To the Editor: Existing data suggest that suicide rates increase during middle age and again among older adults in the general population (
1). Previous research has documented more than a fourfold increase in risk for all-cause and cause-specific mortality associated with homelessness (
2). However, less is known about the relationships between suicide and age among those with history of homelessness. In recent years, identifying risk factors and intervention strategies to eliminate homelessness among veterans has become a national priority (
3). Understanding age-related variability among veterans who have died from suicide will increase understanding of differences in mortality associated with homelessness and opportunities for prevention.
Using data from the 2000–2008 fiscal years, we identified age-based differences in suicide among veterans with and without a history of homelessness. Deaths from suicide were identified by using data from Veterans Health Administration (VHA) administrative records linked with data on cause of death from the National Death Index (
4). History of homelessness was defined as one or more intake assessments conducted by VHA case managers once a veteran was identified as homeless or at risk of homelessness. During the project period, a total of 15,535 suicides occurred and 238,266 unique VHA patients were assessed for homeless services. Among the veterans who died by suicide, 561 (3.6%) had a history of homelessness.
The distribution of suicides among those without a history of homelessness was bimodal, with higher prevalence during middle age and again among older adults. In contrast, suicides among homeless veterans had a single peak during middle age. [Figures illustrating these patterns are available online as a
data supplement to this letter.] Comparisons of average age at time of death confirmed these differences. The difference in average age at death was statistically significant: 48.5 years among veterans with a history of homelessness and 60.8 years among those without a history of homelessness (t=31.61, df=701, p<.001).
Results from this project support recommendations for routine assessment of housing status and awareness of age-related differences in mortality associated with homelessness. It is unclear whether the observed patterns are related to population differences in suicide risk, variability in assessment and service use, cohort-specific exposures, or changes in housing associated with increasing age. Additional research on the relationships between homelessness, suicide risk, and age is recommended.
Acknowledgments and disclosures
The authors report no competing interests.