Routine patient reported outcomes as predictors of psychiatric rehospitalization

Objective: Patient reported outcome measures (PROMs) are increasingly used to measure psychiatric service consumers' progress and to provide feedback to consumers and providers. We tested whether PROMs can predict and be used to identify groups at high risk for future hospitalization. Methods: A total of 2842 Israeli users of psychiatric rehabilitation services reported on their quality of life (QoL) and the effect of symptoms on their daily functioning. Survey data were linked with information on psychiatric hospitalization 6 and 12 months after survey completion. Variables associated with each of the outcomes were tested for significance and entered into a multivariate logistic regression model. Prediction scores were developed to identify the highest-risk groups according to each model. Results: QoL was found to be a significant predictor of future hospitalization within 6 months (odds ratio [OR] = 0.71, 95% CI: 059-0.86), and self-report of the impact of symptoms on functioning significantly predicted 12 month hospitalization (OR = 0.83, 95% CI: 0.74-0.93), controlling for known risk factors. Positive predictive values for the 6- and 12-month risk scores were 31.1 and 40.4, respectively, for the 10% highest risk categories. Conclusions: Reports of psychiatric service consumers on their QoL and on the effect of symptoms on their functioning significantly predict of future hospitalization risk beyond other well-known risk factors. PROMs can identify consumers at high risk for future hospitalization and thus direct interventions for those at highest risk. (C) 2017 Elsevier B.V. All rights reserved.