Application, Review, Funding, and Demographic Trends for Maximizing Investigators’ Research Awards (MIRA): FYs 2019-2021

4 comments

In this Feedback Loop post, we revisit our previous analysis of application, review, funding, and demographic trends for the Maximizing Investigators’ Research Award (MIRA) program over Fiscal Years (FYs) 2019 to 2021. We look at trends for applicants by race/ethnicity and by gender. Due to privacy requirements and small numbers, applicants from underrepresented racial and ethnic groups in biomedical research (Black or African American, Hispanic or Latino, American Indian or Alaska Native, Native Hawaiian and other Pacific Islander) are combined into a larger group that can be reported. Because of the small applicant numbers, we’re unable to show intersectional analyses of race/ethnicity and gender or analyses of applicants with disabilities. 

Table 1 shows the number of new awards made and associated award rates by fiscal year for Established Investigators (EIs) and Early Stage Investigators (ESIs).

Table 1: Number of New (Type 1) MIRAs Funded by Fiscal Year and Cohort (Percentage of Reviewed Applications)

FY Established Investigators Early Stage Investigators
2019 187 (55.5%) 150 (41.9%)
2020 188 (55.0%) 200 (42.4%)
2021 217 (51.5%) 237 (42.5%)

Table 2 displays demographic information for the various steps of the application, review, and award process for EI MIRA applicants over this period. There is no clear statistically significant difference in the proportion of applications reaching review or discussion between groups, although the difference between discussion rates by race and ethnicity groupings nearly meets statistical significance (p = 0.07). We do note a statistically significant difference in the priority scores, with White applicants tending to have numerically lower (better) scores than applicants from other groups.

There are also differences in the percentage of applications awarded, driven mostly by a higher rate for White applicants. This is significant whether we consider all reviewed applications or only those that were scored. While overall we see a statistically significant difference, the only pairwise difference that reaches statistical significance is between White and Asian applicants. However, the small number of EI MIRA applicants from underrepresented groups and the lower funding rates for non-White applicants emphasize the need to strengthen efforts in enhancing the diversity of the biomedical research workforce and ensuring fair and equitable processes for review and funding decisions.

Table 2: Review Process by Race/Ethnicity, EI MIRA. For all tables, p-values reflect Fisher’s exact test results across all groups for review, discussion, and award outcomes, as well as analysis of variance tests for review scores.

White
Asian
Underrepresented
Unknown/Withheld/Multiple
P-Value
Apps Received (N)
774
306
52
69
NA
Apps Received (% of Total)
64%
26%
4%
6%
NA
% Reviewed Within Group
92%
91%
>80%
>80%
0.7
% Discussed Within Group
76%
69%
67%
70%
0.07
Mean Score
36.3
40.9
39.5
38.6
0.002
Awards (N)
414
128
22
28
NA
% Scored Apps Awarded
70%
60%
63%
58%
0.03
% Reviewed Apps Awarded
58%
46%
47%
45%
0.003

Table 3 provides information on the same stages as Table 2, split by applicant gender. Again, we see no statistically significant difference in the proportion of applications reaching review or discussion, but do see a difference in priority score and award rate (percentage of reviewed applications that are awarded). Women applicants tend to have a slightly numerically lower (better) score than men, as well as a higher percentage of both scored and reviewed applications that are awarded.

Table 3: Review Process by Gender, EI MIRA

Men
Women
P-Value
Apps Received (N)
906
291
NA
Apps Received (% of Total)
76%
24%
NA
% Reviewed Within Group
91%
94%
0.1
% Discussed Within Group
73%
77%
0.2
Mean Score
38.2
35.8
0.04
Awards (N)
426
166
NA
% Scored Apps Awarded
65%
74%
0.01
% Reviewed Apps Awarded
52%
61%
0.01

Table 4 displays demographic information for ESI MIRA applicants over this period. We see no statistically significant difference at any stage of the process between groups.

Table 4: Review Process by Race/Ethnicity Grouping, ESI MIRA

White
Asian
Underrepresented
Unknown/Withheld/Multiple
P-Value
Apps Received (N)
832
456
126
152
NA
Apps Received (% of Total)
53%
29%
8%
10%
NA
% Reviewed Within Group
90%
87%
87%
87%
0.2
% Discussed Within Group
60%
58%
58%
59%
0.9
Mean Score
41.8
40.7
43.7
41.5
0.5
Awards (N)
322
168
44
53
NA
% Scored Apps Awarded
65%
64%
60%
60%
0.7
% Reviewed Apps Awarded
43%
42%
40%
40%
0.9

Table 5 provides the ESI MIRA information by applicant gender. Again, there is no difference at any stage of the process among ESI applicants.

Table 5: Review Process by Gender, ESI MIRA

Men
Women
P-Value
Apps Received (N)
977
491
NA
Apps Received (% of Total)
67%
33%
NA
% Reviewed Within Group
89%
89%
0.9
% Discussed Within Group
60%
60%
1
Mean Score
41.7
41.2
0.6
Awards (N)
375
191
NA
% Scored Apps Awarded
63%
65%
0.6
% Reviewed Apps Awarded
42%
44%
0.7

Figure 1 shows the race/ethnicity breakdown of NIGMS R01 and MIRA-supported principal investigators (PIs) over this period for both the EI and ESI pools. The demographics between the EI R01 and R35 (MIRA) portfolios are similar, while the ESI R01 portfolio has a higher percentage of Asian PIs and a lower percentage of underrepresented minority PIs than the ESI R35 portfolio. Note that the ESI MIRA pool is considerably larger than the ESI R01 pool, with MIRA accounting for nearly 75% of all new (Type 1) NIGMS ESI R01, DP2, and R35 awards over the period of FYs 2019 to 2021.

Race/Ethnicity Breakdown of NIGMS R01 and R35 PIs by Cohort

FYs 2019-2021 Only

Figure 1: Race/Ethnicity Breakdown of NIGMS R01 and R35 PIs by Cohort. Illustrated are the percentages of NIGMS R01 and R35 MIRA PIs in each race/ethnicity group split by EI and ESI. The striped purple bars represent the percentage of PIs in each group who are White, the dotted blue bars represent the percentage of PIs who are Asian, the crosshatched green bars represent PIs from underrepresented groups, and the solid yellow bars represent PIs with unknown/withheld/multiple race or ethnicity.

Figure 2 shows the gender breakdown of NIGMS R01- and MIRA-supported PIs over this period for both the EI and ESI pools.

Gender Breakdown of NIGMS R01 and R35 PIs by Cohort

FYs 2019-2021 Only; Excludes Unknown/Withheld

Figure 2: Gender Breakdown of NIGMS R01 and R35 PIs by Cohort. The graph shows the percentages of men and women NIGMS R01 and R35 MIRA PIs split by EI and ESI status. The solid blue bars represent the percentage of PIs in each group who are men, and the solid tan bar represents the percentage of PIs who are women.

Overall, the ESI MIRA pool has the highest percentages of women PIs and those from underrepresented racial and ethnic groups. Although this is encouraging, neither level reflects the demographics of Ph.D. graduates in the life sciences (58% women and 14% individuals from underrepresented groups, excluding temporary visa holders), indicating the need for continued work to enhance the diversity of biomedical research faculty. NIGMS will continue its efforts in this area, which include programs such as MOSAIC, and we will support NIH-wide efforts led by the UNITE initiative.

4 Replies to “Application, Review, Funding, and Demographic Trends for Maximizing Investigators’ Research Awards (MIRA): FYs 2019-2021”

  1. Thanks for this report. Do you have any data on the success rate of MIRA renewals for EIs and ESIs? Asking for a friend…

  2. Have you looked at whether women/URMs submit fewer total grant applications per year? I am wondering it this also might contribute to why grant applicants don’t match the workforce demographics – especially with ESI R35 vs RO1 differences.

    1. We’ve not looked at NIH applications per investigator since 2018; at that time we reported that men do apply more often than women for new and renewal projects.

Submit a Comment

Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.