Predictors of Housing Dissatisfaction Among Persons with Serious Mental Illness (SMI) Enrolled in a Primary and Behavioral Health Care Integration (PBHCI) Program

Background: Individuals with SMI are disproportionately impacted by homelessness and housing instability. This disparity is currently being addressed by integrated primary and behavioral health care services and psychosocial supports in the community mental health (CMH) setting. CMH centers provide whole-person care and are well-positioned to meet the needs of individuals with SMI who often have comorbid conditions, housing insecurity, and other barriers to optimal health. While variables such as socioeconomic status and access to care have been studied in detail in this population, little attention has been dedicated to housing satisfaction and its potential correlation to psychosocial risk factors. The present study examines predictors of housing dissatisfaction among clients with SMI enrolled in a PBHCI program in a CMH setting.

Methods: Employing SAMHSA’s PBHCI National Outcomes Measures dataset, this study examined the prevalence of housing instability and the predictors of housing dissatisfaction among individuals with SMI at program enrollment (N=929). Bivariate statistics were calculated to determine correlates of housing dissatisfaction for inclusion in the logistic regression model. Predictors assessed included demographic and psychosocial characteristics including measures of psychological distress, daily functioning, and social support within the last 30 days. Quality of life (QOL) was assessed with the question, “In the last four weeks, how would you rate your quality of life?” Dissatisfaction with housing was dichotomized (0=no or neutral, 1=yes).

Results: The study sample primarily was female (64%), African American (44%) and identified as Hispanic/Latino (61%). In the past 30 days, about 73% of consumers lived in housing they owned or rented, while 19% lived in someone else’s residence, 4% were homeless, and 4% were hospitalized or living in transitional facilities. About 24% stated that they were dissatisfied with the conditions of their living space. The LR indicated that the overall model correctly classified 78.9% of cases and it was statistically reliable in distinguishing between consumers who were and were not dissatisfied with their housing (-2 Log Likelihood=252.5;χ2(9)=38.23, p<.000). Statistically significant predictors accounted for approximately 20% of the variance in retention. Among predictors, QOL and social connectedness significantly predicted dissatisfaction with housing. For every unit QOL and social connectedness increased, the odds of feeling dissatisfied with housing decreased by 35% and 15%, respectively. Gender, race, Hispanic/Latino ethnicity, sexual orientation, daily functioning and psychological distress did not significantly predict housing dissatisfaction.

Implications: Findings highlight the need for targeted outreach and assessment about consumer’s living conditions in PBHCI programs and its potential correlates with psychological and social functioning. Providers should pay particular attention to the experiences of clients who are dissatisfied with their housing arrangements to better understand the relationship between housing conditions, quality of life and social connections. Future research using longitudinal design is needed to better clarify the direction of these relationships.