Schizophrenia affects approximately 1.1% of the population, or approximately 3.5 million Americans. The life expectancy of persons with schizophrenia is 10-25 years less than that of the general population, and has not improved in recent decades. Sudden cardiac arrest (SCA) and ventricular arrhythmia (VA), which account for 8-10% of all deaths in this population, is three times as common among those with than without schizophrenia. The most clearly established risk factor for SCA/VA in persons with schizophrenia is the use of antipsychotic agents, and this risk is dose-dependent. Second-generation antipsychotic agents (SGAs) are the most common treatment for schizophrenia, and are also used for other common mental health conditions including bipolar disorder, major depressive disorder, and psychosis/agitation associated with dementia. The prevalence of current antipsychotic drug use in US adults is 1.6%, or 3.8 million adults. A recent meta-analysis found that several widely used SGAs are associated with an elevated risk of SCA/VA. Given the high prevalence of polypharmacy in persons with schizophrenia and other mental health conditions, and the high potential for drug interactions in persons taking SGAs, clinicians badly need evidence-based guidance on which drugs do and do not increase the risk of SCA/VA in persons taking specific SGAs. Unfortunately, almost all available evidence about the health effects of potential DDIs involving SGAs comes from either 1) spontaneously reported adverse drug events, which provide limited evidence for causation because of their anecdotal nature, or from 2) pharmacokinetic studies of serum concentrations of SGAs, which include few subjects and do not examine health end-points. A more rigorous approach to identifying and elucidating drug interactions will help to identify drug-drug pairs that truly increase the risk of SCA/VA, and improve the safety of pharmacotherapy provided to persons with schizophrenia and other mental health conditions. To address these critical knowledge gaps about which drugs should and should not be avoided in persons receiving commonly used second generation antipsychotic drugs, we will perform high-throughput pharmacoepidemiology screening to identify drugs that may increase the rate of out-of-hospital SCA/VA in persons taking commonly used second-generation antipsychotic agents. Then, in two independent validation populations, we will conduct hypothesis- driven etiologic pharmacoepidemiology studies to either confirm or refute high-priority potential drug interactions, and elucidate factors that place patients at increased risk of out-of-hospital SCA/VA associated with specific drug pairs. The results of this research will provide drug interaction compendia editors with valid, actionable evidence that will allow them to warn clinicians about truly risky drug combinations. Of equal importance, this research will allow clinicians to be relieved of unnecessary and burdensome alerts about...