let the door hit you on the a** on your way out, OK."

  8. Why is this happening? As far as I can tell, NIH funding showed a sharp rise until about 2002 and then a bell returning to 2002 levels now. But this suggests that some small subset of researchers (trending older so probably the same ones) are taking more and more if the money. Is this some sort of academic capture issue? re funding in real dollars

    1. Big teams using expensive equipment like gene sequencers could also do it. Also a greater insistence on sound experimental design, usually involving larger samples. Just throwing in hypotheses.

    2. I think it is something like 85% of NIH funding that goes to 100 institutions. This does not really explain the age thing, but it is a fact that a small subset of researchers are taking more and more of the money. I think the age thing is largely explained by the increased length of the training period (6+ yrs for PhD average, 5-6 years postdoc), which results in higher age at first R01 application. Add that to a few failed attempts and the average investigator is late 30's early 40's before getting an R01. Combine that with the trend in universities of reducing tenure track lines, in favor of adjuncts being hired to teach. Since they are not firing the old guys and fails, this means more people not getting tenure. So again you are eliminating some 40-50 year olds from the funding pool, and replacing them with many fewer 35-40 year olds.

      It's a great system.

  9. And I suppose all the towers of jello at NIH are lying on the fainting couches, clutching their pearls and wondering how this could have happened. How, oh how!

    Is it just a coincidence that during this same time frame, the number of trainees dramatically increased, the length of the training period drastically increased, the number of visa holders in the trainee population drastically increased, and the percentage of trainees supported by research funds (R grants) rather than training grants or fellowships (T or F funds) drastically increased, and NIH either actively promoted those changes or stood by silently as they happened?

    So, it takes longer to finish a PhD and longer to be competitive on the job market, so investigators are older when they apply for the first R01, and older still when they get that first R01. It's harder to get tenure, R01 or not, so again the pool skews older as younger people give up and drop out.

    Any time anyone mentions this to anyone at NIH, they say "Nothing we can do about it". Really? You don't control something that other people want? You don't have leverage? You just give away free money with no accountability? Really?

  10. This movie was copied without attribution from:

    which is Supplemental material to my paper:

    Gordon, R. (2012). The vanishing physician scientist: a critical review and analysis. Accountability in Research 19(2), 89-113.

    QuickTime movie built from slides (Schafer, 2009b) showing “Age Distributions
    of NIH Research Project Grant Investigators and Medical School Faculty,
    1980–2006,” with kind permission of Dr. Andrew I. Schafer."

    Schafer, A.I. (2009b). How to Succeed as a Clinician Scientist in the Year 2009 and
    Beyond. [PowerPoint presentation]. Available at
    How-to-Succeed-as-a-Clinician-AFMR-Career-Development-Workshop.pdf. Last
    accessed February 13, 2012.

    -Dr. Richard Gordon

    1. Thanks for the links, but they're behind paywalls. I'm interested in the article, and particularly interested in your interpretation of the data. But I'm not $39 interested, nor does my university library subscribe to that Taylor and Francis Journal, and I'm not interested enough to be willing to take up a "fair use" ILL request for the article.

      Your slideshow led to an interesting classroom discussion today — so thanks, and I'll be sure to give my students the appropriate citation on Wednesday.

    2. A journal called "Accountability in Research and it's behind a paywall? Physician, heal thyself. IMHO, one of the principles of accountability in research is to make results available to the public.

      Thanks for the work, and thanks to Keith for letting us see it.

  11. Hey, in the current transitional phase of science publishing, the rules are: author pays or you pay, either directly or via grants. I assure you no one gave me a grant to criticize the grant system, and I can't afford the $500 to $3000 journals charge authors for "open access". But the rules are I can send copies in response to personal requests. Your move. Thanks.
    Yours, -Dick Gordon

Comments are closed.