Top down or bottom up? NIH RFAs are a two-way discussion between Program and Investigators

One of the erroneous claims made by Steven McKnight in his latest screed at the ASBMB President's space has to do with the generation of NIH funding priorities. Time will tell whether this is supposed to be a pivot away from his inflammatory comments about the "riff raff" that populate the current peer review study sections or whether this is an expansion of his "it's all rubbish" theme. Here he sets up a top-down / bottom-up scenario that is not entirely consistent with reality.

When science funding used to be driven in a bottom-up direction, one had tremendous confidence that a superior grant application would be funded. Regrettably, this is no longer the case. We instead find ourselves perversely led by our noses via top-down research directives coming from the NIH in the form of requests for proposals and all kinds of other programs that instruct us what to work on instead of asking us what is best.

I find it hard to believe that someone who has been involved with the NIH system as long as McKnight is so clueless about the generation of funding priorities within the NIH.

Or, I suppose, it is not impossible that my understanding is wrong and jumps to conclusions that are unwarranted.


Having watched the RFAs that get issued over the years in areas that are close to my own interests, having read the wording very carefully, thought hard about who does the most closely-related work and seeing afterwards who is awarded funding... it is my belief that in many, many cases there is a dialog between researchers and Program that goes into the issuance of a specific funding announcement.

Since I have been involved directly in beating a funding priority drum (actually several instruments have been played) with the Program staff of a particular IC in the past few years and they finally issued a specific Funding Opportunity Announcements (FOA) which has text that looks suspiciously similar to stuff that I have written, well, I am even further confident of my opinion.

The issuance of many NIH RFAs, PAs and likely RFPs is not merely "top-down". It is not only a bunch of faceless POs sitting in their offices in Bethesda making up funding priorities out of whole cloth.

They are generating these ideas in a dialog with extramural scientists.

That "dialog" has many facets to it. It consists of the published papers and review articles, conference presentations, grant applications submitted (including the ones that don't get funded), progress reports submitted, conversations on the phone or in the halls at scientific meetings. These are all channels by which we, the extramural scientists, are convincing the Program staff of what we think is most important in our respective scientific domains. If our arguments are good enough, or we are joined by enough of our peers and the Program Staff agree there is a need to stimulate applications (PAs) or secure a dedicated pool of funding (RFAs, PASs) then they issue one of their FOA.

Undoubtedly there are other inputs that stimulate FOAs from the NIH ICs. Congressional interest expressed in public or behind the scenes. Agenda from various players within the NIH ICs. Interest groups. Companies. Etc.

No doubt. And some of this may result in FOAs that are really much more consistent with McKnight's charge of "...programs that instruct us what to work".

But to suggest that all of the NIH FOAs are only "top-down" without recognizing the two-way dialog with extramural scientists is flat out wrong.

15 responses so far

  • MoBio says:

    Yes agree and "they" also read unfunded in point:

    A few years ago I sent in a TRO1 which was triaged...along with probably >70% of the applications.

    Fast-forward 3 years and an RFA arrives which appeared to have been paraphrased from my Intro. I applied for the RFA and received the award.

    My sense from this is that the idea was probably a bit too forward looking at the time it was initially submitted (though we kept working away) and later when 'the time was ripe' I was able to write something believable.

  • DrugMonkey says:

    Or Program decided that the ignorant riff raff on study sections need to be bypassed so they set up something with an SEP and set aside funds to guarantee at least one grant gets funded.

    McKnight seems to be missing this little alignment with his other argument.

  • girlparts says:

    I'd like to know more about this. About six months ago, I called my program officer to ask for help deciphering some reviewer comments. He basically told me to forget it - reviewers weren't interested in my grant topic, and by the way, I should really work on bunny ears, 'cause it is the next big thing. Lo and behold, the PO has convened a workshop for his favorite peeps on "The role of bunny ears in bunny hopping". The aim of the workshop is partly to inform program staff about future directions. So is this top-down, or bottom-up, or both?

  • neuropop says:

    DM: your characterization is accurate enough. In this case, you or investigators like you, benefit from particular RFAs. If indeed there is a critical mass of PIs who are asking related questions that highlight particular needs bubbling up, then the logical outcome is a targeted opportunity (even the SEP/set aside funds). One might even say BRAIN is something like that. But the design of the particular RFA ends up channeling money and ideas along very particular lines that a select few decide and benefit from. That is something to oppose. Better let innovations come from the "riff-raff". Someone brought up the Manhattan project in the previous thread as the top-down initiative that led to something good. I'd have to disagree. Quantum mechanics on the other hand, which eventually led to the bomb, was a classic case of the "riff-raff" coming up with bold ideas. I know, I know, the players weren't exactly riff-raff, but humor me...

  • drugmonkey says:

    The aim of the workshop is partly to inform program staff about future directions. So is this top-down, or bottom-up, or both?


  • drugmonkey says:

    But the design of the particular RFA ends up channeling money and ideas along very particular lines that a select few decide and benefit from. That is something to oppose.

    given the relative proportions at work here....why?

  • neuropop says:

    @dm:"given the relative proportions at work here....why?' Because it can become a slippery slope. In moderation, I suppose targeted RFAs are a good thing, since they can identify emerging opportunities. The drawback is that once a mechanism like that is entrenched, it sets a precedent for all sorts of initiatives that have limited utility -- ENCODE comes to mind. I subscribe to the notion that the empowering individual investigators is better. Once in a while, though if Congress sets priorities, then top-down initiatives are inevitable, but the less they are employed, the wider the benefits.

  • Ola says:

    What's funny is that even with RFAs, the funding situation is subject to the whims of the reviewers on the SEPs, and these are in many ways no different that those found at a regular study section. Why, I was once on an SEP in which "the person for whom the RFA was written" (i.e. just about the entire RFA was pulled from their papers) applied with 2 proposals. They both got terrible scores. You can be knee-deep in program pork and pigs-hit at the NIH, but it doesn't guarantee you'll bring home the bacon.

  • DrugMonkey says:

    Of course. I've also known of cases where someone who was clearly not in the original discussion space won the award. It isn't a fake review.

    This knowledge is what keeps me applying to RFAs even if I don't feel like I've been part of the discussion. Among other reasons, you can go back to a regular section under a general FOA and possibly get funded. I've obtained two awards that way over the years.

  • datahound says:

    Don't forget that IC Advisory Councils also have a role. The concepts for RFAs and PAs need to be "cleared" before the FOAs can be released. See . I am sure, however, that different Advisory Councils show different levels of engagement as well as different philosophical views of RFAs and PAs versus purely investigator-initiated research.

  • Jonathan says:

    Yup, to echo Datahound, you can't discount the role of councils. Often, they'll even charge ICs with areas to develop RFAs for.

  • Davis Sharp says:

    There is a time lapse from IC Advisory Council clearance of an RFA to the RFA submission deadline. So, if you learn, through an IC's website, that their Advisory Council has "green lighted" an RFA in a particular area of your interest, try to get a grant together for review by a regular study section in the interim. Even if you're triaged, it's a "free" set of critiques to improve your grant for the upcoming RFA deadline.

    This, of course, is more likely to work for R01s and R21s than it is for a P01, etc. that involves a lot of coordination with other groups.

  • DoctorD says:

    One more place to look for early information on Institute priorities is the House and Senate comments in the appropriation bills. Advocates frequently use this language to get specific topics prioritized.

  • […] Clock May End Time As We Know It Top down or bottom up? NIH RFAs are a two-way discussion between Program and Investigators US: Obama asks for $6.18 billion to fight Ebola Postdocs speak up Tapatio is the Best Hot Sauce. […]

  • drugmonkey says:

    I guess I never really outlined the consideration McKnight should have. Perhaps if NIH priorities are always running away from him then he is doing low priority science? Just following the thread here....

Leave a Reply