NIGMS blog post on "shared responsibility"

Jan 05 2015 Published by under NIH, NIH Careerism, NIH funding

This is a fascinating read.

Jon Lorsch, current NIGMS Director, spreads around so much total nonsense that I just can't even deal.

And journals, professional societies and private funding organizations should examine the roles they can play in helping to rewire the unproductive incentive systems that encourage researchers to focus on getting more funding than they actually need.

Riiiight. We PIs out here in extramural land are focused on getting more grant money than we feel that we need. Because what? We enjoy grant writing? Is this guy nuts? When I feel like I have enough grant support to keep what is really a very modest operation afloat I quit writing grants! The problem is that Director Lorsch is really, really out of touch with what a PI in today's climate actually needs. "Unproductive incentive systems"? Dude, when NIGMS stops giving grants to anyone who publishes in Science, Nature or Cell, and starts beancounting Supplementary Data for reduced publication output, and punishes PIs for failing to publish data that is "scooped" or "not hot enough" etc, then maybe I will take you seriously. Jeepers. LOOK IN THE MIRROR, NIH!!!!!!!

But to achieve this increase, we must all be willing to share the responsibility and focus on efficiency as much as we have always focused on efficacy. In the current zero-sum funding environment, the tradeoffs are stark: If one investigator gets a third R01, it means that another productive scientist loses his only grant or a promising new investigator can’t get her lab off the ground. Which outcome should we choose?

better to have everyone funded at $50K and sitting around doing nothing, right?

Although certain kinds of research projects—particularly those with an applied outcome, such as clinical trials—can require large teams, a 2010 analysis by NIGMS and a number of subsequent studies of other funding systems (Fortin and Currie, 2013; Gallo et al., 2014) have shown that, on average, large budgets do not give us the best returns on our investments in basic science.

The "2010 analysis" has been discussed here, I recall. It's flawed. It fails to recognize the cost of a Glamour Pub- love or hate, we have to admit that it takes a rich lab to play in that arena. One pub to the accountants has like 6-10 pubs worth of time/effort and probably data (buried in the Supplemental Materials). It fails to recognize there are going to be some scientific advances that simply cannot be accomplished for less. It fails to recognize the "efficiencies" and lack thereof associated with continued funding versus the scary roller coaster of a funding gap.

and Lorsch does a little neat ju-jitsu with this post. Berg's analysis concluded that $700K was the peak of productivity. That is three concurrent full modular R01s. Even with a traditional budget award the PI has to have two of them awarded at $350K per year (and we know it really means more than that because of cuts) to hit this level. So the finger pointing at the investigator who "gets a third R01" doesn't even square with his own citation on "efficiency", now does it?

Furthermore, the larger a lab gets, the more time the principal investigator must devote to writing grants and performing administrative tasks, further reducing the time available for actually doing science.

Good GRAVY man! Do you have any idea what the hell time it is on PI street? A one-grant lab PI is constantly on the edge of disaster and the PI is always writing furiously to increase that funding level to where she can finally breathe for six months. Protocols and registrations most often are one per lab, so that "administrative tasks" claim is also nonsense. It is the smaller PI who spends more effort per-grant on keeping the approvals in place.

These and other lines of evidence indicate that funding smaller, more efficient research groups will increase the net impact of fundamental biomedical research: valuable scientific output per taxpayer dollar invested.

[citation needed]

It may sound truthy but it is not at all obvious that this is the case. "More efficient" is tautological here but the smaller=efficient is not proven. At all. Especially when you are talking about the longer term- 30 years of a career. There is also the strong whiff of Magic Unicorn Leprechaun money about this comment.

My main motivation for writing this post is to ask the biomedical research community to think carefully about these issues.

You know what I would really like to ask? For the NIH to actually think carefully about these issues. Starting with Director Lorsch.

But reshaping the system will require everyone involved to share the responsibility.

Somehow I doubt he means this. Is there any evidence that NIGMS actually denies the super awesome PIs their extra R01s? Is there any evidence they do anything more than handwring about HHMI investigators with NIGMS grants? Is there any evidence they mean to face down the powerful first and make them equal to the rest of the drones before they take it out of the hide of the less-powerful? HA!

h/t: PhysioProf

24 responses so far

  • lurker says:

    For fukkes sake, DM, get back on your rocker! Lorsch WAS a real PI with NIGMS funding only just 2-3 years ago. I'm sure he knows exactly the gauntlet we all are going through and how much $ a lean yet mean lab needs to not starve to death, and he knows when extra $$$ breeds bloating and inefficiency in the BSD lab. Need 2, no 3 postdocs working on the same damn project, pitted against each other so the best PD wins? Yes, a single modular is no longer keeping up with inflation, and yes, more grant $ gives you more bollocks to take risks that lead to breakthroughs. But you are completely missing the point of Lorsch's new year plea.

    The first real point is that the complete NIH funding system is so way out of balance, beyond the booms and busts cycle you write about. It's becoming the gilded age juxtaposed to the okkie dust bowl. It's that we write grant app after app on thought experiments on paper that either are already done or will be done by the time the A0 masquerading as an A6 gets funded.

    The second real point you missed is that Lorsch is in somewhat Sisyphusian fashion trying to drum support for the MIRA program, which is right now only deck chair shuffling so as to get away from Project based grants towards Investigator-program grants. And he wants to extend a bone to ESI's to be eligible for MIRA. BUT, the catch to MIRA right now is that only established PI's eligible to enter MIRA must already have TWO concurrent R01's, and they must give up the pony from TWO grants and "self regulate" with a single MIRA. Only then will there be extra cheddar to feed the few ESI's who can compete for a MIRA as well.

    But which BSDs with TWO R01's (or a giant Nonmodular) are going to buy into MIRA? Will a BSD like DM read Lorsch's new year plea feel sympathetic or completely miss the point and rant on their blog that Lorsch is out of his fukken mind? Oh wait....

    MIRA is a great idea doomed to fail if it is just deck chair reshuffling, but I applaud Lorsch for his bollocks to at least try something, even if he's like Leo DiCaprio eventually drowning. But there are alot of BSD sharks circling the cold waters who will sink MIRA if the NIH budget stays flat. This is obvious, even at the NIGMS council meeting video where MIRA was presented, and a BSD at UCSF asks how any lab eligible for MIRA is going to be willing to participate in it if it means going from TWO R01s to 1.5'sh grants:

  • Drugmonkey says:

    oh I'm well aware this is about the clusterbork of MIRA. Which follows their alleged "special scrutiny" lie. My criticisms apply regardless of which porker is being lipsticked at any given moment.

  • Drugmonkey says:

    and BTW my fix to the churning problem is far superior. Just increase the use of R37s, preferably using some fancy algorithm to chose the candidates based on past experiences with the continuously funded. Berg's "mind the gap" work up is the starting point for that.

  • becca says:

    Which institute is better than NIGMS for denying special PIs their extra R01?

  • Philapodia says:

    How big of a slice of the NIGMS pie are these 3+ R01 folks anyway? Probably won't make much of a dent to take a few of these away in the short or long run. Another thing about multiple grants is institutional salary structures for faculty. I know TT faculty who have to bring in 50% of their salaries in high cost areas, making it hard to survive and be productive on just 1 R01. Seems like this is an important aspect not taken into consideration in the post when it comes to needing multiple grants to survive.

  • drugmonkey says:

    If it is NIGMS' intent to not fund soft money PIs then they should just go ahead and make that the policy, instead of this absurdness about "shared responsibility". My state taxes fund profs who use that support to do the NIGM's research business whilst barely trying at all to educate my state's undergrads. Who is sharing what responsibility now?

  • drugmonkey says:

    becca- NIGMS is the place trumpeting the special review policy, that's why they get picked on here. I bet it is bogus. We've never seen really hard data and my recollection is that JB gave us some rough numbers testifying to my belief. I.e. That it was actually quite rare for $$$ PIs to be denied through the special review process.

  • datahound says:

    In my previous experience at NIGMS, the "well-funded investigator policy" was applied thoughtfully and with some teeth. Each case of an investigator with more than $750K in annual total costs from all sources (both within and outside NIGMS) was examined by staff and Council. It is difficult to quantify how much impact it had because the results were not quantized. In some cases, applications with percentiles better than the nominal payline were definitely not funded. I would estimate this occurred 30-40% of the time. In other cases, the policy was used in negotiations that led to funding at much reduced levels (e.g. 50% of requested budget) or commitment not to seek renewal for an ongoing grant. I would estimate that this occurred another 30-40% of the time. In the remaining cases, the application was deemed to be sufficiently novel and important that it was treated as if it were a normal application. Another important point is that this policy was only applied to new (Type 1) applications and not to competing renewal (Type 2) applications.

    Does this policy have an effect? Not as much as it could, but substantially more than the NIH-wide pilot which has a larger threshold and considered only NIH funded. The policy also forces staff and Council to become aware of funding levels. It is unfair to call in bogus, but it is not as strong as it could be. I am working on some analysis that may bear on these issues to some extent.

  • rxnm says:

    "My state taxes fund profs who use that support to do the NIGM's research business whilst barely trying at all to educate my state's undergrads."


  • drugmonkey says:

    Ok, ok. 30% is better than nothing JB. I retract the "bogus" claim and replace it with "disappointing". 🙂

    Was HHMI or other non-NIH funding considered? How about soft/hard money status? Overhead rate? Cost of the research models? Number of fellowship funded postdocs in the lab?

  • dsks says:

    Relating to Philopodia's comment, I thought this kill-the-rich meme was binned by that graph showing that the vast majority of PIs with RO1s had 1-2, and that this made up a huge fraction of the investigator-initiated funding? I think it was a DM post, but I can't find it. i.e. even if you trimmed the fat from the >3 RO1 BSDs, it wouldn't make a whole lot of different to the overall pot available for everyone else. I might be wrong, though.

    Anyway, as much as it would be nice if the gummint could stop effectively cutting the budget every year, the main problem is too many pigs at the trough; a problem better solved in the long term bottom-up by reducing the currently deleterious incentives to constantly recruit new graduate students.

  • Evelyn says:

    After working on the other side of the grant machine for almost a year now, I can say that "spreading the wealth" can be tricky. We give out some medium-sized grants and it is tough to deny someone a well written and interesting idea. The only thing I can come up with is to state in the call that the investigator cannot have an active R01 or equivalent. In our last round, our 3 best grants (2 of them we funded) come from folks who would be considered well funded (although nobody had 3 R01s).

  • Dave says:

    Kill grad students and soft-money faculty now!!!!

  • Joe says:

    "My state taxes fund profs who use that support to do the NIGM's research business whilst barely trying at all to educate my state's undergrads."
    That's because all the incentives are for research, and there is no reward for working on teaching. Well, you might get a nice Teacher of the Year plaque, but you won't get promoted or get your salary increased if you invest your efforts in teaching. Furthermore, large reductions in state funding mean that teaching faculty have been fired and replaced with research faculty who also have to teach.

  • rxnm says:

    DM raises an interesting possibility: kill scientists who teach.

  • drugmonkey says:


    It was Sally Rockey's "mythbusting" post

    (although I may have reposted one of the figures at some point)

  • drugmonkey says:

    That's because all the incentives are for research, and there is no reward for working on teaching.

    Exactly. So any professor at a University with any sort of research mission is almost inevitably giving short shrift to the teaching in favor of research and publishing. If they have hard money salary that is for their teaching responsibilities* and they spend less effort on this than is reasonable for the ratio of hard-money versus grant-funded salary, they are robbing the State taxpayer to contribute to the Federal Government's mission.

    Now, sure, these are the same taxpayers in one viewpoint so it shouldn't matter, right?

    Except that the taxpayers of States which support their University missions are shouldering more of the Federal good/service burden than are taxpayers of States which do not support their University professors' salaries.

    *and yes, I grasp that some University hard money deals explicitly say they are paying professors in part to do their research. I still bet a close time-accounting would not falsify my assertion that many hard money profs are giving shorter shrift to actual student instruction than is justified by their percent hard money salary.

  • datahound says:

    DM: HHMI and other non-NIH funding was/is considered. Soft money-hard money status, F&A rate was not considered, and outside fellowship positions were not considered.

  • Pinko Punko says:

    I have heard that some institutes, that have in the past not been as tough as GMS in scrutinizing total funding, are now doing so. I think the teeth are being sharpened for HHMI types, which I think is sort of reasonable. The reality is they have funding for some of these projects. It is incredibly bleak right now.

  • rxnm says:

    " they spend less effort on this than is reasonable for the ratio of hard-money versus grant-funded salary"

    I think this is wrong, at least for many. I work in a 100% hard money environment where everyone teaches, from the chair down, and everyone is expected to have research grants and trainees. teaching and service expectations are a little higher for tenured than for untenured.

    If I compare the target operating budget of my lab and my salary to one in a research only institute of comparable reputation, I think my salary is pretty much right for the teaching and educational mission-related service I do. The taxpayer is getting my research expertise and trainee supervision dirt cheap.

  • drugmonkey says:

    I think my salary is pretty much right for the teaching and educational mission-related service I do.

    I have little doubt that you are one of the good folks. But I have seen a metric boatload of hardmoney profs in research intensive Universities that phone in the teaching*.

    *of course, admittedly, my frame of reference for proper undergraduate instruction is most definitely a non-research-intensive setting**.

    **yes, including the public-funded variety

  • jmz4 says:

    " I work in a 100% hard money environment where everyone teaches, from the chair down, and everyone is expected to have research grants and trainees."
    -That sounds ideal to me. How common of a position is this?

  • […] has been much discussion (including recently e.g. here and here) about the distribution of the number of grants per PI. Here I examine the distribution of the […]

  • […] have been exceptionally dishonest about interpreting the the efficiency data they produce and slippery as otters about resulting policy on per-PI dollar limitations. Nevertheless, one interpretation of their data is that $750,000 in […]

Leave a Reply