Success in Attaining Independent Funding Among National Institutes of Health K Grant Awardees in Ophthalmology: An Extended Follow-up

Key Points

Question  How successful are ophthalmology clinician-scientists receiving National Eye Institute career development awards (K08 or K23) in obtaining R01 grant funding?

Findings  In this follow-up analysis of a cohort of National Eye Institute career development awardees, there was a 93% increase in the number of awardees who received an R01 grant. The mean time of receiving an R01 was nearly 3 years after completion of a K award.

Meaning  These findings suggest National Eye Institute career development awardees are more successful at obtaining independent funding than previously reported, but this can occur many years after the K award ended.

Importance  Understanding factors associated with attaining independent research funding by ophthalmology clinician-scientists receiving National Eye Institute career development awards (K08 or K23) in ophthalmology can be important to maintaining the pipeline of clinician-scientists.

Objective  To provide continued follow-up of a cohort of ophthalmology clinician-scientists who received National Institutes of Health (NIH) K career development grants.

Design, Setting, and Participants  Cohort study from an electronic database review of ophthalmologists who have received either a K08 or K23 career development grant from the NIH. Data were analyzed between December 30, 2015, and December 30, 2017.

Main Outcomes and Measures  Receipt of an NIH R01 grant.

Results  We previously characterized a group of more than 100 ophthalmologists who received K awards from 1996 to 2010, of whom 29 were awarded R01 grants. In follow-up of this cohort in 2017, 27 additional K awardees of this initial cohort were awarded an R01 from 2011 to 2017, leading to a total of 62 of 128 ophthalmologists receiving an R01. The mean time to receiving an R01 grant after the K award ended was 2.8 years. The data did not identify a definitive association with sex, having a PhD degree, or research tier of university in obtaining an R01 grant in this cohort.

Conclusions and Relevance  In comparison with our previous report of the same cohort, there was a 93% increase in the number of K awardees who have received an R01 award, with the mean time to award being nearly 3 years after completing their K grant. This suggests that most K awardees in ophthalmology are successful in obtaining R01 grants, but one should recognize this may be several years after their K grant has ended.

Clinician-scientists are highly trained physicians who use both their clinical and scientific expertise to advance research. These individuals play a vital role in ophthalmology, often serving as a bridge between basic scientists and clinicians to translate discoveries into new treatments for ophthalmic diseases. Much has been written about the dearth of clinician-scientists in ophthalmology and whether it is still a viable career pathway given the protracted training, financial pressure owing to educational debt, decreased financial remuneration compared with private practice, and intense competition for research funding.^1-4 The National Institutes of Health (NIH) K awards are a common pathway for development of clinician-scientists. This involves a 5-year career development award providing protected time and mentorship to develop an independent research program in either clinical or laboratory research.

In 2013,⁵ we characterized a cohort of ophthalmologists who have received NIH K awards and assessed their ability to get NIH R01 grants, a key benchmark for having an independent research program. We found a substantial decline in the proportion of K awardees who obtained R01 grants over time, with 63% of K awardees from 1996 to 2000 obtaining an R01 grant, while only 13% of K awardees from 2001 to 2005 obtained an R01 grant.⁵ In this report, we provide continued follow-up on this cohort 5 years later and expanded the analysis to include ophthalmologists who received K awards since our previous report.

The methods for this study were described in detail previously.⁵ K awardees were identified using the NIH’s online Research Portfolio Online Reporter Tools database (REPORTER).⁶ Search terms used were K08 and K23 award recipients from the National Eye Institute (NEI) from 1996 to 2017. Data, including the recipient’s name, year beginning the K award, training institution, and project title, were exported into a Microsoft Excel spreadsheet. These individuals were then individually searched in the REPORTER database to identify whether they received R01 or U10 grants. Demographic data were collected including sex, educational degrees, training institution, year of start of their K award, and clinical specialty. Because this was a search of a public online database, no institutional review board approval was needed.

K12 awardees were identified by emailing K12 directors at each of the respective programs with NEI K12 awards. A list of K12 awardees was provided by 4 of the 8 programs, which included the recipient’s name, sex, educational degrees, and award date. These names were then cross-referenced with the K08 and K23 awardees and searched in the NIH REPORTER database to determine whether they had received R01 funding. National Eye Institute budget and award data were also retrieved from the NIH REPORTER database.

Descriptive statistics for the K08 and K23 awardees are presented. Comparisons between K awardees who did and did not receive R01 grants were performed with the χ² test for categorical variables. Independent predictive factors for success in obtaining an R01 grant were assessed with logistic regression. Statistical analysis was performed using SPSS PASW statistics, version 17.0 (IBM). Odds ratios with 95% confidence intervals were calculated.

The demographic characteristics of ophthalmology K awardees are shown in Table 1. Previously, we had identified 106 ophthalmologists who had received an NIH K award from 1996 to 2010. We identified 75 additional ophthalmologists who received K awards from 2011 to 2017 from the NEI. Among this cohort of new K08 or K23 grant recipients, 27 were women (36%). Thirty-nine (52%) had a PhD degree in addition to their MD degree, and 57 (76%) came from ophthalmology departments ranked in the top 20 for NIH funding based on the Blue Ridge Medical Institute.⁷ We also identified 21 individuals who had received K12 grants during this period. Twelve had also received K08 or K23 awards and thus were already included within our analysis. Nine K12 awardees who did not receive an individual K08 or K23 award did not have a change in their R01 grant status.

We explored the NIH REPORTER database to determine whether previous K awardees had received new R01 grants since our last analysis in 2013. Previously, 29 of 105 K awardees from 1996 to 2010 had received R01 grants. In requerying the database, we identified 34 additional K awardees who received R01 grants since our last analysis (2 in the 1996-2000 cohort, 14 from the 2001-2005 cohort, 11 from the 2006-2010 cohort, and 7 in the 2011-2017 cohort).⁶ All R01 grants were awarded from the National Eye Institute. The proportion of K awardees who obtained R01 funding was 67% from 1996 to 2000, 46% from 2001 to 2005, 47% from 2006 to 2010, and 9% from 2011 to 2017 (Table 1). The Figure illustrates the time to obtain an R01 grant for K awardees. The mean (SD) time to get an R01 award after the K award ended was 2.8 (3.3) years. We also performed multivariate logistic regression analysis to identify factors associated with obtaining an R01 grant. Sex, having a PhD degree, research tier of university, or whether awardees had a K08 or K23 award were not significantly associated with attainment of an R01 grant in this cohort (Table 2).

In addition, we also explored the NIH REPORTER database to look at other NIH grants awarded to NEI K awardees other than the R01. U10 grants are large collaborative grants for investigator-initiated clinical vision research including clinical trials and large epidemiologic studies.⁸ We identified 4 K awardees of 180 K awardees from 1996 to 2017 who received U10 awards but not R01 grants. Because U10 grants are large multiyear grants for clinical research, we have considered them an “R01 equivalent” in terms of statistical analysis for Table 2. Additionally, we searched for NIH R21 awards, which are 2-year nonrenewable exploratory grants given to pilot projects to generate data for larger R01 grants.⁹ We identified 4 individuals who were awarded R21 grants but not awarded R01 grants. Because R21 grants are pilot project grants and are nonrenewable, we did not include them in our statistical analysis.

We used the NIH REPORTER to determine NEI budget expenditures from 1997 to 2017, and found that R01 award costs increased from $55.7 million in 1997 to $105.3 million in 2017 (Table 3).¹⁰ During the period for which K award data are available, the number of awards from the NEI increased from 43 in 1998 to 79 in 2017 (Table 3).

We have extended our follow-up of ophthalmologists who received NIH K awards to better understand the outcomes of clinician-scientists pursuing independent research within ophthalmology. The newest cohort of ophthalmologists from 2011 to 2017 who received K awards had similar characteristics to those from previous years in terms of sex, whether they had a PhD degree, and the research tier of institution where they received the grant. Interestingly, in our follow-up analysis of K awardees from 1996 to 2010, a substantial number of K awardees who did not obtain R01 grants in our last analysis were found to have received NIH R01 grants when we requeried the NIH REPORTER database 5 years later. This might be surprising because most of these individuals had already finished their K award and would presumably no longer have salary support for protected research time. This suggests that they may be receiving bridge funding to continue their research programs from the department, private foundation awards, or Veterans Affairs awards. In support of this notion, we found the mean time to obtain an R01 grant after completion of the K award is 2.8 years, indicating that there is often a significant gap between completion of the K award and receiving an R01 grant. Therefore, many K awardees continued with their research projects even after their NIH salary support ended. In addition, 18% of the K awardees (11 of 62) received their R01 grants 5 years or more (the longest being 12 years) after their K awards had elapsed, indicating a persistent desire to obtain research funding. Expanding on this, the 1996 to 2000 cohort had a higher rate of obtaining an R01 than the later groups, which on initial approach seems intuitive given these individuals have simply had more time in which to obtain funding. However, in examining those who received K awards during 2001 to 2005, they do not have an increased rate of obtaining an R01 compared with those from 2011 to 2012. Thus, the mean length of time from the end of a K award to obtaining an R01 does not fully illustrate how large the gap is between these awards for clinician-scientists in ophthalmology because those who received a K award more than a decade ago have no higher likelihood of obtaining an R01 compared with those whose funding only recently ended. Although interesting, it would be difficult to ascertain exactly why this is the case, although the aforementioned disincentives likely play a role.

One interesting finding from our study was that K awardees with combined MD/PhD degrees did not have an increased chance of obtaining an R01 grant compared with MD degrees alone. This was also found in our previous analysis. While our numbers are small, larger analysis by the NIH in general showed that MD/PhDs and MDs with prior K awards both had a 66% R01 award rate.¹¹

Comparison of R01 award rates for established vs new investigators lends additional context to these findings. A separate large NIH analysis in 2012 found the award rate for both MDs and MD/PhDs with a prior R01 was 23.3%, compared with an award rate of 14.9% for MD/PhDs and 14.7% for MDs without a prior R01.¹² While the overall award rate has generally decreased across all types of investigators, there has been a reduction in the gap between the award rates for new vs established R01 applicants, which the NIH notes is partly attributable NIH’s New Investigator policies implemented in 2007 and strengthened in 2009.¹²

One consideration that determines, in part, the success of obtaining an R01 grant is the fluctuating NIH budget. The NIH budget has remained relatively flat in inflation-adjusted costs since 2003, despite rapid increase in the number of applications. Indeed, the rate of funding for new applications at the NEI is 20% vs 48.6% for renewal applications.¹⁰ We examined NEI budget expenditures from 1997 to 2017 and found that R01 award costs increased from $55.7 million in 1997 to $105.3 million in 2017. During the period for which K award data are available, the number of awards increased from 43 in 1998 to 79 in 2017. In addition, the number of new K awards from the NEI each year from 2008 to 2017 has remained steady. Hence, while there are year-to-year fluctuations in budget and awards, the overall trend has been markedly positive.

There is limited information assessing how K awardees perform in comparison with other scientists and how K awardees from the NEI perform in obtaining research funding compared with those of other NIH institutes. Data looking at the outcomes of K awards from the National Cancer Institute (NCI) indicate 56% of K awardees in the NCI received further NIH funding (including T32, R01, and U10 grants).¹³ However, this comparison cannot be fully conducted because the NCI data include other K awards that do not apply to clinician-scientists, such as K01, K11, K22, and K25 mechanisms, although it appears the NEI and NCI have comparable rates of K awardees obtaining future independent NIH funding. Further research in comparing clinician-scientist outcomes between institutes would be interesting. Another study¹⁴ looking at K08 and K23 awardees across the entire NIH from 1997 to 2003 showed a success ratio of 30% in obtaining R01 grants (n = 2784).¹⁴ This suggests that K awardees from the NEI were at least as successful in obtaining R01 grants, and possibly more successful if followed up over a long enough period.

There are multiple limitations of this study. We focused our outcomes on receipt of the NIH R01 or U10 grant, large multiyear grants that we defined as the benchmark of a substantial research program. We did not assess other major grants, such as Veterans Affairs grants and private foundation grants, because these are not easily accessible in a publicly available database. We also did not assess receipt of other NIH grants, such as R21 and R03 grants, because these are generally considered precursors to obtain R01 funding. Optometrists were not included in this study because they may have differing scope of clinical responsibilities and challenges compared with ophthalmologist clinician-scientists.

Beginning in 2013, numerous publications have discussed strategies to retain and nurture clinician-scientists in ophthalmology and whether the NIH K awards are a successful pathway to develop clinician-scientists in ophthalmology.^1-4 The data presented here suggest that NEI K awardees’ ability to obtain NIH R01 grants is better than we had previously reported, although it may take several years after the K award is finished for individuals to obtain independent research funding. Continued study of the development of clinician-scientists is needed to develop best policies to maintain this vital cohort of ophthalmologists.

Corresponding Author: Daniel L. Chao, MD, PhD, Shiley Eye Institute, University of California, San Diego, 9415 Campus Point Dr, La Jolla, CA 92093 (dlchao@ucsd.edu).

Accepted for Publication: July 12, 2018.

Published Online: September 27, 2018. doi:10.1001/jamaophthalmol.2018.3887

Author Contributions: Dr Chao had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Protopsaltis, Chen, Gedde, Chao.

Acquisition, analysis, or interpretation of data: Protopsaltis, Chen, Hwang, Gedde.

Drafting of the manuscript: Protopsaltis, Chen, Hwang.

Critical revision of the manuscript for important intellectual content: Protopsaltis, Chen, Gedde, Chao.

Statistical analysis: Chen, Hwang.

Supervision: Protopsaltis, Gedde, Chao.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: Supported by National Institutes of Health center core grant P30EY014801, a Research to Prevent Blindness unrestricted grant, and Department of Defense grant W81XWH-09-1-0675.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.