NIH and NIGMS have policies to promote the successful entry of junior investigators into independent biomedical research careers. NIH classifies investigators who have not previously had a major NIH grant into two categories: new investigators (NIs) and early stage investigators (ESIs), a subset of NIs who are within 10 years of completing their terminal research degree or medical residency. The goal of these policies is to support R01-equivalent awards to both of these categories of investigators at success rates (the percentage of new Type 1 R01 applications that were funded) similar to those of established investigators (EIs) who submit new R01 applications.

Given that the NI and ESI policies have been in effect for some time, we wanted to update and extend an analysis of success rates by investigator status performed in 2010 to see if NIGMS has been able to meet these objectives. While we found that the success rates for all NIs were comparable to or greater than that of EIs, our new analysis also revealed that the subset of NIs who completed their terminal research degree at least 10 years ago (non-ES NIs) had consistently lower success rates in obtaining R01s relative to both ESIs and EIs.

We focused our analysis on NIGMS Type 1 R01 applications for Fiscal Years 2011-2014. Figure 1a shows the success rates for EIs and NIs. During the time period analyzed, success rates for both EIs and NIs were comparable. However, when the NIs are separated into ESIs and non-ES NIs, the data show a more nuanced result (Figure 1b). ESIs consistently had higher success rates than either EIs or non-ES NIs when applying for new R01s.

NIGMS has maintained a policy of “reaching” to fund meritorious non-ES NI and ESI applications at higher rates across a range of percentiles; thus, the lower success rates for non-ES NIs were confounding. A graph of the percentage of applications funded by percentile for each investigator category was generated for Fiscal Year 2011, and the evidence of our actions is seen in Figure 2, which shows a rightward shift in the funding curves for applications from non-ES NIs and ESIs relative to the curve for applications from EIs. The patterns for Fiscal Years 2012-2014 paralleled that of Figure 2 (data not shown).

To better understand the difference in success rates for the two categories of NIs, we analyzed their application distribution by percentile. Interestingly, an average of 52% of the non-ES NI applications were unscored over this 4-year period. While this rate was comparable to that of EI applications at 53%, it was moderately greater than that of ESI applications, which was only 40%. The larger number of unscored applications could be one contributor to the lower application success rates for non-ES NIs. Another factor driving down the success rate for the non-ES NI applications appears to be the smaller proportion of scored applications that are funded. On average, only 24.8% of non-ES NI applications that were scored over the time period analyzed were awarded, compared to 32.8% of ESI applications and 33.5% of EI applications. In addition, the distribution of non-ES NI applications is skewed toward the higher percentile range. This result is demonstrated in the funding distributions for Fiscal Year 2011 shown in Figure 3a as well as in the table in Figure 3b, which provides the mean and median percentiles of scored applications from both ESIs and non-ES NIs for Fiscal Years 2011-2014. While the mean and median percentiles for funded applications are equivalent for both investigator categories, the mean and median percentiles for the entire non-ES NI application pool are higher than those for ESIs. This suggests that the non-funded, non-ES NI applications are more heavily skewed toward the higher percentile range.

We next looked at the age distribution of NIGMS-funded investigators over the time period analyzed to further understand the difference in age between these groups and to see if there are any trends. While other reports have shown that the median age of first-time R01 grantees is around 42 years, we again found that the distinction between ESIs and non-ES NIs is an important one to make. As Figure 4 shows, there is a significant difference in the age distribution of investigators at the first year of the new award among the three categories. Funded ESIs have a median age of approximately 38 years, while funded non-ES NIs and EIs have median ages of approximately 44 and 51 years, respectively. There is no significant difference between the median ages of those funded and the applicant pool of each category as a whole. This result suggests that there is no bias toward funding a specific population of applicants within each investigator type based on age.

As these data show, NIGMS has supported ESIs at success rates comparable to those of Type 1 applications from EIs, which is consistent with NIH goals. Although the success rates of non-ES NI applications have been lower than those of other investigator types, NIGMS has consistently "reached" to fund applications from non-ES NIs as well. These data bring to light distinctions between the ES-NI and non-ES NI populations and indicate that considering them separately in ongoing monitoring and evaluation at NIGMS is useful.

Acknowledgements

Age data were provided by the Statistical Analysis and Reporting Branch, NIH Office of Extramural Research, and analyzed by the NIGMS Office of Program Planning, Analysis, and Evaluation.

Definitions

New investigator: An individual who has not previously competed successfully as a program director/principal investigator for a substantial NIH independent research award.

Early stage investigator: An individual who is a new investigator and is within 10 years of completing the terminal research degree or medical residency (or the equivalent).

Non-early stage new investigator: An individual who is a new investigator and is beyond 10 years of completing the terminal research degree or medical residency (or the equivalent).

Type 1 application: An application for a new grant.

Success rate: The percentage of new Type 1 R01 applications that were funded.