Risk of death over six months in users of conventional and atypical antipsychotic drugs

Purpose: To compare the safety of conventional and nonconventional antipsychotics in older persons. Specific aims included to i) compare the risk of all-cause mortality among users of conventional versus atypical antipsychotic drugs, ii) investigate whether any increased mortality risk was limited to specific causes of death, and was still present after adjusting for comorbidities and, iii) examine mortality risks among users of conventional versus atypical antipsychotic in people with Alzheimer's Disease (AD). Method: The study included a retrospective cohort study of residents in Milan, Italy, aged 60+ who were prescribed an antipsychotic drug for the first time during April 1, 2002 to June 15, 2005. Data were obtained from the health authorities' information systems. 7739 persons using antipsychotic drugs (n = 2908 conventional, n = 4831 atypical) were included. Overall and cause-specific mortality over six months following first prescription of an antipsychotic drug was investigated. Covariates included, age, sex, and comorbities (cancer, cardio/cerebrovascular and other diseases). Mortality was assessed both in the whole cohort, and a subset of people with AD who were prescribed cholinesterase inhibitors. Results: People prescribed atypical antipsychotics had 65% lower all-cause mortality (HR 0.35, 95% CI 0.31-0.39) than those prescribed conventional antipsychotics. This mortality risk was explained by cancer deaths, with no differences within deaths for other conditions. Among people using cholinesterase inhibitors, no statistically significant difference in mortality was observed for users of conventional versus atypical antipsychotics. Conclusion: The characteristics of the observed differences in mortality suggest that there might be a preferential prescription of conventional drugs in terminally ill cancer patients. There is no evidence of a differential survival in AD patients using conventional or atypical antipsychotics.