Relapse in schizophrenia: The role of factors other than non-adherence to treatment.

Abstract

Relapse rates are very high in schizophrenia. However, little is known about the predictors of the time to relapse other than treatment non-adherence. We investigated possible risk factors for the time to relapse in patients with first-episode schizophrenia (n = 107) who received assured treatment by way of long-acting injectable antipsychotic over 24 months and who underwent regular clinical, cognitive, and metabolic assessments. Using Cox regression analyses we assessed selected premorbid and baseline potential predictors of time to relapse. Relapse was defined using operationally defined relapse criteria. In the primary analysis only neurological soft signs total score retained significance, with higher scores predicting shorter time to relapse (HR = 1.05, 95% CI = 1.01-1.10, p = .029). In a more detailed secondary analysis poorer social relationships predicted shorter time to relapse (HR = 0.85, 95% CI = 0.76-0.95, p = .003). Our predominantly negative findings suggest that many of the previously implicated risk factors for the time to relapse are mediated by non-adherence rather than having a direct effect on relapse-proneness. Neurological soft signs, and perhaps quality of life in social relationships appear to play a role and merit further investigation.