Comparative effectiveness of pharmacologic strategies to treat first episode psychosis

Antipsychotic medications are a cornerstone of care for FEP patients because they reduce the risk of relapse and hospitalization. Therefore, strategies to enhance adherence to antipsychotics are an urgent need in psychiatric practice. Ideally, antipsychotic medications would be studied in randomized trials. However, progress in the management of FEP cannot rely exclusively on randomized trials, which cannot answer all clinically relevant questions. Because randomized trials cannot realistically answer all questions about the effectiveness of early FEP interventions in all clinical populations and for all outcomes in a timely way, the findings from randomized trials need to be complemented with those from observational studies. Causal inference from observational data can be viewed as an attempt to emulate a hypothetical pragmatic randomized trial—the target trial. We will use data from the FEP-CAUSAL Collaboration to emulate target trials of initiation of oral and long-acting injectable antipsychotic medications. To overcome concerns about confounding due to noncomparability of individuals in different treatment groups, we will conduct our observational analyses in two steps. First, we will identify the observational data required to replicate findings from two flagship randomized clinical trials in this area: the EUFEST and PRELAPSE trials. This benchmarking of the observational effect estimates to existing randomized trial estimates allows to calibrate, and increase confidence in, the observational analyses. Second, we will emulate target trials that extend the results from previous trials for the management of individuals with FEP. Specifically, we will study the risks of nonadherence, relapse, and hospitalization over longer follow-up periods in adolescents and adults under different treatment strategies after FEP diagnosis. The findings from these analyses will help guide the choice of antipsychotics to enhance adherence and clinical outcomes. This Project 2 also complements the findings from the randomized trial in Project 1 using the FEP-CAUSAL Collaboration, an international consortium of prospective cohorts of individuals with FEP that is coordinated by LEAP’s Methods Core. Our target trial emulations will pioneer the implementation of causal inference methodology to observational prospective cohorts in FEP. The superiority of this methodological approach has been demonstrated in several areas of medicine but not yet in psychiatry. We expect that our work will result in methodological advances and in case studies that will be generally applicable across many areas of mental health research.