Rates and predictors of relapse in first-episode non-affective psychosis: a 3-year longitudinal study in a...

Relapses may represent a critical hazard in schizophrenia spectrum disorders as they are associated with an increased risk of a clinical and functional deterioration. Preventing relapse after recovering from a first psychotic episode has become a major challenge due to its critical impact on lifelong functionality. This study explored the rate of first and second relapses and the predictors associated with these relapses in a large cohort of non-affective psychosis patients during a period of 3 years after the first break of the illness. From February 2001 to May 2014, sociodemographic and clinical data from an epidemiological cohort of 341 non-affective first-episode psychosis patients at risk of relapse were analysed at a specialized early intervention service. Logistic regression, Cox regression, and Kaplan-Meier survival analyses were performed to compare non-relapsed and relapsed patients. One hundred and sixty-six (48.7%) individuals relapsed at least once. Median time to relapse was 17.0 months in non-adherent patients and 40.0 months in adherent patients (log-rankχ ²: 51.36; p < 0.001). Non-adherence to medication (odds ratio-OR 2.979; p < 0.001), schizophrenia diagnosis (OR 2.173; p = 0.002), and age of onset (OR 1.020; p = 0.033) were the main predictors of the first relapse. Fifty-six subjects experienced a second relapse (33.73%) predicted by diagnosis (OR 1.975; p = 0.074), age of onset (OR 1.078; p = 0.003), and positive symptoms (OR 0.863; p = 0.03), but not adherence. Non-adherence is the main predictive factor of first relapse after a first episode of psychosis. Second relapses were not often and not related to modifiable factors, suggesting that multiple relapsed patients may comprise a subgroup with a higher biological risk.