Please, Just Kill Me Instead
January 23, 2008
The Center for Scientific Review at NIH just released a two-page PDF pamphlet called the “Insider’s Guide to Peer Review For Applicants”. This publication contains little squibs written by former and current study section chairs intended as advice for applicants about how to write a good R01 research grant application. (h/t WriteEdit.)
Much of the advice is trite, but unobjectionable. However one of the suggestions relates to something that is one of PhysioProf’s bêtes noires.
Here is the suggestion that has me grinding my teeth:
Spend time on the Expected Outcomes, Data Interpretation, Pitfalls, and Contingencies section for each aim. The “expected outcomes” section shows you’ve got a logical strategy. The section on Data Interpretation gives insight into your depth of understanding the problem. The Pitfalls section shows how familiar you are with the proposed techniques and methodologies. Finally, in discussing alternative strategies, you can give us confidence you are able to deal with the problems that arise when experiments don’t work as expected.
This is so wrong, on so many levels, it makes me want to poke my eyes out with a red hot poker.
First, this is the kind of overly pedantic, clenched-sphincter approach to scientific research that is threatening to kill the NIH research enterprise, by driving out all of the biomedical scientists doing creative novel research with a high potential upside, and leaving behind only the bunumbed goldbrickers doing the same boring crap day after day, month after month, year after year. If you are proposing a research project that is designed to generate new information by exploring new areas with new techniques or model systems, then how the heck are you supposed to know what to expect, what might mess it up, and how to get around any problems that might arise?
The only kinds of studies in which you can really describe any of that crapola in detail without just pulling it randomly out of your ass are ones in which you basically already know to a high degree of certainty what the experiment is going to tell you. That sure isn’t the “crosscutting, groundbreaking, innovative” research that NIH says they are looking for. Sounds more like ditch-digging to me.
And “Data Interpretation”? You know what they’re talking about? “We will compare the measured values for the experimental and control groups using an unpaired t-test. If the p value if less than 0.05, then we will conclude that the null hypothesis that the two groups do not differ must be rejected. Alternatively, if the p value is greater than 0.05, we will conclude that the null hypothesis must be accepted.” Yeah, that’s a useful way to fill limited page space.
Second, what kind of people on a study section are going to be able to even slog through this treated sewage sludge, digest it, assess it, determine if it is “satisfactory”, and then give you a better score in light of it? Exactly, it’s the same kind of halfway-to-rigor-mortis boring twits that think science amenable to this kind of ass-backwards pre-performance analysis is the only kind worth doing, and that all the interesting science is a “fishing expedition”, “not hypothesis driven”, or “too speculative”.
Third, and finally, who in their right mind wants to write this kind of useless garbage? It is crap like this that makes writing a grant application even more painful than a root canal. And it makes it nearly impossible for the writer to get excited and stay excited for the sustained period of time it takes to write an R01. I would rather drive nails through my foot than write this kind of a grant application.
All this kind of overly-pedantic formulaic gotcha crapola is a corollary of the extremely detailed and long-winded 25 page application format, with junk like “In Specific Aim #3, sub-Aim (A)(c)(i)(2), we propose to use 50 mM NaCl, whereas in sub-Aim (A)(c)(i)(1), we propose to use 25 mM NaCl”. And guess where the 25 page application is going? Yep, right in the toilet.
So, the good news is that this “suggestion” by some superannuated ossified study section chair about “Expected Outcomes, Data Interpretation, Pitfalls, and Contingencies” is, soon enough, going to just be superfluous, rather than actively pernicious, because when the R01 application is 10 pages, there ain’t gonna be room for any of this verbal and scientific torture.
Instead, you’re gonna have to impress the study section with the novelty, creativity, and possibly groundbreaking nature of your proposed studies, and not with the fact that the research is so pedestrian that you already know exactly what is going to happen, or at least have some enumerated list of possible “pitfalls”.
Frankly, even with the 25 page R01, I have considered this experimental design kabuki to be an insult to my intelligence and creativity, and an insult to the intelligence and creativity of the members of the study sections that review my grant applications. I have written a substantial number of 25-page R01 applications that were well-liked by study sections, have had multiple R01s awarded, and I have never taken this “suggestion” to heart. It is just too painfully boring and, in my opinion, unnecessary if you take other steps (which we will get into in future posts here) to make it clear to a study section that they can feel confident that you will be able to handle these things as they arise in the course of the research even in the absence of all the sphincter-clenching kabuki suggested in the CSR pamphlet.