Predictors of medication adherence in a large 1-year prospective cohort of individuals with schizophrenia: insights from the multicentric FACE-SZ dataset

Abstract

Abstract Schizophrenia is characterized by the most salient medication adherence problems among severe mental disorders, but limited prospective data are available to predict and improve adherence in this population. This investigation aims to identify predictors of medication adherence over a 1-year period in a large national cohort using clustering analysis. Outpatients were recruited from ten Schizophrenia Expert Centers and were evaluated with a day-long standardized battery including clinician and patient-rated medication adherence measures. A two-step cluster analysis and multivariate logistic regression were conducted to identify medication adherence profiles based on the Medication Adherence rating Scale (MARS) and baseline predictors. A total of 485 participants were included in the study and medication adherence was significantly improved at the 1-year follow-up. Higher depressive scores, lower insight, history of suicide attempt, younger age and alcohol use disorder were all associated with poorer adherence at 1 year. Among the 203 patients with initially poor adherence, 86 (42%) switched to good adherence at the 1-year follow-up, whereas 117 patients (58%) remained poorly adherent. Targeting younger patients with low insight, history of suicide, alcohol use disorder and depressive disorders should be prioritized through literacy and educational therapy programs. Adherence is a construct that can vary considerably from year to year in schizophrenia, and therefore may be amenable to interventions for its improvement. However, caution is also warranted as nearly one in five patients with initially good adherence experienced worsened adherence 1 year later.