Predictors for starting depot administration of risperidone in chronic users of antipsychotics.

BACKGROUND Risperidone long-acting injectable (RLAI), the first second-generation depot antipsychotic, has extensively been studied before introduction. Thereafter, questions about the type of patients actually treated with RLAI in daily practice remain to be answered for making valid antipsychotic treatment comparisons involving RLAI in observational studies. OBJECTIVE We aimed to determine in chronic antipsychotic users who switched treatment, predictors for the prescription of (1) depot versus oral antipsychotics and (2) RLAI versus first-generation antipsychotics (FGAs) depot. METHODS We used pharmacy dispensing data from 53 community pharmacies in the northeast of the Netherlands containing approximately 500,000 persons. Chronic antipsychotic users were defined and followed up for a switch in antipsychotic treatment within the first period that RLAI was on the market. Multivariable analysis was performed to relate patient, prescriber, and medication characteristics to prescription of a new antipsychotic drug. RESULTS Predictors for switching to depot versus oral antipsychotics were male sex, previous use of depot antipsychotics, recent anticholinergic drug use, and a gap in antipsychotic dispensation history. Predictors for switching to RLAI versus FGA depot were previous use of depot and consulting a specialist. CONCLUSIONS The results suggest that, compared with oral antipsychotics, patients receiving a depot are less compliant users, with more extrapyramidal side effects. Compared with FGA depot, patients receiving RLAI tend to be more severely ill patients. We conclude that RLAI may be partly channeled to patients as a last resort, which may have important consequences for the interpretation of observational effectiveness comparisons between RLAI and other antipsychotics in daily practice.