Your Grant in Review: The impact of institutional affiliation

A comment on a recent post discussing the impersonality of perceived bias in the NIH grant game asks:

I would be interested to know your (and others) opinion(s) on the weight that is given to the institution, aka, environment. In other words, how important is your ZIP code?

My answer: Crucial, immaterial….meh. More after the jump.

There has, probably since forever, been a phenomenon in the NIH funding race: the rich get richer. Some Universities or research institutions get a lot of NIH money and they tend to do so year after year, decade after decade. For a hint of which institutions are at the top, you may check the ARRA/Stimulus dole out and how that compared to NIH $$ for FY 2008, for example. My posts list allocations by Congressional District; these are bigger than ZIP codes but probably capture the regional flavor of the original question.
The rich get richer so your affiliation must be crucial, right?
Not exactly.
In case you are unaware, the NIH has several strategies to combat what might otherwise be a reinforcing spiral. Unfettered, geographic regions or individual Universities might attract the better scientists, who get the most NIH Grant dollars, conduct the more cutting edge science, get asked to advise the NIH on the best science….and continue to award the grants to people like themselves. This is deemed unacceptable, going by certain efforts of the NIH.
At the primary level of review, the CSR has an explicit mandate for geographic diversity in the reviewer pool. According to the official rules for reviewer selection.

There must be diversity with respect to the geographic distribution

In my experience this is taken quite seriously. I’ve had some conversations with SROs about possible adding Dr. So-and-so to the panel and had it pointed out to me that because of someone else from the same region, they would have to wait to empanel my suggested person. To address the more general point, my experience from looking over the rosters of the panels reviewing my proposals suggests that size, type or NIH funding status of home institution is very diverse.
My view is that this type of representation at the primary point of review is an absolutely critical step in fairness for those at less research focused institutions.
I would also point out that there are specific grant mechanisms reserved for the less well-funded Universities, institutions or regions. The AREA / R15 mechanism is one such: eligibility requires an academic unit (Med schools can be distinct from the traditional part of the University) that does not exceed a certain amount of NIH funding. There are pools of money limited to “Underserved States”. It is not clear to me if funding exceptions, aka “pickups”, for grants scoring over the nominal payline are influenced by a desire for geographic diversity. I bet they are, because NIH seems to have so many ways to spread the money around that there must be a firm committment. Also, it takes no genius to understand that the NIH is a federal agency which depends on Congress to appropriate money to fund it. Keeping as many CongressCritters as possible happy with your work is crucial. Not spreading the grant awards across Congressional Districts would be a very stupid move indeed.
Fairness at review and special award mechanisms…so your institutional affiliation must be immaterial. Right? Well, these factors can help keep the $$ from all accruing to a narrow set of SuperUniversites over time but none of this is a guarantee. Particularly when it comes to your specific grant review.
Meh.
I should specify that I have no data on the more proximal question seemingly posed in the comment. At the individual applicant level, do they still pay a price for being at a less well-funded institution? If the CSR keeps track of success rates by institution size or amount of total NIH $$ I haven’t seen it or I can’t recall it.
So I’ll stick you with my anecdote and I’m sure some others will chime in with a comment.
I’ve never been struck by any overt bias against an application because the University / institution in question was too poorly funded, too unknown or anything that smelled like unjustifiable snobbery. I’m not saying it hasn’t happened, just that I don’t recall anything glaring. On the contrary, I can remember a time or two when the location was a plus, along the lines of “Look how much good stuff the PI has accomplished despite the environment”.
There are, certainly, times when specific environment support issues arise. No doubt about that. If the institution just doesn’t have the resources for BigTicketItem and it is not dealt with in the application, this is legit. If the application requires a collaboration with someone at BigU, that falls under the usual criticisms to show that it will work. I don’t get the sense, however, that the original question is asking about such matters.
Final thought is that I have heard rumour of fairly flagrant biases being directed against the applicant who is perceived as being in a very well-funded institutional setting. Not personally, that I can recall, but second hand accounts of “they have enough grant money already”, “they are just greedy” and the like. We get comments around here from people who like the idea of capping a given PI’s total award amount and from those who don’t like the soft-money job category fairly frequently. Therefore it is entirely unsurprising to me that such sentiments would occasionally bubble up at study section discussion.