A Longitudinal Study of Suicidality in a Homeless Population Sample

Little research has been conducted on the timing of the onset and course of suicidality relative to the timing of the onset and temporal status of homelessness. Therefore, this longitudinal study investigated suicidal ideation and plans and suicide attempts in a homeless population in relation to housing attainment, psychiatric disorders, and substance use/disorders.

Prospective longitudinal follow-up data were collected from a representative sample of literally homeless adults in St. Louis (N = 255) using the Diagnostic Interview Schedule/Homeless Supplement, the Composite International Diagnostic Interview-Substance Abuse Module, and urine drug screens. Associations among suicidal symptom variables, housing status, psychiatric/substance use disorders, and substance use were examined at baseline and longitudinally.

Lifetime prevalence of suicidal ideation/plan and suicide attempts in this homeless population were much higher than in the general population. Onset of suicidal symptoms had typically preceded onset of homelessness. Few individuals experienced suicidal ideation/plans or attempted suicide during this study, and even fewer experienced new suicidal symptoms after baseline. Securing stable housing during the study follow-up was associated with lower rates of suicidal ideation/plans.

This study's findings contradict assumptions that the high prevalence of suicidal symptoms in homeless populations can be explained by the difficulties and miseries of homelessness. Psychiatric illness, substance abuse, and psychosocial factors associated with homelessness may be direct contributors to suicidal symptoms and thus represent strategic intervention targets.HIGHLIGHTSMost suicidality reported at baseline first emerged long before first homelessnessAfter baseline, few reported new suicidal symptoms or had active suicidal symptomsSignificantly fewer reports of suicidal ideation/plans over time were found in those with the most stable housing outcome.