To the Editor: Stefan Leucht, M.D., et al. (1) concluded their meta-analysis by comparing the difference in relapse rates between atypical and typical antipsychotic agents to that produced by aspirin in preventing vascular events. But this comparison does not support the widespread use of atypical antipsychotics. A year’s supply of enteric-coated aspirin costs less than $10 and reduces the risk of fatal or disabling myocardial infarction and stroke by 8% per year. A year’s supply of an atypical agent costs thousands of dollars more than a typical agent, but when compared to a typical agent, it reduces the risk of psychotic relapse by 8% per year.

This is not to diminish the impact of psychosis nor does it serve as an argument for reducing expenditures for those with serious mental illness. Rather, it questions whether the billions of dollars currently spent on atypical antipsychotics might not produce a greater reduction in mortality, morbidity, and misery if spent on more robust interventions, such as assertive community treatment or supported employment and adequate housing. Perhaps providing atypical antipsychotic medication to a population that is 85% unemployed, has 10–20 times higher rates of homelessness, 8–10 times higher rates of criminal justice involvement, and 2–3 times higher rates of substance abuse is more like "giving an aspirin" than the authors had intended.