PROJECT SUMMARY/ABSTRACT Asthma is a leading cause of morbidity in the US and worldwide, affecting almost 10% of the US population. Individuals with severe asthma, who are uncontrolled on conventional therapy, account for a disproportionate share of asthma-related morbidity and mortality. For these patients, the advent of multiple monoclonal antibodies “biologics” has provided additional therapeutic options to improve outcomes in this potentially fatal disease. While these biologics have helped to improved outcomes in many patients with asthma, there is considerable uncertainty on the optimal choice of biologic in patients who meet criteria for two or more biologics, a common occurrence. There are no head-to-head trials of these therapies and results of indirect treatment comparisons have been limited or conflicting. Moreover, indirect comparisons are fraught with multiple limitations, including that the populations recruited into different trials for the therapies being compared may differ in ways which may influence the outcome(s) of interest. Consequently, current asthma guidelines have limited evidence-based guidance on how to approach biologic initiation in patients who are eligible for multiple biologics. To optimize the value of these expensive biologics, data on their comparative effectiveness are solely needed. In the absence of randomized trials, observational data can be used to generate evidence on the comparative effectiveness of these biologics using ‘real-world’ data. However, to mitigate biases in non-experimental research and to ensure sound conclusions are drawn from the data, we need to leverage subject matter expertise with advanced and innovative causal inference and pharmacoepidemiology techniques. This proposal leverages the PI and study team’s clinical expertise in asthma and monoclonal antibody research with expertise in causal inference and epidemiologic study design to emulate hypothetical randomized and adaptive trials in answering questi