SU60. Risk and Protective Factors Associated With Psychotic Symptom Profiles of Marginally Housed Adults

Abstract

Background: This study examined the characteristics and associated risk and protective factors of distinct psychotic symptom profles exhibited by marginally housed adults. Method(s): The Hotel Study is a longitudinal observational study of adults living in marginalized housing. Five psychosis symptoms (delusions, conceptual disorganization, hallucinations, suspiciousness, and unusual thought content) from the Positive and Negative Syndrome Scale (PANSS) were assessed monthly. Sociodemographic, psychiatric, medical, developmental, social, and substance use factors were also assessed. Two-step cluster analysis was employed to identify groups of people that shared similar symptom profles at the time of their maximum (MaxTime) and minimum (MinTime) total symptom severity (PANSS) in 1 year. Multinomial logistic regression analysis was used to estimate the associations between factors and cluster membership. Paired Wilcoxon and McNemar?s tests were used to compare substance use at MaxTime and MinTime within each cluster. Result(s): In the frst year of study, 404 participants had at least three 5-item PANSS assessments. Cluster analysis of the PANSS scores identifed 3 clusters at MaxTime. The Severe Cluster (n = 74) endorsed severe psychosis symptoms, while the Variable Cluster (n = 147) endorsed supra-threshold delusions and hallucinations only. The Low Cluster (n = 183) did not endorse psychosis at MaxTime. Variable Cluster membership was associated with methamphetamine (OR, 95% CI: 2.28, 1.26?4.10) and cannabis use in the past week (OR, 95% CI: 2.69, 1.39?5.24), a history of traumatic brain injury (OR, 95% CI: 3.14, 1.41?7.01), and low social support at study entry (OR, 95% CI: 0.79, 0.63?0.99). The Variable Cluster experienced frequent transitions between psychotic and nonpsychotic states (median, IQR: 3, 1?5; P < .001), possibly exacerbated by methamphetamine (X2 = 6.86; P = .009) and alcohol use (X2 = 7.90; P = .005). Severe Cluster membership was associated with antipsychotic treatment (OR, 95% CI: 7.34, 3.21?16.81), methamphetamine (OR, 95% CI: 2.84, 1.36?5.90) and cannabis use in the past week (OR, 95% CI: 2.69, 1.39?5.24), and low social support at study entry (OR, 95% CI: 0.73, 0.53?0.99). The Severe Cluster had high rates of primary psychosis diagnosis (P < .001) and poorer psychosocial functioning (P < .001) than the Low Cluster. Symptoms may be exacerbated by recent cannabis (X2 = 5.06; P = .024), opioid (X2 = 4.00; P = .046), or alcohol use (X2 = 4.65; P = .031), but may be unaffected by methamphetamine (X2 = 0.44; P = .505) and antipsychotic use (X2 = 0.00; P = 1.000) in the Severe Cluster. Conclusion(s): A subset of marginally housed adults living with complex multimorbid illness experience severe psychotic symptoms that may be unresponsive to both antipsychotic treatment and methamphetamine use. These individuals may need alternative or additional forms of mental health care and rehabilitation support.