Background Eating disorders (EDs) likely affect a significant proportion of female and male Veterans; however, they remain understudied and under-treated in VA. Individuals with EDs are likely to be high users of medical care, which is at least partly due to the physical complications of EDs. Nonetheless, EDs often go undetected in general medical settings, unless providers screen for them. Currently, the VA does not routinely screen for EDs; thus, enhancing VA's capacity nationally to provide comprehensive care for Veterans with EDs is a priority area of focus. This proposal will address four major gaps: the lack of prevalence estimates of EDs in nationally representative samples of Veterans, the lack of knowledge of healthcare needs and preferences of Veterans with EDs, the paucity of findings regarding military/Veteran-specific risk factors for EDs, the lack of a validated screening measure for EDs in Veterans, and the lack of knowledge regarding Veteran preferences for ED screening. These gaps are barriers to screening for and treating Veterans with EDs in VA healthcare. Objectives The specific aims are to establish a nationally representative cohort of male and female Veterans in order to: 1) a) examine the prevalence of male and female Veterans reporting full and subthreshold EDs and b) to examine the potential prevalence of EDs among vulnerable subgroups of Veterans; 2) to identify service use and needs among male and female Veterans with full and subthreshold EDs; 3) to estimate models of risk and protective factors for EDs, including military sexual trauma, combat trauma, and unit and family deployment support; and 4) to validate an existing screening measure of EDs in a Veteran sample and assess screening preferences. Methods In order to achieve these aims, we will recruit a nationally representative sample of 1500 male and female Veterans. We will request that the VA/DoD Identity Repository (VADIR), a VA office with access to Department of Defense records, assist in randomly selecting 3750 individuals from the national Veteran population. We will stratify the sample based on race/ethnicity and oversample women to achieve a 1:1 ratio. We will mail a comprehensive survey to potential participants using a multi-pronged, six-stage mailing strategy. Participants will complete survey measures of ED symptoms, risk and protective factors, and VA and non-VA healthcare service and needs. Aim 1: sample weighting procedures will be used in order to calculate nationally representative prevalence estimates of EDs among male and female Veterans. Aim 2: Poisson regression will be used to compare healthcare service use for Veterans with EDs to Veterans without EDs. In addition, we will use descriptive statistics to report service use among male and female Veterans with full and subthreshold EDs and which services, if available, they would like to use. Aim 3: structural equation modeling will be used to estimate models of risk and protective factors for EDs. Aim 4: Receiver Operating Characteristic curves will be used to estimate sensitivity and specificity of the ED screening measure, regression models will be used to investigate criterion validity, and descriptive statistics will be used to describe Veteran screening preferences. Dissemination Plan We will share our findings with our Operational Partners on monthly workgroup calls and make recommendations based on these results. Our findings regarding the scope of EDs among Veterans and their healthcare use and preferences will be used to inform ongoing provider trainings in EDs. We also will disseminate findings during Women's Health Cyberseminars and via newsletters such as the Women's Health Roundup. We will develop clinician and executive briefs in order to disseminate findings to both clinicians and policy-makers. Thus, our results will be summarized and disseminated to VA leadership in a timely manner. !
NIH Reporter Project Information
None at this time.
Health Systems, Mental, Cognitive and Behavioral Disorders
Epidemiology, TRL - Basic/Foundational
Addictive Disorders, Gender Differences, Utilization
None at this time.