T51. SEMANTIC FLUENCY PERFORMANCE IS PREDICTIVE OF RE-HOSPITALIZATION IN PATIENTS WITH RECENT ONSET PSYCHOSIS

Abstract

Background: The presentation of psychotic symptoms among young adults often indicates the onset of a chronic mental illness, which frequently requires multiple hospitalizations. Patients with serious mental illnesses involving psychosis tend to present with global impairment of neurocognitive functioning (NCF), which is associated with functional impairment. However, it is unclear whether measures of NCF are predictive of rehospitalization early in the course of illness. This study presents preliminary data regarding NCF impairment in patients with a recent onset of psychosis and relationships between NCF and rehospitalization. Method(s): Retrospective review identified 28 patients admitted within 5 years of initial onset of psychosis (M age=24.5) who completed neuropsychological screening as part of the Early Onset Treatment Program (EOTP), an inpatient program for serious mental illness. Patients were administered the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), which involves 12 subtests, 5 domain index scores, and 1 global index. Age-adjusted standard scores were converted to z-scores (M=0, SD=1) to facilitate interpretation. Z-scores at or below -1.5 were considered impaired. The number of rehospitalizations at 1-year follow-up after discharge was tabulated. Associations between RBANS z-scores and rehospitalization rates were determined with Pearson product-moment correlations. RBANS subtests significantly associated with rehospitalization were included as predictors of rehospitalization in follow-up stepwise linear regression analyses. Result(s): Most patients were diagnosed with a schizophrenia-spectrum disorder (57% schizophrenia, 29% schizoaffective) and over 60% were hospitalized for the first time. A majority (71%) showed global impairment on the RBANS Total score with a median of 2 indices impaired. All patients exhibited impairment on at least 1 individual subtest measure with a median of 8 impaired. Impairment was most frequent and severe across measures of delayed memory [List Recall: 82% impaired; M=-2.21 SD=0.70] and verbal fluency [Semantic Fluency: 79% impaired; M=-1.94, SD=0.90]. Only 25% of patients were rehospitalized in the follow-up period (median= 1 rehospitalization). Rehospitalization was significantly associated with measures of verbal fluency [Semantic Fluency: r(26)=.67, p < .001]. Rehospitalization was not associated with specific diagnosis. Discussion(s): Consistent with prior investigations, NCF impairment in patients presenting with psychosis is ubiquitous and diffuse. Close inspection of patient neuropsychological profiles suggests that aspects of memory and language may be particularly impacted. Although such NCF deficits have been associated with poorer functional outcomes, these preliminary results suggest that greater NCF impairment is not predictive of rehospitalization. Semantic fluency accounted for 42% of the variance in readmittance, suggesting that better language functioning is significantly predictive of being rehospitalized within 1 year of discharge from early intervention. It is possible that patients with relatively preserved language abilities exhibit more positive symptoms and are better able to vocalize distress, ultimately facilitating their presentation to medical attention. Further investigation appears warranted.