[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure 1.  National Institute of Child Health and Human Development Training Award Expenditures as a Percentage of Extramural Funding
National Institute of Child Health and Human Development Training Award Expenditures as a Percentage of Extramural Funding

The National Institute of Child Health and Human Development has maintained a strong investment in extramural training and career development over the last several decades. In 2016, The National Institute of Child Health and Human Development invested 5.5% of its extramural funding in training and career development programs.

Figure 2.  Proportion of Training Dollars Allotted to the Different Training Mechanisms for Several National Institutes of Health (NIH) Institutes
Proportion of Training Dollars Allotted to the Different Training Mechanisms for Several National Institutes of Health (NIH) Institutes

Institutes support training and career development through a number of award mechanisms including career development awards (K) and training mechanisms (F and T). K awards include those that are individual (K08 and K23) and institutional (K12), and training awards also include those that are institutional (T32) and individual (F). The National Institute of Child Health and Human Development (NICHD) allocates a higher percentage (38.3%) to institutional career development awards (K12) than other NIH institutes (range 0-11.2%). NCI indicates National Cancer Institute; NHLBI, National Heart, Lung, and Blood Institute; NIAID, National Institute of Allergy and Infectious Diseases; NIAMS, National Institute of Arthritis and Musculoskeletal and Skin Diseases; NIDA, National Institute on Drug Abuse; NIDDK, National Institute of Diabetes and Digestive and Kidney Diseases; NIGMS, National Institute of General Medical Sciences; NIMH, National Institute of Mental Health; NINDS, National Institute of Neurological Disorders and Stroke.

Figure 3.  National Institute of Child Health and Human Development Success Rates of Individual Career Training Awards Over Time
National Institute of Child Health and Human Development Success Rates of Individual Career Training Awards Over Time

The National Institute of Child Health and Human Development success rates for individual career development awards (K08 and K23) have declined since 2000.

Figure 4.  Person Application and Funding Rates for 257 Physician-Scientists With an MD Only, Supported by Career Development Awards Between 1999 and 2001 and Applying for Subsequent National Institutes of Health Research Grants
Person Application and Funding Rates for 257 Physician-Scientists With an MD Only, Supported by Career Development Awards Between 1999 and 2001 and Applying for Subsequent National Institutes of Health Research Grants

Scholars who had an MD degree only and received an individual K award or both an individual K award and institutional K12 support were more likely than those who received only institutional K12 support to apply for subsequent research project grant funding (88.0% vs 82.8% vs 60.1%, respectively; P < .001) and to receive it (60.0% vs 60.9% vs 32.9%, respectively; P < .001).

Figure 5.  Subsequent Careers of National Institute of Child Health and Human Development Physician-Scientists Supported by Career Development Awards
Subsequent Careers of National Institute of Child Health and Human Development Physician-Scientists Supported by Career Development Awards

Both individual K and institutional K12 programs produced a cohort of scholars who largely continued in academic careers. Physician-scholars in the K12-only group were more likely to go on to careers in clinical care compared with physician-scholars who received an individual K award only or both individual K and institutional K12 support (16.8% vs 12.3% or 7.3%, respectively; odds ratio, 2.61; 95% CI, 1.05-6.76).

1.
 Century Cures Act. Pub L No. 114-255, 130 Stat 1033-1344 (2016). https://www.congress.gov/bill/114th-congress/house-bill/34/text#toc-H3ED134CB78594644AE2028D8BCE93275. Accessed October 19, 2017.
2.
National Institutes of Health.  Physician Scientist Workforce Report. https://report.nih.gov/workforce/psw/, Published 2014. Accessed June 15, 2017.
3.
Marsh  JD, Todd  RF  III.  Training and sustaining physician scientists: what is success?  Am J Med. 2015;128(4):431-436.PubMedGoogle ScholarCrossref
4.
National Institute of Child Health and Human Development.  Review of NICHD Training and Career Development Programs. https://www.nichd.nih.gov/training/extramural/Documents/NICHD_training_review.pdf. Accessed June 15, 2017.
Original Investigation
March 2018

Association of National Institute of Child Health and Human Development Career Development Awards With Subsequent Research Project Grant Funding

Author Affiliations
  • 1Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland
JAMA Pediatr. 2018;172(3):226-231. doi:10.1001/jamapediatrics.2017.4305
Key Points

Question  What is the success rate of individuals supported by the National Institute of Child Health and Human Development career development and training awards?

Findings  In this cohort study of 355 physician-scholars, those who received an individual K award or both individual K and institutional K12 support were more likely to apply for and receive subsequent research project grant funding compared with those who received only institutional K12 support.

Meaning  Physician-scholars are more likely to apply for and receive an independent National Institutes of Health research grant if they are trained on an individual career development award, with or without an institutional training award.

Abstract

Importance  Investing in the next generation of researchers is essential, as recently highlighted in the 21st Century Cures Act. From its inception, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) has supported training and career development to ensure a robust pipeline of investigators who are prepared to lead their respective fields of inquiry. In recent years, the NICHD has emphasized institutional over individual training awards to a greater degree than many other National Institutes of Health institutes of similar size.

Objective  To evaluate the success of individuals supported by NICHD career development and training awards, as measured by subsequent application and receipt of independent National Institutes of Health research project grant funding.

Design, Setting, and Participants  This retrospective cohort study identified 417 physician-scholars who were supported by NICHD career development awards between October 1, 1999, and September 30, 2001. This period was selected to allow adequate follow-up of research project grant applications and funding. Among these physician-scholars, 355 met inclusion criteria.

Main Outcomes and Measures  The primary outcomes were the numbers of research project grant applications submitted and, of these, the numbers that were successfully funded.

Results  Among 355 physician-scientists, scholars who had an MD degree only and received a K award or both an individual K award and institutional K12 support were more likely than those who received only K12 support (individual K only vs institutional K12 only: odds ratio [OR], 4.86; 95% CI, 1.83-13.62; both K and K12 vs K12 only: OR, 3.19; 95% CI, 1.46-7.10) to apply for subsequent project grant funding (88.0% vs 82.8% vs 60.1%, respectively; P < .001) and to receive it (60.0% vs 60.9% vs 32.9%, respectively; P < .001). For physicians with both MD and PhD degrees, neither application nor funding rates were statistically significantly different whether their career development was supported by individual or institutional awards.

Conclusions and Relevance  Physician-scholars are more likely to apply for and receive a National Institutes of Health research grant if they are trained on an individual career development award, with or without an institutional training award. Based on the data, the NICHD intends to provide a greater proportion of its career development fund allocation to individual awards. The NICHD recognizes the importance of institutional awards and will continue to support them. The NICHD remains committed to training and intends to maintain its investment in training and career development awards going forward.

Introduction

Training the next generation of physician-scientists is vital for the research enterprise. From its inception, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) has emphasized support for training and career development to ensure a robust pipeline of scientists and clinician-scientists who are prepared to lead their respective fields of inquiry. The 21st Century Cures Act, signed into law on December 13, 2016, also highlights the importance of investing in the next generation of researchers, specifically calling to provide and promote opportunities for new researchers and earlier research independence.1

During the past 3 decades, the NICHD has maintained a strong investment in extramural training and career development (Figure 1). In 2016, the NICHD invested 5.5% of its extramural funding in training and career development programs, a proportion that exceeds or is similar to the proportions at other institutes of like size at the National Institutes of Health (NIH), including the National Institute on Drug Abuse, the National Institute of Mental Health, and the National Institute of Digestive Diseases and Kidney Disorders. The NICHD is committed to spending a similar percentage of its extramural funding on training and career development activities going forward.

At the NIH, the institutes and centers use grant award mechanisms that are best suited to their constituents’ needs and research goals. The NICHD grant mechanisms for training and career development fall into 2 main categories: individual and institutional. To apply for an individual fellowship or career development award, an early-stage investigator identifies both a project and a mentor and applies directly to the NIH using mechanisms such as the F32, K08, or K23. For an institutional award, an academic institution applies to the NIH under the T32 or K12 mechanism, providing a description of the overall program. If the application is successful, the institution is allocated a specific number of positions; individual trainees or scholars are then selected by the institution to fill those positions. For some K12 programs, scholars are selected from the institution that receives the K12 grant. These local K12 programs include the Child Health Research Career Development Award, Women’s Reproductive Health Research, and Building Interdisciplinary Research in Women’s Health. Other K12 programs coordinate and manage career development activities for scholars located at many different institutions across the country. These national K12 programs include the Pediatric Scientist Development Program, Reproductive Scientist Development Program, and Rehabilitation Medicine Scientist Training. In addition, training and career development are often conducted within research project grants, at clinical research centers, and in university settings not supported by the NIH.

There are benefits and limitations to both institutional and individual career development awards. In fields in which few mentors are available, institutional awards may help attract scholars and in the long term may increase the number of mentors. Institutional awards also may provide opportunities for training in areas outside a specific clinical specialty, such as basic or cross-disciplinary research. In some situations, institutional awards are used for 2 to 3 years to provide basic research skills and to prepare scholars to apply for an K award. Local institutional awards may not be ideal in cases in which the mentors do not match well with a scholar’s scientific interests. Furthermore, institutional awards provide resources only to select institutions.

Individual awards provide scholars with opportunities, such as direct experience writing a grant application, that allow the individual to develop a specific project tailored to his or her career needs and goals. Limitations to individual awards include the absence of a peer group and the potential lack of opportunities to learn from shared experiences, such as didactic programs that larger institutions may be able to provide. However, individual awards allow all institutions to be competitive in terms of funding scholars.

In recent years, the NICHD has emphasized the institutional training and career development awards to a greater degree than many other NIH institutes and centers. In fiscal year 2014, institutional programs accounted for 73.7% of the NICHD’s expenditures on training and career development awards (Figure 2). This difference is largely based on the NICHD’s strong investment in K12 career development programs. The K12 programs are targeted to clinician-scientists in areas of the NICHD mission, including pediatrics, obstetrics and gynecology, and rehabilitation medicine. Over time, as new programs have been added and others expanded, expenditures on K12 programs have grown at a faster rate than expenditures in other training and career development mechanisms, resulting in K12 awards accounting for a higher proportion of the overall training and career development budget (eFigure 1 in the Supplement). Concurrently, the success rate for individual NICHD career development and training applications has fallen (Figure 3). In a comparison of the NICHD’s 2016 success rates for K08 and K23 awards (13.6% and 19.7%, respectively), success rates at other institutes were higher. Specifically, the success rates for K08 and K23 awards were 44.6% and 45.2% for the National Institute of Digestive Diseases and Kidney Disorders, 41.7% and 35.5% for the National Institute of Mental Health, and 25.0% and 40.5% for the National Institute on Drug Abuse.

Another factor affecting the success rate for individual K awards is a change introduced in fiscal year 2017 that requires institutes to contribute a higher amount toward K08 and K23 salaries. This requirement is the result of the conclusions of the 2014 NIH Physician-Scientist Workforce Report,2 which identified substantial salary disparities between clinical and research careers. To encourage more scholars to continue in research, the panel recommended increasing salary support within the K08 and K23 grants to at least $100 000, an increase from $75 000 for the NICHD. However, no additional funds were provided to carry out this requirement. With these increased costs, fewer individual K grants can be supported. Because of the currently low rates of success for individual K08 and K23 awards (Figure 3), there is mounting concern that scholars and early-stage investigators will not receive essential research training and may even be discouraged from continuing in research careers.

The goal of career development and training awards is to provide training and mentoring so that trainees develop the skills to become independent investigators. One important measure of success is becoming a principal investigator on a federally funded research project grant. Other measures of success may include receiving funding from nonfederal entities such as foundations or industry, achieving academic promotion, producing peer-reviewed publications, and serving as a coinvestigator.3 For our analysis, we defined success for physicians who completed career development training programs—through individual K or institutional K12 mechanisms—as the proportion who applied for and received independent NIH research project grant funding. We also identified whether these physicians went on to academic positions. As part of an assessment of its overall training portfolio, the NICHD conducted an analysis of its investments in different training mechanisms. Given the NICHD’s decline in success rates for individual training awards and the current ratio of institutional to individual career training programs, we reviewed the NICHD-supported training pathways to determine how best to support the next generation of physician-scientists.

Methods

This analysis focuses on a physician cohort that included all individuals who were supported under institutional K12 career development awards or who received individual K career development awards (K08, K23, or equivalent) between October 1, 1999, and September 30, 2001. These years were chosen for the analysis so that former scholars would have sufficient time to progress through subsequent stages of career development, apply for research project grants, and establish an academic or clinical career path. From 1999 through 2001, K awardees may have received up to 9 years of career development funding. We chose to limit the analysis to scholars who had a maximum of 6 years of career development support to be consistent with current limits on combined individual and institutional career support. We did not seek institutional review board approval or an exemption because this project did not meet the definition of research per National Institute of Health policy and 45CFR46. Please note that the data were internal to the institute or publicly available and considered to be nonsensitive.

Physician-scholars supported by NICHD career development awards were identified by searching NIH administrative data, as well as progress reports submitted by individual and institutional career development awardees. The NIH administrative data include the IMPAC II system, the NIH’s main administrative system for grants; eCommons, which includes information on individuals associated with NIH grants and review processes; data files specific to training and career development appointments; and data systems developed by the NIH to link funding data with publications. The scholars were divided into groups based on type of career development funding (institutional K12 only, individual K only, or both). When the subsample size was large enough, analyses were conducted separately for physician-scholars with only an MD degree and those with both an MD and a PhD. Primary outcomes were subsequent application and receipt of NIH research project grants (application and funding rates) and category of employment at the time of analyses. Post hoc analyses were conducted to determine whether scholars, sex, type of K12 program (local vs national), or prior postdoctoral training funded by a T32, F32, or F35 award affected subsequent funding rates.

Subsequent NIH grants and grant applications were identified through NIH administrative data, using each person’s unique NIH profile identification number. In addition, each name was separately searched and verified to account for cases in which 1 person might have been assigned multiple identification numbers.

Subsequent professional employment was searched by matching the scholar’s name, university or other affiliation, years of training, area of research, sex, degree, and other characteristics. Data sources included the faculty directory of the Association of American Medical Colleges; university directories and other databases; LinkedIn and other social media websites; publication databases, including Scopus, PubMed, and Web of Science; Open Researcher and Contributor ID profiles; NIH administrative data; and information retrieved from internet searches. Two team members conducted separate searches on the outcomes for each scholar; this information was then verified by a third team member.

Additionally, we collected and analyzed data relating to NICHD training and career development programs. To assess these programs in the context of the overall NICHD budget, we downloaded financial reporting data from the NIH’s historical budget files. Data on the NICHD’s training and career development expenditures, funding rates, and funding mechanisms were calculated from information downloaded from the NIH’s administrative data systems. Information on the specific programs and individual trainees supported was derived from NIH administrative data, as well as from progress reports submitted by individual and institutional career development awardees.

Differences between 3 groups were analyzed via χ2 test Fisher exact test was used for pairwise comparisons. The significance level was set at P < .05.

Results

Of scholars trained between October 1, 1999, and September 30, 2001, 417 physician-scientists were supported by K12, K08, and/or K23 (or their equivalents) awards. Of these, 355 met inclusion criteria, and 62 were excluded because they received more than 6 years of career development support. Of these 62 individuals, 59 received both individual K and institutional K12 funding and 3 received only institutional K12 funding, but from multiple programs. The 355 scholars analyzed included 257 who had solely an MD degree and 98 who had both an MD and a PhD (eTable 1 in the Supplement).

Group differences in research project application rates were found for MD-only scholars but not for scholars with both MD and PhD degrees. Among MD-only scholars, those who received an individual K award or both an individual K award and institutional K12 support were significantly more likely to apply for subsequent research project grant funding compared with individuals who received only K12 support (individual K only vs institutional K12 only: odds ratio [OR], 4.86; 95% CI, 1.83-13.62; both K and K12 vs K12 only: OR, 3.19; 95% CI, 1.46-7.10) (Figure 4). Among scholars with both MD and PhD degrees, application rates did not differ significantly by group (institutional K12 only, 80%; individual K only, 93%; both, 91%; P > .28). Within the K12-only group, a higher percentage of scholars with both MD and PhD degrees than MD-only scholars applied for NIH research grants (80% vs 60%, respectively, OR, 2.72; 95% CI, 1.19-6.38). Within the groups that had an individual K award only or both and individual K award and an institutional K12 award, there were no significant differences between MD-only scholars and scholars with both MD and PhD degrees. Research project grant application rates were also higher for men compared with women among scholars who received only institutional K12 support (73% vs 59%; OR, 2.16; 95% CI, 1.13-4.1) and both institutional K12 and individual K support (96.0% vs 76.0% respectively; OR, 6.95; 95% CI, 1.34-48.16).

As with application rates, group differences in research project grant rates were found for MD-only scholars, but not for scholars with both MD and PhD degrees. In part because they are more likely to apply, MD-only scholars who received an individual K award only or both an individual K award and institutional K12 support were significantly more likely to receive an NIH research project grant than those who only had an institutional K12 award (individual K only vs institutional K12 only: OR, 3.04; 95% CI, 1.66-6.16; both individual K and institutional K12 vs institutional K12 only: OR, 3.19; 95% CI, 1.66-6.16; P < .001) (Figure 4). For physician-scholars with both MD and PhD degrees, application and funding rates between groups were not significantly different (P > .11). Funding rates did not differ between men (55%) and women (43.0%).

We also analyzed a subset of data from the K12-only group to determine if application and funding rates were different for national and local K12 career development programs. Within the K12-only group, 24 physician-scholars were funded by a national K12 program and 157 were funded by a local K12 program. Of note, 13 scholars were excluded because they had participated in both national and local K12 programs. Although research project application and funding rates were greater for local institutional K12 training programs compared with national K12 training programs, the difference was not significant (application rates, 68.1% vs 50.0%; P = .11; and funding rate, 40.8% vs 20.8%; P = .07) (eTable 2 and eFigure 2 in the Supplement).

Approximately 37.2% of physician-scholars included in the analysis had prior postdoctoral training via an individual fellowship (extramural F32 or intramural F35 [n = 18]) and/or institutional training grant (T32 [n = 124]). Within-group comparisons showed that MD-only scholars had higher application rates if they had previous F32 and/or T32 training (80.0% vs 54.9%; OR, 3.29; 95% CI, 1.16-9.79); there was no difference in the funding rate in this group. In addition, there was no difference in application or funding rates within any of the other groups (eTables 3-7 in the Supplement).

Both the individual K and institutional K12 programs produced a cohort of scholars who largely continued in academic careers (Figure 5). Of scholars who received support from an individual K award, institutional K12 award, or both between 1999 and 2001, approximately 75.2% of physician-scholars went on to academic positions. Physician-scholars in the K12-only group were significantly more likely to go on to careers in clinical care, compared with individuals who received both individual K and institutional K12 support (16.8% vs 7.3%; OR, 2.61; 95% CI, 1.05-6.76). In each group, 4.1% to 7.3% of scholars went on to careers in industry. A smaller number worked in government or nonprofit agencies, and a few were deceased or could not be traced. Within and across groups, there were no significant differences between MD-only scholars and scholars with both MD and PhD degrees with respect to positions. There was no difference with respect to positions among scholars with a history of postdoctoral fellowship (F) or institutional training (T32) support.

Discussion

Physician-scholars are more likely to apply for and receive an NIH research grant if they are trained on an individual career development award, with or without an institutional training award. Previous T32 experience may be beneficial for MDs who received institutional (K12s) support but not individual K awards. Although physician-scholars supported by a local K12 career development program appear to be more likely to apply for and receive funding compared with scholars from a national K12 career development program, a large portion (68.0%) of the scholars from national K12 programs were not included in the analysis because they also received a local K12 award, an individual K award, or more than 6 years of career development training. Physician-scholars with both MD and PhD degrees had similar research project grant application and funding rates regardless of the type of training support received (institutional K12 only vs individual K only vs both). These findings suggest that physician-scholars with solely an MD degree may need additional time to become competitive for subsequent research project grant applications; thus, they should use a pathway with both institutional and independent K awards. Scholars with both MD and PhD degrees do not appear to need additional time, presumably because of the additional training they acquire during their PhD project and thesis.

Limitations

Several limitations should be considered when interpreting these results. First, the study cohort included physician-scholars with an MD or equivalent degree who received a maximum of 6 years of career development support (individual, institutional, or both), and were actively receiving that support between fiscal years 1999 and 2001. Therefore, these results are not applicable to individuals who received support before or after that time, to PhD scientists without an MD degree, or to individuals who received more funding than is permitted under current policy (6 years). Considerable changes have taken place both in career development programs and in the biomedical research enterprise since these scholars received training support, and these differences may limit the applicability of past experience to current programs. Second, because of limitations in the available data, not all relevant variables could be examined. Specifically, the analyses do not include data on non-NIH research funding or support on NIH-funded grants in roles other than principal investigator. These analyses also do not account for other relevant factors, such as the quality of mentoring, resources available at the institutional level, and didactic research training outside of PhD programs (for example, MS degrees in clinical research). The limited number of scholars who were followed precluded more sophisticated multivariate analysis; thus, the analyses reported here should be interpreted with caution. Third, efforts were made to collect all available information on each scholar. Follow-up data on positions were found for over more than 96.6% of former scholars. The NIH databases were searched for all scholars, by name and identification number, to identify subsequent grant applications. However, it is possible that some scholars changed their name (eg, on marriage) and may not have been identified by subsequent searches. Finally, the metrics of success for these analyses were subsequent research project grant applications, awards, and professional positions. There are many other markers of success for career development programs that were not evaluated, including funding from nonfederal entities, academic promotion, peer-reviewed publications, and serving as a coinvestigator.3

Conclusions

The NICHD remains firmly committed to training and intends to maintain its investment in training and career development awards at 5% to 6% of its extramural budget. As highlighted in the 21st Century Cures Act,1 supporting the career development of the next generation is essential to the scientific enterprise. The recent review of the NICHD training programs has highlighted a number of opportunities and provided recommendations.4 Given the evidence shown here regarding the importance of the individual career awards, the NICHD intends to shift some of its focus from institutional training awards and provide a greater proportion of its career development allocation to individual awards. However, the NICHD also recognizes the contributions of institutional career development programs, both at the local and national levels, and will continue to offer these mechanisms to support promising physician-scholars who are interested in the NICHD’s research mission.

Back to top
Article Information

Corresponding Author: Diana W. Bianchi, MD, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, 31 Center Dr, Room 2A03, Bethesda, MD 20892 (diana.bianchi@nih.gov).

Accepted for Publication: September 21, 2017.

Published Online: January 16, 2018. doi:10.1001/jamapediatrics.2017.4305

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

Concept and design: Glavin, Taymans, Spong, Bianchi.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Twombly, Glavin, Spong, Bianchi.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Glavin, Guimond, Spong.

Administrative, technical, or material support: Twombly, Glavin, Taymans, Spong.

Supervision: Twombly, Glavin, Spong, Bianchi.

Conflict of Interest Disclosures: None reported.

Additional Contributions: We gratefully acknowledge the contributions of the following members of the Training Task Force: Sherry L. Dupere, PhD, Melissa Gilliam, MD, MPH, Lisa Halvorson, MD, A. Tyl Hewitt, PhD, Charisee Lamar, PhD, Ruth Lehmann, PhD, Ming Lei, PhD, Ralph Nitkin, PhD, Stephen A. Petrill, PhD, Christine M. Rogers, MA, George R. Saade, MD, and Jennifer Sutton, PhD. Paul Williams and Lisa Kaeser, JD both from the National Institute of Child Health and Human Development participated in the task force and provided helpful review of the manuscript. They received no compensation for these contributions.

Additional Information: This study was performed as part of official duty work of the authors, who are all employed by the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

References
1.
 Century Cures Act. Pub L No. 114-255, 130 Stat 1033-1344 (2016). https://www.congress.gov/bill/114th-congress/house-bill/34/text#toc-H3ED134CB78594644AE2028D8BCE93275. Accessed October 19, 2017.
2.
National Institutes of Health.  Physician Scientist Workforce Report. https://report.nih.gov/workforce/psw/, Published 2014. Accessed June 15, 2017.
3.
Marsh  JD, Todd  RF  III.  Training and sustaining physician scientists: what is success?  Am J Med. 2015;128(4):431-436.PubMedGoogle ScholarCrossref
4.
National Institute of Child Health and Human Development.  Review of NICHD Training and Career Development Programs. https://www.nichd.nih.gov/training/extramural/Documents/NICHD_training_review.pdf. Accessed June 15, 2017.
×