Association Between Use of Services To Address Adverse Social Determinants of Health and Documented Suicide Attempt Among Patients in the Veterans Health Administration

Le résumé fourni par la source

Suicide prevention is a top priority for the US Department of Veterans Affairs (VA), and suicide is often associated with adverse social factors (e.g., financial, legal, and housing problems). The VA provides social services integrated with healthcare services, which may increase the opportunities to detect and document suicide attempt in EHR records. Using VA administrative data, we examined three cohorts of all patients from 2014 to 2018 who had housing instability (n = 659,987), justice involvement (n = 200,487), and unemployment (n = 346,556). Administrative records were used to determine ordinal indicators of receipt of VA social services (no services, low, or high). The outcome was suicide attempt noted in the healthcare record (i.e., documented suicide attempt) in the 1-6 months following the incident adverse social factor. We conducted logistic regressions utilizing a discrete-time survival framework with person-month as the unit of analysis, which facilitated accounting for covariates while isolating the independent association of social service utilization. After adjusting for covariates, high receipt of housing services (vs. no services) was significantly associated with documented suicide attempt during the 6-month observation period (aOR = 1.14, 95%CI = 1.06-1.22). A similar association was observed for high vs. no use of justice programs (aOR 1.24; 95% CI:1.12-1.37). There was no significant association between employment services utilization and documented suicide attempt during the 6-month observation period. Our finding that utilization of social services as positively associated with documented suicide attempt likely reflects increased suicide attempt surveillance and documentation with social service involvement. Future research should explore operationalizing patient-level distress in administrative data.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.