Career development awards are associated with lower career-long research attrition risk in Veterans Affairs investigators.
- 1Office of Research & Development, US Department of Veterans Affairs
Abstract
Objective:To investigate research attrition and the associated factors among US Department of Veterans Affairs investigators with and without a prior Career Development Award (CDA) from fiscal year 2008 to 2026. Design: Retrospective cohort study. Methods: Investigators whose CDA ended between October 2008 and March 2026 were identified from the Research Analysis and Forecasting Tool (RAFT) database and compared to non-CDA peers. Extracted descriptive variables included sex, investigator type, doctoral degree, salary source, specialty, and research area. Attrition, derived from the electronic Project Management and Information System, was defined when an investigator ceased serving as a principal or coinvestigator on VA-funded research. Analysis utilized attrition percentages, crude count ratios (CR), chi-squared statistics, and multivariable binomial regression to assess adjusted odds ratios (ORadj). Results: Investigators with a prior CDA were 42% less likely to attrite from VA research engagement compared to non-CDA counterparts (CR 95% CI, 0.50-0.67, p<.001). They maintained high engagement regardless of demographic or professional variables (engagement ranges: 66.7%-93.8%). Clinician-scientists were 34% less likely to attrite from science (ORadj 0.54) compared to non-clinician scientists. The 1.68-times higher crude attrition risk for non-VA salaried investigators became non-significant after adjusting for other factors. Attrition was significantly higher in basic science and engineering (ORadj 3.90), endocrinology, metabolism, and nutrition (ORadj 4.24), infectious disease (ORadj 3.80), aging and geriatrics (ORadj 3.56), and cardiopulmonary and vascular research (ORadj 2.70) compared to health services and epidemiology. Conclusion: Prior CDA recipients demonstrate significantly lower research attrition in VA research.
Related
- Distribution of industry-sponsored research payments to US physician principal investigators
- Assessing methods to identify Veterans Health Administration populations with Veterans Justice Programs contact to improve health research
- Development of a tool to measure patient-oriented outcomes for Veterans with inflammatory arthritis initiating self-administered biologic therapies.
- Factors associated with academic-medicine faculty attrition before promotion to senior rank: A national-cohort study
- Narrowing the research to practice gap for trauma-focused psychotherapies in the Veterans Health Administration: A narrative review
BibTeX
@misc{fraser2026career,
title = {{Career development awards are associated with lower career-long research attrition risk in Veterans Affairs investigators.}},
author = {J. J. Fraser and S. Hack},
howpublished = {medRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.03.26367173},
url = {https://www.medrxiv.org/content/10.64898/2026.11.03.26367173v1},
note = {CC0},
}
RIS
TY - UNPB
TI - Career development awards are associated with lower career-long research attrition risk in Veterans Affairs investigators.
AU - J. J. Fraser
AU - S. Hack
PY - 2026
DA - 2026/11/03
PB - medRxiv
DO - 10.64898/2026.11.03.26367173
UR - https://www.medrxiv.org/content/10.64898/2026.11.03.26367173v1
AB - Objective:To investigate research attrition and the associated factors among US Department of Veterans Affairs investigators with and without a prior Career Development Award (CDA) from fiscal year 2008 to 2026. Design: Retrospective cohort study. Methods: Investigators whose CDA ended between October 2008 and March 2026 were identified from the Research Analysis and Forecasting Tool (RAFT) database and compared to non-CDA peers. Extracted descriptive variables included sex, investigator type, doctoral degree, salary source, specialty, and research area. Attrition, derived from the electronic Project Management and Information System, was defined when an investigator ceased serving as a principal or coinvestigator on VA-funded research. Analysis utilized attrition percentages, crude count ratios (CR), chi-squared statistics, and multivariable binomial regression to assess adjusted odds ratios (ORadj). Results: Investigators with a prior CDA were 42% less likely to attrite from VA research engagement compared to non-CDA counterparts (CR 95% CI, 0.50-0.67, p<.001). They maintained high engagement regardless of demographic or professional variables (engagement ranges: 66.7%-93.8%). Clinician-scientists were 34% less likely to attrite from science (ORadj 0.54) compared to non-clinician scientists. The 1.68-times higher crude attrition risk for non-VA salaried investigators became non-significant after adjusting for other factors. Attrition was significantly higher in basic science and engineering (ORadj 3.90), endocrinology, metabolism, and nutrition (ORadj 4.24), infectious disease (ORadj 3.80), aging and geriatrics (ORadj 3.56), and cardiopulmonary and vascular research (ORadj 2.70) compared to health services and epidemiology. Conclusion: Prior CDA recipients demonstrate significantly lower research attrition in VA research.
ER -