Defunding the NIH

A article in the Pacific Standard magazine by Michael White provides an update on my prior post on The NIH Un-Doubling. The primary point in that post was a graph published in 2007 in

Heinig SJ, Krakower JY, Dickler HB, Korn D. Sustaining the engine of U.S. biomedical discovery. N Engl J Med. 2007 Sep 6;357(10):1042-7. [Publisher Link]

which presented the NIH budget allocations in dollar amounts adjusted for inflation* (expressed in 1998 dollars). The "undoubling" part reflected the 2007 allocation and 2008 Bush administration request in comparison with a trendline established from the early 1970s until the beginning of the doubling. It's worth revisiting the graph from that article

Figure 1. NIH Appropriations (Adjusted for Inflation in Biomedical Research) from 1965 through 2007, the President's Request for 2008, and Projected Historical Trends through 2010.
All values have been adjusted according to the Biomedical Research and Development Price Index on the basis of a standard set of relevant goods and services (with 1998 as the base year). The trend line indicates average real annual growth between fiscal years 1971 and 1998 (3.34%), with projected growth (dashed line) at the same rate. The red square indicates the president's proposed NIH budget for fiscal year 2008, also adjusted for inflation in biomedical research.

because the updated one, below, only starts in 1990.

NIHBudget-MAW-edit-497x400This new article How We're Unintentionally Defunding the NIH provides the update, now represented in 2011 dollars. I'm not immediately seeing whether Michael White made this graph himself or sourced it from somewhere else but he does cite a Congressional Research Services report by John F. Sargent Jr which is worth a read.

This is fascinating. We've discussed historical funding trends and success rates under NIH extramural grant awards in the past. One post I wrote is highly pertinent:

The red trace depicts success rates from 1962 to 2008 for R01 equivalents (R01, R23, R29, R37). Note that they are not broken down by experienced/new investigators status, nor are new applications distinguished from competing continuation applications*. The blue line shows total number of applications reviewed...which may or may not be of interest to you. [update 7/12/12: I forgot to mention that the data in the 60s are listed as "estimated" success rates.]

The bottom line here is that looking at the actual numbers can be handy when playing the latest round of "We had it tougher than you did" at the w(h)ine and cheese hour after departmental seminar...Things are worse than they've ever been and these dismal patterns have bee sustained for much longer. ... Anyone who tries to tell you they had it as hard or harder at any time in the past versus now is high as a kite. Period.

One key takeaway from this new graph is a consideration for those who insist that the NIH doubling interval was a poisoned gift. There are those that claim that our current woes are because research Universities and Medical Schools built up tremendous amounts of new infrastructure and personnel during the doubling, with the expectation that that rate of NIH budget escalation would continue. The thinking is that we experienced a bubble and the only reason we have problems now (during this extended interval of budget flatlining and therefore slipping purchasing power**) with dismal success rates. Too many mouths at the trough, is the way I put the situation, even if I don't specifically blame the doubling interval for this.

This new graph makes it very clear that we have not just returned to the 3.3% growth trendline for the NIH budget. We have fallen off that line. Furthermore, the stimulus funding and the modest increases the Obama Administration have bruited as an initial budget offering are insufficient to change this divergence. It is absolutely clear that the NIH purchasing power is shrinking. Shrinking below the trends established from 1971 to 1998.

This is not a contraction relative to the doubling interval anymore! We're way beyond that. We look to be as far below the historical trendline as we were above the line at the peak (end) of the doubling interval. We're something on the order of $8-$10 Billion in the hole, something around 75% of where the historical trendline would have taken us. That seems like a lot of money until you realize

*from here:

**using BRDPI (Biomedical Research and Development Price Index)

49 responses so far

  • Dave says:

    Well that's all rather depressing.....

  • Jim Woodgett says:

    Good information but it's likely an even worse scenario given that how and what things are funded (and included in the budget) has also changed. I don't have any NIH examples, but in Canada, various program "initiatives" such as "Patient-Oriented Research" have been earmarked and indirectly reduce funds from other pots. Moreover, this is a politically safe way of reducing traditional funding as politicians are able to say with hand on heart that total funding is being maintained. Of course, this can work to research benefit in some cases (e.g. DoD breast cancer funding). But the devil in these graphs are the details - and we live off those same details. A dollar that is siphoned off into some obscure pork-bellied program is a dollar lost.

    The other complication of multi-year data is when programs are significantly reformed (akin to devaluation of currency). The Canadian Institutes for Health Research (our bobble-head sized version of NIH) is about to change their programs as well as the entire mechanism of peer review in an effort to improve allocation and fairness. The uncertainty and distrust that accompanies any wholesale change is, of course, having an immediate impact as people submit more applications before the changes take place (next Fall). As a consequence, the current system is imploding (also helping to justify the need for reform). What effect that has on our longitudinal graphs is unclear, but methinks it'll be a good approximation of a cliff.

    The impact of increased funding pressure also leaks over to other agencies. We have to look at the entire system to see what the true impact is. The tight budget at NIH is clearly causing significant stress at research support-oriented charities and other sources.

  • Joe says:

    You can draw a trend line from 2003 through 2013 and beyond and see that in about 21 yrs we will be at the level of funding we were in 1990. The additional data we need now is on when research institutions will cease hiring or start firing scientists (reducing the number of applicants submitting) and when congresspeople will start noticing that people are dying of antibiotic-resistant infections and that the private sector is not taking care of that.

  • GAATTC says:

    I still stick to my adage that they can't fire all of us. At least I still think I do.bToo bad none of this information seems to elicit much change in Uncle Francis et al. Last one left turn out the lights.

  • Ola says:

    I've been pushing the "until we deal with out of control military spending, all else is moot" line for years, but unfortunately the wackaloons have made it such that anyone who says this too loudly is labeled as unpatriotic.

    At the height of the Talibanistan boondoggle, we were burning $12m an hour in the desert. 12 fucking R01s an hour! I can't remember who said "if you're not outraged, you're not paying attention". A very apt phrase for the times. Maybe it was Howard Zinn?

    Anyway, fuck the DOD.

  • Grumble says:

    No, they won't fire us all. They will keep everyone with 3 R01s, all the center grant directors, etc. Those of us who fall short of that standard will eventually be told to go fuck ourselves.

    I've already started seeing it happen at my institution (med school): several underfunded faculty have been forced out, while at the same time the administration is awarding some extravagant retention packages to well-funded faculty who threaten to leave.

    "Imploding" is a word I keep hearing. It might be time to jump ship before there's too much competition for all the "alternative PhD career" jobs.

  • Grumble says:

    At the recent SfN meeting, Navy warships passed by the convention center regularly, cruising under the Coronado Bridge. They have big numbers on the side. A friend who was in the Navy long ago googled one of those numbers, and found out that the ship (I think it was a destroyer) burns $230,000 worth of fuel per week.

    One R01 per week.

    And that's just one ship out of thousands.

  • DJMH says:

    Rachel Maddow's book Drift does an excellent, if equally depressing, job of showing how the military spending got to its current heights. Shorter version: thanks, Reagan.

  • Ass(isstant) Prof says:

    I hear from some of my senior colleagues that "we had it bad, too, and still made it." Of course, knowing that it is false doesn't help anyone not established get funded.

    I'm telling students that going into grad school now might be OK, though. Once the Boomers retire/senesce and the big cull takes 90% of the Xers and 70% of the Millenials, they might be in good shape. Not sure what to tell my tech, who loses his job next week, partly due to lack of new grant funding, partly due to sequester delays.

  • drugmonkey says:

    I hear from some of my senior colleagues that "we had it bad, too, and still made it." Of course, knowing that it is false doesn't help anyone not established get funded.

    This is why I like to show the longitudinal trends reaching back as far as possible. Because at all times, as far as I can tell, the new entrants have it harder than the established investigators. So memory of their youth spurs the above sentiment and a refusal to understand that there is a historical decline in grant success layered on top of this new/old investigator effect.

    So whenever you hear this, make sure to point out the relative success rates now and when they were noobs. Also, point out the lasting duration of the current downturn versus prior blips. And then, as always, point out the declining purchasing power of the average R01. As I pointed out elsewhere in a post, you need something around $350-375K per year right now to equal the $250K full modular when they first established it.

  • Mikka says:

    Regarding this last point, how much does submitting a non-modular budget in the $350K range hurt the perception and score of the grant? If one submits a reasonable budget, well justified, does the reviewer still get the idea that you are asking for too much? The CSR must have data about this.

    NIGMS is cutting all budgets across the board 20%-25%, pushing the modular budget almost into NSF numbers. Since an R01 funded to the tune of $160k today dollars is definitely not self-sustainable, shouldn't we all be submitting non-modular budgets? Of course, even if everybody asks for more the money is not going to magically materialize, but at least any biases against non-modular budgets would go away?

  • Grumble says:

    Mikka, I've reviewed literally dozens of R01s, and I can remember exactly two that exceeded the modular limit. One I loved and it got a great score and was funded. The other was OK, but it seemed to me like they were adding all sorts of extra collaborators onto it. The collaborators were actually justified by the experimental plan, but my feeling was that the experimental plan was not very focused - almost as if they came up with the plan to justify all the collaborators, rather than the other way 'round. That grant didn't do very well. (Also, there was some nitpicking by the reviewers - the grant asked for a few thousand dollars per year for an overseas adviser - not even a collaborator - to come visit them. It was an easy target. So, by spelling out all the budget items, you set yourself up for this kind of criticism unless everything is very well justified.)

    As a reviewer, I'm not automatically opposed to non-modular grants - but I want the budget to be very well justified. It should be glaringly obvious why a larger budget is required. Basically, by including a detailed budget, you are just begging reviewers to review it, even though budget issues are not supposed to be the focus of scientific review. When everyone is struggling for money, a request for a lot of it is going to invite extra scrutiny and skepticism from reviewers.

  • old grantee says:

    Just watching Washington Journal dealing today with NIH activities and the effect of sequestration on biomedical research and researchers. A young researcher asked Dr Collins on the prospects for young investigators having to leave their field and going into related, if not completely different type of jobs. Collins sounded kind of hopeful and sent a call: “Hang in there, things are going to get much better soon”. Is this real or just the result of his optimistic nature?.

    On a different avenue, yesterday there was a Senate hearing on the prospects of cutting unemployment compensation for 1.3 millions Americans out of work and desperately looking for reemployment. It was heartbreaking to hear their testimonies on the effect on unemployment on their lives, sense of purpose and self-esteem. Two of them were scientists having been laid off from university and pharma companies and their struggle to overcome their now long-term unemployed situation and to re-invent themselves to fit their experience into new avenues. Collins talked about NIH biomedical research being a source of economic potential and wondering whether some of these people, well into their late fifties, might find an opportunity to contribute within the NIH mission.

    Just wondering if anyone has any insights.

  • drugmonkey says:

    I think Collins has a professional obligation to be optimistic and upbeat no matter the reality or his private opinion. So I listen for specifics rather than his attitude and affect.

    And being the cynical person I am, things that sound like realistic analysis coupled to moderate changes are more convincing than soaring rhetoric.

  • neuropop says:

    And now this?? Fund the superstars. The rich will get richer indeed!!

    In some ways it makes sense, future funding is based on past productivity. But the Pioneer model is not the right way to go.

  • Grumble says:

    I can't access the Chron article, but "fund people, not projects" is indeed the right approach, IMHO. But if it's done at a small scale, you end up with just the biggest of the BSDs getting funded - which defeats the whole purpose as far as I'm concerned.

    Not sure what Collins has in mind - but from what I glean from the opening paragraph, Collins wants to expand the Pioneer program (or something like it) from the small program it is now to something NIH-wide. Presumably that would make it a bit more egalitarian. That's at least a step in the right direction.

  • […] We are seriously defunding the NIH […]

  • drugmonkey says:

    It *will* be on a small scale. Favorites will be picked for protection and the rest of us left to fight for a smaller pool of funds. Make no mistake about that.

    Buy in from investigators will be 99.87% based on self-assessment of which side of the line one will fall.

  • Joe says:

    @Mikka & Grumble,
    With regard to grants with non-modular budgets. I have one. No one dinged the budget at the 01 or A1 stage (when it was funded). I have a good, but not fundable score on another application with a non-modular budget. $250K just isn't enough money to get the work done, so ask for what you need and provide the justification. At a study section this year, I saw only two applications get recommendations for reducing the budget. Those got other complaints, so they were unlikely to be funded anyway.

  • neuropop says:

    DM: Did you do an analysis of productivity broken down by HHMI BSDs, Pioneer/New Innovators and the well-funded hoi-polloi? I seem to recall that you had done one of this and found that the biggest correlation to productivity (even judged by the dreaded impact factor) was \(\) but I can't find it.

  • Grumble says:

    DM, I appreciate your skepticism, but is it based on any knowledge about what Collins actually has in mind? Or is it just general cynicism?

  • Ola says:

    Re: modular vs. non...... I tried for non-modular and study section said go modular. Was it a factor in the scoring? No idea, but when I went modular on the A1 then it got funded. N=1, but probably more generalizable than one would think.

    As a reviewer, the non-modular awards I have seen fall into 3 categories:
    (i) Clinical studies - seriously, trials are expensive and a non-mod' budget is often necessary.
    (ii) Basic scientists, A0, trying their luck for a bit more cash (275-325 range). They expect they''ll get cut at the funding stage, but almost all get told to go modular on the A1, and they do. I haven't seen a non-modular budget at A1 stage for the past few cycles (and let's not forget that A1s make up the bulk of what actually gets funded these days).
    (iii) BSD basic scientist with a completely crazy budget in the 400s, with the justification including 3% effort for 20 of his best friends, and no-one below faculty level to actually do the science. Thankfully these often get triaged for other reasons (such as the BSD, despite doing good science, having a complete inability to string a fucking sentence together).

    I wouldn't mind the status-quo if they actually funded modular budgets to the full amount, but as we all know that's just not happening nowadays. It's like your employer not giving you a pay raise for 10 years - you can live with it if you know about it in advance, so you can budget accordingly, but if they then randomly dock your pay further, it becomes very hard to plan your life.

  • labrat says:

    My institution does not allow any modular budgets to be submitted any more... We all have to go detailed and ask for more.

  • qaz says:

    Re: non-modular vs. modular. Legally, study section is not supposed to take budget into account when scoring the grant. Scores get announced, voted on, and then the chair asks if there are any budget issues. At which point, the three primary reviewers are supposed to "make suggestions to program about the budget".

    Note: whether the budget actually affects one's perception of the grant is another story. In my experience, the members of the study section I am on tend to be very cognizant of whether a grant is in the right budget ball park or not. Grants way out of line (a $495k grant that should be modular) tend to get noticebly worse scores. Whether that's because the study section members are mad about the budget or whether that's because people who pad budgets tend to write bad grants, I don't know.

    Generally, in my experience something slightly more expensive than modular (less than $325k) doesn't get scored badly but might get a "this should be modular" comment in the budget discussion.

    Budget for what you actually need. Don't pad the budget. We'll notice.

  • drugmonkey says:

    Only thing I've seen draw consistent fire on high $300K+ proposals are the ones larded up with several techs, grads and postdocs. From a budget perspective they look like "fund my big ass lab for another five years" type of thing. The irritation definitely translates to score, of course it does. Reviewer psychology is ignored at your peril, no matter the supposed rules.

    With that said I've written more traditional budget apps in the last couple of years than ever before. The declining purchasing power of the full-mod has become untenable.

  • drugmonkey says:

    Anyone ever gotten nailed by the maximum percent-increase rules of a competing continuation?

  • Mikka says:

    Thanks. I guess next one will be non-modular. Otherwise I'll be paying people to sit on their asses. In any case my institution makes us craft a detailed budget even for modular budgets, for internal purposes. So for me time gain is zero.

  • drugmonkey says:

    What "internal purpose"?

  • qaz says:

    We have the same thing Mikka does. Our internal processing department (the one that actually submits the grant) demands a complete detailed grant budget in order to know which accounting strings to assign to the grant. They then use the detailed budget to determine how much to ask for in the modular budget. A lot of people waste a huge amount of time building detailed funding numbers for modular budgets, including delicately changing the number of dollars assigned to "lab resources" or percentage of time assigned to "technician 1" so that the modular budget comes out the way we want. It's really messy because the program they have automatically calculates the various overheads that get counted in direct and indirect costs. Too little and you can find yourself in the $225 k/yr but small changes can throw you into the non-modular range. It's a real pain.

    In my department, our (wonderful!) support staff has figured out ways to make sure there's a dollar in each account string and stuff like that to make dealing with our internal bureaucracy easier.

  • mikka says:

    I have no idea but it could be something similar to what qaz describes. Why they would want it for all proposals, not just awarded grants, is what has me stumped.

  • Grumble says:

    They want it for all proposals because they are accountable to NIH for how the money will be spent. Let's say you budget for an outside subcontract, three monkeys and $50,000 in travel expenses. The outside contract incurs extra F&A, the monkeys require an approved (or soon-to-be-approved) animal protocol, and the travel expenses need solid justification. If you just submit a modular budget, there's no way for the college to know what you plan to spend the money on, and therefore no way for them to make sure that they (and you) are following the NIH's rules. And in the case of the outside contract, the F&A going to the subcontractor would come out of YOUR direct costs unless the institution specifically asks for those extra F&A expenses (which can exceed the modular limit even in a modular grant). So in the end all that scrutiny can benefit you.

  • drugmonkey says:

    That's not necessary in advance. And it is not really any protection- the PI could still try and spend the money on new rims for the Beemer. Universities have to have scrutiny at the stage of approving purchases. So why bother specify in advance?

  • mikka says:

    Because administration? Hey, those indirects aren't gonna spend themselves!

  • job hunting says:


    Wondering if the available certificates for HAS and MO, advertised at NIH Jobs, are for people already in the Intramural NIH or are available to anyone interested in training and education jobs (see below)

    "Global Recruitment for HSA and Medical Officers; Pathways Recruitment for Program Support; HR Modernization Program"

    I appreciate info from you or anyone participating in your blog.

  • Grumble says:

    I agree that, from the PI's point of view, providing the administration with a detailed budget at grant submission is mostly a complete waste of time. It would make some sense to go through budget review (and other administrative review, such as compliance with animal and human subjects rules) only once the grant gets a score that might be fundable. But what if the PI has made some mistake (e.g., not asking for subcontract F&A) that will cost the college and/or the PI money?

  • qaz says:

    Since most grants (including all NIH and NSF grants) are essentially fungible and budgets can mostly be reworked as needed [you don't even have to work on the project you propose, so how can you they worry in advance about whether you've correctly budgeted the post-doc cost?], it always seemed stupid to me to fill out the detailed budget plan.

    But, as Mikka says, "Because administration?" (That's going to be my new goto line whenever anyone says the university is being stupid.)

  • Joe says:

    The answer to the bother of doing the detailed budget is to have boiler-plate info available. Do the detailed budget once and then you only have to change it if there is a big difference in the way you going to spend the money (grad student instead of senior tech salary, or equipment in first year). I am lucky to have dept staff that do this for me, but if I tell them the personnel I want on an application, they can respond with a detailed budget in an hour. I tell them to shift a $1000 here or there, and we're done. The non-modular budget is more work, since I try to get all the numbers as close as possible for the study section.

  • Dave Bridges says:

    General point on the dotted line. Is there any reason to believe that inflation for scientific research is the same as general inflation in the economy. I would guess it is much higher.

  • drugmonkey says:

    Yes. These analyses use BRDPI to adjust the dollars.

  • old grantee says:

    It seems as if there are good news, though not as good as one would want, for NIH with the announced budget deal. Let's hope for the better.

  • old grantee says:

    Let's hope for the better, if not the best.

  • meshugena313 says:

    Has anyone seen details on the appropriation for the NIH in the proposed budget deal? I can't find anything...

  • Grumble says:

    Supposedly the deal lifts 50% of the sequester cuts for 2013 and 25% of them for 2015. I think it also allows Congress more latitude in moving around the cuts (i.e., cutting more from one agency to increase the budget of another).

    While any reduction to the sequester is welcome, I see little to be really grateful for. "Thanks for hitting me less." And the money is coming from federal workers, who've already had a lot taken from them.

    Fuck you, Congress.

  • […] rapidly un-doubled and fell back on the historical trend line. [see update suggesting we are now defunding the NIH] That analysis depended on the Biomedical Research and Development Price Index or BRDPI. This […]

  • […] See this graph for a recent historical trend on success rates and application submission numbers. With respect to the latter, you can see that the small decrease to 49,581 is not hugely significant. We'll have to wait for a few more years to be convinced of any trend. Success rates are at an all-time low. This is rather unsurprising to anyone of you that has been paying attention to doing at the NIH and is a result of the long trend toward Defunding the NIH. […]

  • […] to applications submitted from FY2007 onward? I mean, we're seeing evidence in this time of pronounced budgetary grimness that the NIH is slipping on its rather overt efforts to keep early stage investigator success rates […]

  • […] that we should also review the subsequent update on the Undoubling graph, the even more sinister Defunding Graph. Via Michael White, presumably via John F Sargent, […]

  • […] billion slated for next year.  This is great news for scientific research.  However, history has many lessons about the potential hazards of a big bump in science funding.  In the coming days, […]

Leave a Reply