The suicide rate among female Veterans increased 63% between 2000 and 2014 - significantly higher than the 30% increase observed among male Veterans during this period. Despite a large volume of work examining risk factors, barriers to care, and care utilization among Veterans, little research has examined these issues as they relate to females. Available research has been limited by small female sample sizes, cross-sectional analysis, and other methodological limitations. As such, we know surprisingly little about the health and psychosocial factors, barriers to care, and healthcare utilization patterns associated with suicidal behaviors among female Veterans. Data on female risk for suicide and their healthcare utilization is needed to direct valuable suicide prevention resources and help VHA address this growing concern.
In a large, national sample of female and male Veterans with recent non-fatal suicidal self-directed violence (SSV: fatal and non-fatal suicide attempts), we aim to: 1) Develop and test explanatory models of female and male risk for repeat SSV over 12 months, and 2) Identify similarities and differences in patterns of healthcare utilization, coping strategies, and symptom change over time between female and male Veterans at risk for SSV.
This is a mixed-methods, longitudinal cohort study of Veterans with a history of SSV, guided by a public health, social-ecological framework to facilitate examination of the range of proximal and distal risks for SSV. First, we will identify and enroll 30 female and 30 male Veterans for whom a non-fatal SSV event was recorded in a suicide behavior report in VA's Corporate Data Warehouse (CDW). These 60 Veterans will participate in qualitative interviews to gather data on Veterans' perspectives and experiences with suicidal thoughts and SSV, their recovery needs and experiences with the recovery process, barriers and facilitators to care, and how clinicians and the healthcare system could better identify and address the needs of Veterans like them. We will use a grounded theory approach to analyze transcripts to develop a theoretical model of risk for SSV among female Veterans, directly informing survey construct selection, quantitative analysis plans, and interpretation of quantitative findings. For the longitudinal survey, we will use suicide behavior reports in CDW to identify and enroll at least 480 female and 480 male Veterans, who will complete health and psychosocial measures at baseline, 6- and 12-month follow up. Self-report questionnaires will be informed by the qualitative findings and include psychosocial and health-related measures such as coping efficacy, interpersonal conflict, positive relations with others, trauma, occupational problems, barriers to care, and mental health symptoms. Participants will be followed for 12 months to assess and document all SSV events (primary outcome: an SSV event following baseline), which will be ascertained via multiple sources. Health and utilization data will be obtained from CDW and medical record progress notes. Main quantitative analyses will use a latent variable modeling framework to simultaneously model males and females in a multi-group format to test models of female and male Veteran suicide risk. We will then use latent class and latent transition analysis to identify differential responses to healthcare utilization and how certain health and psychosocial variables cluster together by gender.
None to date.
Findings from this study will provide previously unavailable evidence to support the selection of intervention targets as well as identify high-priority services and barriers to care to direct programing and research priorities for female Veterans at risk for SSV. This work has the potential to also benefit the broader population of female Veterans with mental health conditions or other risk factors for SSV.
HSR&D or QUERI Articles
- Hoffmire CA, Denneson LM. Concerning Trends in Suicide Among Women Veterans Point to Need for More Research on Tailored Interventions. HSR&D FORUM: Research Highlights. 2018 May 1; Spring(2018).