T126. PSYCHIATRIC PREDICTORS FOR BECOMING HOMELESS AND EXITING HOMELESSNESS: A SYSTEMATIC REVIEW AND META-ANALYSIS

BACKGROUND: Homelessness is an increasing societal problem in high-income countries and often linked to psychiatric disorders. However, a study compiling the existing literature is lacking. The aim was to identify individual-level predictors for becoming homeless and exiting homelessness in a systematic review and meta-analysis. METHODS: We searched PubMed, EMBASE, PsycINFO, and Web of Science (up to January 2018). Becoming homeless and exiting homelessness were the outcomes. Observational studies with comparison groups from high-income countries were included. The Newcastle Ottawa Quality Assessment Scale was used for bias assessment. Random effects models were used to calculate pooled odds ratios (ORs). In all, 116 studies of predictors for becoming homeless and 18 for exiting homelessness were included. RESULTS: Psychiatric problems, especially drug use problems (OR 2.9, 95% confidence interval (CI) 1.5–5.1) and suicide attempts (OR 3.6, 95% CI 2.1–6.3) were associated with increased risk of homelessness. However, the heterogeneity was substantial in most analyses (I2>90%), and the estimates should be interpreted cautiously. Adverse life-events, including childhood abuse and foster care experiences, and past incarceration were also important predictors of homelessness. Psychotic problems (95% CI 0.4, 0.2–0.8; I2=0) and drug use problems (OR 0.7, 95% CI 0.6–0.9; I=0) reduced the chances for exiting homelessness. Female sex and having a partner increased the changes of exiting homelessness. DISCUSSION: Evidence for several psychiatric predictors for becoming homeless and exiting homelessness was identified. Additionally, socio-demographic factors, adverse life-events, and criminal behavior were important factors. There is a need for more focus on psychiatric vulnerabilities and early intervention to reduce the risk of homelessness.