Time to first R01-level NIH grant

I’ve started thinking about lab budgets and how far you can get on startup funds / how long you may need to try to stretch them out. Clearly one of the biggest factors in making a projected budget (say, a 5 year budget) is when you’re able to obtain your first R01-level external grant, which really helps you sustain a small-to-moderate size lab for 4 to 5 years. The general impression I’ve gotten is that many institutions want / expect you to get one within 3 years or so, so I started building this assumption into my projected budget. But Anna reminded me that I may already have this data (see this post), so I could take a data-driven perspective on this. Turns out that I didn’t quite have the relevant data in hand, but it did become pretty clear that I could quite easily obtain this dataset.

I went bak to NIH RePORTER (I’m getting so much mileage out of this site!), and the NIH grant funding history for 59 late assistant, associate, or early full professors. I then looked at their LinkedIn profiles (again, what a useful resource) to figure out when they started their faculty positions. With both datasets in hand, I calculated how many days it was between when they started their position, and when they first started seeing funds from an R01 / DP2 / R35 level grant. Three individuals were instructors when they got their first R01 (and seemingly turned these into tenure-track faculty positions), so I left them out of my analysis. Six had not received their first grant yet (the shortest was ~550 days, whereas the longest was ~2500 days… yikes!) leaving me 50 data-points (mostly people from UW, Boston, my PhD field (molecular HIV research), or schools where I had interviewed). When plotted as a histogram, this is what it looked like:

The earliest was 408 days (there were also 428 and 456 days; one of them was Jesse Bloom, of course…), and the latest was 3179 (though they were at a smaller, R2 institution, so perhaps the rules / expectations / possibilities are different there). This is still a relatively small and potentially biased dataset, and it only looks at NIH funding and not NSF (though I tried to keep to biomedical researchers to avoid this confounder), or DOD. With those caveats in mind, the median ended up being just about three years. But as the plot makes clear, it really is a distribution. What kind of stress were the latter half of individuals experiencing as they were frantically writing grants while wondering how they were going to keep their lab together without money? Statistically speaking, it’s kind of a scary situation: small n values (maybe a dozen or so attempts), with low probabilities of success (a vast minority of submissions are funded), a large stochastic component (reviewers), and very large effect (rejected = no money, accepted = considerable monetary support). But, that is how some systems in the world are structured (whether they should be changed is a different matter).