222 Grants to 22 PIs

BB brings to our attention the “item in the latest Nature re 222 NIH grants going to 22 investigators.”

As Nature notes, the first two PIs with the most awards are conference organizers (including Keystone Symposia) juggling lots (n=32 to Keystone) of little R13 grants. Busy but not necessarily abusive to the system.

The big question is whether big names go on applications to secure the NoA, with the actual work completed by others in the lab. Consider, for example, Nobel laureate and former NIH Director (& budget doubler) Harold Varmus holding on to 8 awards worth $13.1M. John Reed, as director of a 35-person lab and PI on 11 NIH awards (~$10.9M), feels he deserves all his awards. The question is would those 11 awards still go to the Burnham Institute if the person who did most of the work were the PI on the application. If the science is sound, the answer should be yes. So why not let the science and the actual investigator running the show stand on their own merits and apply directly?

And then there is Sten Vermund at Vanderbilt (11 grants, $24.1M), who bluntly “acknowledges that a former stint at the NIH overseeing a $50-million grant portfolio in AIDS vaccine trials taught him a lot about how successful grant applications are packaged and marketed. “I don’t want to make myself sound like a grant-writing technician, but let’s be honest: that is a nontrivial part of success in biomedical research.””

The NIMH currently does not want its funded PIs coming back to ask for a 5th award without prior approval, and I feel the 20% PI effort requirement is a good step toward equitably allocating limited NIH resources, certainly as applied to research project grants (including projects that are components of P mechanism awards, U54s, and the like). As I’ve noted previously, Brian Martinson et al. suggest this would be a welcome shift from the perspective of procedural and distributive justice.

The US DHHS Office of Inspector General is very interested in effort reporting, so it could be they will offer indirect help to the NIH’s movement toward limiting awards per applicant PI. Harold, John, and Sten clearly have a few obligations to their institutions outside the lab for which they are paid. Aside from fairness in allocating resources, managing – and maintaining funding for – so many projects and people can take its toll on even the most vigorous researcher, as Eric Poehlman can probably attest (falsified/fabricated data on 17 grant applications to the NIH).


  1. BB said

    Thank you for posting. I’m eager to read other opinions from other folks on this topic. A few thoughts: 1st, if a PI lists him/herself for at least 20% salary (and effort) on a grant, the spector of >100% salary is there. How is that possible or justifiable? My university does not allow over 100% salary support, and the VA, the second institution I work for, allows PIs one and only one award (at mandatory 5/8 salary support minimum – you can easily see why the limitation is there). I’m floored that other places allow >100% support; I’m writing a DoD grant and worried because I’ll have to ask for less than 20% support- and I want to be seen as taking the grant very seriously! 2nd, if one is at 100% salary support, one has no incentive to carry out the work of a university- teach, patient duties, committees, etc. So the university workload is no longer shared among colleagues, as it should be. Me, I’m on 2 big committees though I could teach more but that’s political at my school. 3rd, it’s even more disincentive for young people thinking about scientific careers. The view that the plum prize of an RO1 -or 5 or 6- goes to the established labs with RO1 grants has never been truer. And obtaining an RO1 has never been harder. Last, Sten Vermund should package and sell his technique. That would be a way to “spread the wealth.”

  2. bikemonkey said

    direct cost dollars matter as much as the number of grants. if someone trades in two $250K modulars for a single $499K traditional budget R01, what have we gained? this is what ticks me off about many NIH actions. every “fix” leaves a barn door’s worth of loopholes.

    BM, this doesn’t make a lot of sense. Trade in 2 modular R01s for a single larger budget R01? How, unless the two projects have scientifically merged into one larger effort that justifies the single larger budget? Most folks I know would rather keep 2 smaller R01s on the books, especially in case only one of them continues to produce promising data to support renewals. If you are suggesting, for example, that folks with 8 awards combine projects and budgets to wind up with 4 awards (20% effort on each) with the same level of funding, I don’t think this is a viable option in most cases. More likely they would transfer PIship to someone else in the lab prior to the Type 2 submission and hope that person can keep the project going. In this case, more – hopefully younger – investigators are funded … and they’ll have awards they can take with them should they get itchy feet or better offers from Universities trying to buy PIs with NIH awards.

    On the other hand, what BB notes rings quite true. What BB probably doesn’t realize is that these big names do not hesitate to put themselves down for only 5-10% effort on all these awards (obviously less in the case of those with more than 10 awards). I don’t care how many hundreds of hours they are willing to put in personally, the committed effort on paper needs to be comparable for all applicants who take responsibility for serving as principal investigator on the project. I look at this in the same way the uniform page margins, font size, character spacing, etc. level the playing field in terms of the submitted narrative. – writedit

  3. BB said

    So if you are “Big Name” you can get an RO1 with only 5-10% effort as PI on that RO1? How would that ensure success of the project in the mind of the study section? I do not get it at all. I don’t get a lot of things about grants, I admit.

  4. Neuro-conservative said

    Cross-posted from Drugmonkey:
    This conversation is proceeding from a false premise, instigated by poor reporting in the original Nature article and exacerbated over the course of this thread. None of these guys has 10 R01’s. They don’t even have 10 independent awards.

    The search performed by Nature, presumably run through CRISP, brings up multiple records for the same Center grant — and all of these dudes are Center heads. For example, a CRISP search on Harold Varmus for the year 2007 brings up 8 hits, as designated in the Nature table. But 6 of them are for separate components of P30CA008748, imaginatively titled “Cancer Center Support Grant.” He also has one R01 and one U01.

    Several other names on the list are PI’s of CTSAs, which seem to come up as 3 separate CRISP entries. Throw in a T32 and a regular Center (P30) grant, and your total can easily seem astronomic with only one or two R01s.

    In pointing this out, I’m neither defending nor attacking the “Big Science” Center system, just clarifying the underlying facts. I have seen both major advantages and major disadvantages to the way Centers are currently structured. Net-net, though, I think the Center construct is indispensible, and will necessitate some species of big kahuna to run them.

    You are right, Neuro-conservative. I would point out, though, that projects on a Center award (& not all Center awards include separate CRISP listings for their affiliated projects) should be treated as the equivalent of an R01 in scope; during the submission of a Center/PPG grant application, the individual projects can be simultaneously submitted as separate R01s and funded individually if the Center/PPG isn’t awarded … the one exception to the no duplicate submission/review rule. Many institutions view the Center/PPG mechanisms as ways to get their promising junior investigators on the road to independence as the named project PIs (versus shadow PIs behind the big kahuna’s name).

    However, I see Harold’s cancer core center grant has 5 (five!) supplements appended. Not an effort issue probably, though I can’t imagine it hurt to have the Varmus name on all those supplemental requests. Ditto for Sten Vermund and his 6 (six!) supplements to a D43 program (international investigators come to US awardee, train, & go back home with some start-up $). Athlete-investigator John Reed, on the other hand, does need some scrutiny with how he can genuinely fulfill PI responsibility (while serving as President & CEO for his Institute) for 1 program project grant; 4 PPG projects (on 4 different P01s); a U54 molecular screening library (plus core & 2 supplements); 4 R01s (each from a different IC); 2 R03s; and a U19 (another multiproject research program mechanism, in this case a cooperative agreement). [So actually, he does essentially have 10+ R01s.] And I know he is not alone in wearing multiple institutional leadership hats while serving as PI on 4 or more R01-level or higher awards. These people all work insane hours – but the question remains is this the best use of limited (& dwindling) NIH dollars, especially in view of the “broken pipeline” … which is the context in which the issue was raised by the peer review advisory committee et al. – writedit

  5. bikemonkey said

    writedit, I think you are overlooking the medium term outlook and strategy. My point was that instead of submitting $250K modular grants, the bigger PIs would simply shift to submitting $499K traditional grants. The bigger PI might fight harder to get the next renewal accepted at a significant bump from the prior interval (actually, I’m not sure how seriously the ICs take the rules in this regard at present time-any idea?).

    You also overlook the degree to which there is overlap across projects once a lab gets up into the 3+ R01 level of funding. Grants are not contracts and there is very little obligation to stick close to the letter of the proposal. the more basic the science, the less possible it is to analyze contributions to each project specifically.

    So if you are “Big Name” you can get an RO1 with only 5-10% effort as PI on that RO1? How would that ensure success of the project in the mind of the study section? I do not get it at all. I don’t get a lot of things about grants, I admit.

    Oh. Boy. To be as neutral as possible about this, the rationale is based on “track record”. As in “The PI produced all this great stuff in the past 5 year interval with the same effort so s/he can do the same with the next interval”.

    After that things get super fun. One can delve into other commitments, the way junior investigators get hammered over percent-effort (now “calendar months allocated”), how to assess productivity in the context of multiple concurrent awards…go bait the DM over at DrugMonkey and I’m sure you can entice a rant or three.

    BB, please do!

    BM, certainly a PI can try to double his/her award size – so long as the science supports it. Bloating the budget won’t (shouldn’t) affect the priority score, but it won’t automatically sail through intact and certainly can be trimmed, with rationale documented in the summary statement. An exception would be moving from bench/preclinical to clinical testing/intervention, in which case the budget would indeed be expected (& allowed) to increase exponentially. But, with program officers fielding calls from distraught long-term PIs pleading for help so they don’t have to lay off their best people, I don’t think the big kahunas can expect a handout just because they asked. Given the negative mental health effects of being a program officer these days, I’d love to know if they view the 20% per PI requirement as a godsend of sorts.

    On the 3+ R01s to a single lab to fund overlapping work … not sure what I’m overlooking. Sure, grants are grants over the award period, with the moment of truth at renewal time. Drawing separate funding for similar projects from different ICs – e.g., NIDA and NIMH, or NIGMS and NCI – makes maintaining multiple awards to the same lab/PI easier. Lesson for all … find program officers in at least 3 different ICs willing to advocate for your applications. – writedit

  6. bikemonkey said

    writedit we continue to speak at cross purposes. The multiple, smaller grant strategy only has advantages until Program says “Sorry, you only get two grants”. The intent of such a plan might be to reduce the degree to which BigCheezery can accumulate vast tracts of research funding. I mean, what relevance is the number of awards if not the dollars? Who would care and what would it affect otherwise?

    My skepticism that this is window dressing arises from the fact that they chose to focus on the number of awards instead of the research dollars. This is a giant loophole if the goal is to limit the $$s.

    No, I think we just disagree. I believe the goal should be indirectly to limit the number of awards (dollar amount appropriate for the work proposed) – and, more importantly, to ensure % effort matches the award to which it is yoked. The 20% figure is the minimum per award. If BigCheezery were PI on a P or U54 grant plus a PI on one of the component projects, that would be 40% effort. Some study sections could look at really massive projects as needing at least 50-60% effort in reality. On the other hand, there are some really huge-budget R01 awards for clinical trials that probably only need 20% effort from the PI since most of the $ is needed to pay clinical coordinators, patient costs, lab tests/Luminex, etc. I don’t think setting a limit on the dollar amount is a fair or workable solution. Ensuring realistic percent effort (starting with a minimum of 20% for the PI) does begin to level the playing field, especially if study sections and/or program can also take into account the PI’s day job (i.e., dedicated researcher vs dean vs CEO-President etc.) when evaluating the effort proposed and looking at the biosketch for current research support (or the full Other Support via JIT in the case of Program). – writedit

  7. DaveK said

    NIGMS has an interesting policy about this:

    “The Council expects the Institute to support new projects in well-funded laboratories (defined for at least 5 years as >750k in annual direct costs
    — dk) only when distinctively promising work will be pursued. The Institute’s default position is to not pay such applications. However, under special circumstances and with strong justification, staff may recommend overriding the default position. Council must approve the recommendation.”

    Further, “the Council expects the Institute to implement, where appropriate, reasoned budget reductions greater than those dictated by the cost-management principles for both Type 1 and 2 awards made to well-funded laboratories.”

    The link is http://www.nigms.nih.gov/Research/Application/NAGMSCouncilGuidelines.htm
    Note the typo in the link, repeated in the header.

    Interesting. Thanks for pointing this out, Dave. I suspect other ICs have similar policies tucked away – or will soon as they realize they need to spread the wealth a little better. However, not a typo: National Advisory General Medical Sciences (NAGMS) Council. A bit awkward, but maybe the point was just to replace the “I” with an “A”. – writedit

  8. writedit said

    From Nature last week, an interesting expansion of the % effort discussion to holding Universities accountable for hires (i.e., don’t rely on soft money) and to stop using NIH funding to expand their research enterprise. What a novel idea – use the NIH funding exclusively to support/conduct the research for which the award was granted!

    “To address the issue of a stagnant NIH budget, the draft suggests limiting the number of grants an investigator can hold, or stipulating a 20% time commitment per grant. These changes could free up funds for more (and perhaps younger) investigators. But the NIH is also concerned about how much of the money awarded to an investigator is spent by institutions-to cover salary, for example-and about the increasing reliance on NIH funds to facilitate university expansion projects. Placing limits on grants for salary recovery might change the current culture of university reliance on NIH money as a revenue stream and force the universities to take greater responsibility for their hires.”

  9. writedit said

    Both Harold Varmus and Sten Vermund take Nature to task for not explaining the types of grants involved, and as I noted above, both have legitimate reasons to complain about being lumped in the same category as a John Reed.

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: