S26. PSYCHOSIS-SPECTRUM SYMPTOMS AMONG ADOLESCENTS PSYCHIATRICALLY HOSPITALIZED FOR ACUTE SAFETY CONCERNS ARE LINKED TO READMISSION WITHIN SIX MONTHS OF DISCHARGE

BACKGROUND: The emergence of psychosis during adolescence is linked to particularly poor long-term outcomes. Teens at clinical high risk for psychosis (i.e. with attenuated positive symptoms) often experience other psychiatric symptoms that motivate help-seeking behavior. This suggests that inpatient settings may be key points of contact for identification of at-risk youth. Research exploring the utility of psychosis risk screening in pediatric inpatient settings in the U.S. is relatively unexplored. Given evidence of high rates of acute care use in early psychosis, research exploring the occurrence of psychosis-risk symptoms in pediatric inpatient settings and the implications for clinical outcomes may inform risk screening practices. METHODS: Participants were 656 adolescents (ages 11–18) hospitalized on a psychiatric inpatient unit in the Northeast United States. This IRB-approved retrospective chart review explores the associations between psychosis-spectrum experiences and subsequent re-hospitalization among teens with acute safety concerns. All adolescents admitted to the unit were asked to complete a psychosis-risk screening measure, the PRIME Screen-Revised, and a brief psychiatric interview, the Children’s Interview for Psychiatric Syndromes (ChIPS), within 72 hours of admission. Demographic information (i.e. age, race, ethnicity, sex) was extracted from medical charts. Adolescents with primary chart diagnoses including psychosis were excluded from the sample. Responses on the PRIME were explored in relation to rehospitalization during a 6-month follow-up period. RESULTS: Rates of rehospitalization within 6 months after discharge were higher for teens who screened positive on the PRIME (n = 246) compared to those who did not (Χ2 = 4.27, p < .05; 20% versus 14%). Item-level analyses demonstrated that endorsement of two out of the twelve PRIME items had small, significant correlations with 6-month rehospitalization: 1) something may be interrupting or controlling me (r = .08), and 2) my mind is playing tricks on me (r = .12). Age, but no other demographic variable, was significantly correlated with rehospitalization (r = -.11). Logistic regression results indicated that rehospitalization was significantly predicted by endorsement of “mind tricks” (B = .64, p < .05; OR = 1.89), but not the “interruption/control” item, when controlling for age (B = -.13, p < .05; OR = 0.88). “Mind tricks” remained a significant predictor of rehospitalization when controlling for age plus the non-significant effects of diagnostic variables correlated with rehospitalization (i.e. PTSD and behavioral disorder diagnosis). DISCUSSION: Findings indicate that screening for psychosis-spectrum experiences in acute care settings may be helpful for identifying teens at risk for repeat use of high-level care. Moreover, specific types of psychosis-risk symptoms, such as experiencing mind tricks, may be indicative of higher risk for readmission in the months following hospital discharge. Results: suggest that screening and assessment for psychosis-risk symptoms in pediatric psychiatric settings may be important for initiating early psychosis intervention and mitigating future risk for hospitalization.