A longitudinal analysis of NIH R-Funded Investigators: 2006-2013

Apr 28 2014 Published by under Uncategorized

Datahound's first post...

Most of the data readily available from NIH RePORTER are on a 'per grant' basis. However, these data may be more revealing when they examined on a 'per investigator' basis. To obtain such data, the results for all R mechanism grant awards were downloaded from NIH RePORTER and aggregated the data for each unique investigator.

Disclaimer:  While I have done my best to perform this analysis fairly and accurately, there are a small number of records for which the interpretation is problematic. These represent small amount of funding although their inclusion/exclusion could influence the inferred number of investigators to a small degree. Only the contact PI was included for multiple PI awards.

R mechanisms from R01 through R56 were included with the exception of R13 awards (conference grants) and RC1 and RC2 awards made for FY09 and FY10 as part of the American Recovery and Reinvestment Act (ARRA). Approximately 75% of the awards examined were R01s, 10% R21s, 5% SBIR/STTR (small business) awards, and 10% other mechanisms. Note that the number of awards includes all award types including new, competing continuation, non-competing, and supplements rather than unique grant numbers.

The number of awards and the number of investigators are shown below.

2006 2007 2008 2009 2010 2011 2012 2013
Awards 38904 38941 38534 46478 42692 36577 36938 35131
Investigators 27300 27417 27189 29082 28954 26525 26362 25361

Aggregated over all 8 years, a total of 53526 investigators were funded by these mechanisms.

From these data, the number of investigators who left the R-funded investigator pool each year (that is, those who were funded one year but not funded the next year) as well as the number of investigators who joined the funded investigator pool (that is, those who were not R-funded one year but were in the subsequent year) could be determined. The dynamics of the investigator pool determined in this way is shown below:

2006-2013 investigator dynamics-2

The results of this analysis reveal the impact of ARRA and the sequester. From 2006 to 2007, the number of R-funded investigators increased by 117 as 5317 investigators funded in fiscal year 2006 were apparently no longer funded in 2007 while 5434 investigators not funded in 2006 were now funded in 2007. A drop of 228 investigators occurred from 2007 to 2008.  In contrast, with the advent of ARRA, the large bolus of investigators joined the funded pool in 2009.  Note that this analysis does not include RC1 ("Challenge") and RC2 ("Grand Opportunity") grants which would presumably swell these numbers further. The number of investigators held fairly steady from 2009 to 2010 but then fell dramatically after ARRA with 7059 investigators leaving the funded pool and only 4628 investigators added in 2011. The number of investigators dropped slightly again from 2011 to 2012 and then more substantially in 2013 due to the sequester with a net loss of 1001 investigators. Overall, over this period, the number of R-funded investigators dropped by approximately 6%.

Since there are a total of 53526 unique investigators and approximately 27000 investigators for each of the eight fiscal years, each investigator, on average, was funded for a total of four years during this period ((27000 X 8)/53528 = 4.03).  More on the distribution of funding for these investigators soon...


14 responses so far

  • drugmonkey says:

    Welcome aboard, Datahound!

  • odyssey says:

    Welcome! I'm looking forward to this.

  • drugmonkey says:

    To business!

    I'm surprised that ARRA caused so many temp PIs to enter/exit. I'd be curious what this was... I thought ARRA was heavily dominated by the usual suspects. Did it just drop dollars on some of the usual ~5,000 year to year churn? Or were these new folks who never moved beyond the ARRA largesse?

  • DJMH says:

    Yes, I am with DM. Everyone I knew who got ARRA money was already funded. Was there another pool of people I didn't see, or did the presence of the ARRA money mean that some investigators didn't have to go after a second R01, thereby freeing up that money for other folks?

  • Datahound says:

    Hi DM: This is an extension of the analysis that I have done. Let me see what I can do.

  • thorazine says:

    Does including DP1/DP2 recipients make any significant change in these numbers? Certainly, these are R01-comparable, and I'd expect including DP1 to reduce apparent churn by at least a bit.

    • datahound says:

      I agree that these are conceptually similar. There are too few DP1s to make much difference and DP2s are a bit of a technical challenge as all 5 years of each award is funded out of 1 fiscal year. I may try to include them if I redo that analysis.

  • PI in RI says:

    It does look like ARRA money went more towards investigators already in the pool ("shovel-ready", wasn't this the point?) since there were ~8000 more awards but less than 2000 more investigators.

    So, on average investigators were R-funded half the time. Anxious to see if distinct populations are apparent in the histograms.

    Really looking forward to seeing your consistently enlightening analyses on this blog, Dr. Berg!

  • whizbang says:

    Welcome to Scientopia. Saw you in the elevator at EB this morning, but didn't realize you were debuting with us today!

  • namnezia says:

    Welcome to scientopia!

  • […] it comes to climate change, the intellectual bankruptcy of the conservative movement is stunning A longitudinal analysis of NIH R-Funded Investigators: 2006-2013 NDM: “A Great Challenge for the Future of […]

  • […] Jeremy Berg has been doing a good job of starting the process of examining the career aspects, see here and here. I think understanding the PI dynamics is critical to achieving some sort of stabilization […]

  • […] has been a bit over 2 years since Datahound was born. I have greatly enjoyed being brought into the Scientopia team and sharing data that I have gathered […]

Leave a Reply