Study objective: To explore the suggested association between atypical neuroleptic use and the development of diabetes mellitus, with a focus on older adults.
Design: Retrospective cohort study.
Subjects: Eleven thousand one hundred four older (> 65 yrs) residents of long-term care institutions in Ontario, Canada, who received either atypical neuroleptic agents, typical neuroleptic agents, benzodiazepines, or corticosteroids.
Measurements and main results: Each subject was followed for the development of a diabetic event, defined as newly prescribed antidiabetic drug therapy. Our Cox regression model was adjusted for age, sex, socioeconomic status, comorbidity, and concomitant use of beta-blockers, thiazide diuretics, and antiepileptic agents. The adjusted hazard ratio for the development of diabetes in patients receiving atypical neuroleptics compared with those receiving benzodiazepines (control group) was 0.89 (95% confidence interval [CI] 0.66-1.21). The adjusted hazard ratio for typical neuroleptic users compared with the benzodiazepine group was 1.27 (95% CI 0.91-1.77). As expected, patients receiving corticosteroid therapy were almost twice as likely to develop diabetes as those receiving benzodiazepines (adjusted hazard ratio 2.2, 95% CI 1.41-3.12). For patients receiving atypical neuroleptic agents, no statistically significant difference in the percentage of diabetic events was found among individual agents (2.1% olanzapine, 1% quetiapine, 2.1% risperidone).
Conclusion: Drug therapy with atypical neuroleptic agents in older adults did not increase their risk of developing diabetes mellitus.