Descriptive epidemiology of homelessness among veterans in the United States: Methodology and risk assessment

Homelessness within the US veteran population remains a significant public health and social justice problem. A major obstacle to progress in this area is the lack of comprehensive estimates of the prevalence of homeless veterans. Due to unique methodological advantages and limitations inherent in prior research, sometimes widely varying estimates of veteran homelessness have been obtained (range: 10%-23%). The purpose of the present study was to use a multi-faceted approach to capture the advantages, but avoid limitations, of previous efforts in order to produce more definitive estimates of the prevalence of 1) US veterans within the homelessness population and 2) homelessness among US veterans. To this end, we combined data from several administrative sources including 10 homelessness-reporting jurisdictions throughout the US, the Department of Veteran Affairs Grants and Per Diem program, and the US Census Bureau to obtain estimates of veteran homelessness. We report and describe 1) the methodology used to derive estimates of veteran homelessness prevalence (eg, merging data sources, imputing and extrapolating data, minimizing presence of double counts), 2) prevalence estimates of veteran homelessness for substratifications of age, gender, race, and poverty-level, and 3) estimates of relative risk for homelessness among veterans for fine-grained, demographic substrata. Results represent more precise estimates of veteran status within the homelessness population than previously obtained. Findings from the present study will inform policy decisions regarding targeting of resources for the prevention and elimination of homelessness among various groups of veterans in specific geographic areas.