Identifying Some Risk Factors of Time to Relapses in Schizophrenic Patients using Bayesian Approach with Event-Dependent Frailty Model.

OBJECTIVES Schizophrenia patients often experience relapses once and even more with no limit on number of relapses. The time among relapses are rarely considered in studies. The aim of this study was to identify some risk factors of time to elapses in schizophrenic patients with recurrent events model in survival analysis. METHODS In a retrospective longitudinal study, the medical records of 159 schizophrenic patients who referred to Razi hospital in Tehran from 2003 to 2005 were conveniently sampled, investigated and followed until the end of 2009. The time to recurrent relapses were considered in weeks. The patients with at least one relapse in this duration were included in the study. Event-dependent frailty model, using Bayesian approach, was applied to fit the data and identify the risk factors of time to relapses. RESULTS In this recurrent failure time model, the effects of age of onset (95% CI = (0.058, 0.086)), gender (95% CI = (0.146, 0.686)), marital status (95% CI = (0.475, 0.965)) and family history (95% CI = (0.115, 0.543)) were significant on the hazard time to relapses. According to the credible interval of frailty variance, elapsed time to relapse is dependent on patients' characteristics (95% CI = (0.084, 0.369)). Subsequent relapses are likely to be influenced by the occurrence of the first relapse, too (95% CI = (2.504, 3.079)), with decreasing hazard of time to relapse. CONCLUSIONS Subsequent relapses are likely dependent on the first and previous relapses. Age of onset, gender, marital status and family history are important risk factors influencing hazard of time to relapses. More studies are required to clear out the effect of other covariates with this model.