Associations between Homelessness and Psychiatric Symptom Severity and How Homelessness Interacts with Risk Factors among First-Admission Psychiatric Patients

Individuals who experience homelessness are exposed to stressors that have the potential to precipitate mental and physical health complications that can present serious threats to multiple areas of these individuals' lives. Although the path to becoming and remaining homeless is different for every individual, there may be some commonality specific to homelessness that makes certain individuals who experience homelessness more vulnerable than other individuals to the expression and potentially devastating course of psychopathology. Guided by the diathesis-stress model, the intent of this study is to explore the question, following a first hospitalization for mental illness ("schizophrenia spectrum disorder" or "other diagnosis"), what are the associations of homelessness with psychiatric symptom severity and how do risk factors of family history of mental illness, low intelligence, housing independence, hopelessness, and lack of social support interact with homelessness? To answer this question, a secondary analysis was performed on data from the Suffolk County Mental Health Project, a longitudinal study that sought to investigate the course of schizophrenia in an epidemiologic sample of first-admission patients. A sub-sample of 548 cases taken from the parent study were further subdivided into two cohorts; individuals diagnosed with schizophrenia spectrum disorder and those diagnosed with non-schizophrenia mental illness. A multilevel analysis was conducted for four different outcome variables, measuring psychotic, disorganized, and depressive symptoms as well as global functioning, in order to determine whether or not there is a significant association between homelessness and the severity of psychiatric symptomatology. Through the continued use of multilevel analytic models and interaction plots, significant confounding variables were examined to determine whether or not they facilitated interactions with homelessness. Homelessness was found to produce a small, but significant effect, for both cohorts across all four outcome variables. Additionally, applied to different outcome variables, the five confounding variables were also found to have small-to-medium-sized, significant interactions with homelessness. The results of this study provide additional statistical support to other studies looking at similar populations.