Compliance with refilling prescriptions for atypical antipsychotic agents and its association with the risks for hospitalization, suicide, and death in patients with schizophrenia in Quebec and Saskatchewan: a retrospective database study.

OBJECTIVE The objective of this analysis was to describe the patterns of compliance with atypical antipsychotics among patients with schizophrenia in actual practice in 2 Canadian provinces and to examine the relation between degrees of compliance and the risks of hospitalization, suicide, and death. METHODS Adults with a diagnosis of schizophrenia who filled at least 1 prescription for risperidone, olanzapine, or quetiapine were identified in the Quebec public prescription drug insurance plan database (from July 1, 2001, to December 31, 2004) and the Saskatchewan Health database (from January 1, 1999, to December 31, 2003). Compliance was assessed based on the medication possession ratio, which was estimated as the proportion of days for which medication was available over each month of follow-up (> or = 80% = good compliance; 50%-79% = moderate compliance; < 50% = poor compliance). The association between the early and long-term effects of compliance and the risks of hospitalization, suicide, and death were examined using Cox regression, with adjustment for baseline age, sex, and use of antidepressants, sedatives, and lithium (Quebec only). RESULTS respective 41,754 and 3291 patients were identified from the Quebec and Saskatchewan databases. Approximately half of the patients in each cohort were male and were aged < 45 years. Many patients had good compliance over the full follow-up period (Quebec, 61%; Saskatchewan, 45%); however, poor compliance was seen in 23% of patients from Quebec and 34% of those from Saskatchewan. Compared with poor compliance, the long-term effect of good compliance was associated with a significantly decreased risk of all-cause hospitalization (Quebec: adjusted hazard ratio [HR] = 0.60; 95% CI, 0.57-0.64; Saskatchewan: adjusted HR = 0.81; 95% CI, 0.69-0.95) and psychosis-related hospitalization (Quebec: adjusted HR = 0.37; 95% Cl, 0.34-0.40; Saskatchewan: adjusted HR = 0.45; 95% CI, 0.32-0.64). The Quebec data also indicated a significant association between good versus poor compliance and a decreased risk of death (adjusted HR = 0.58; 95% CI, 0.51-0.66) and suicide (adjusted HR = 0.68; 95% CI, 0.55-0.84) that was not observed in Saskatchewan. CONCLUSION In this retrospective analysis of patients with schizophrenia in Quebec and Saskatchewan, good compliance with atypical antipsychotic medications was associated with substantial reductions in the risk for all-cause and psychosis-related hospitalizations.