Emulated Target Trials of Steroids in Patients with Acute Respiratory Distress Syndrome

PROJECT SUMMARY/ABSTRACT Acute respiratory distress syndrome (ARDS) is a severe form of lung injury with significant public health implications due to severe morbidity and mortality. The need to utilize existing data to inform prospective research and clinical decision making was emphasized during the COVID pandemic, when ARDS became a leading cause of death, and clinicians were forced to operate outside of existing evidence. `Dynamic treatment regimes' (DTRs) are rules for making treatment decisions sequentially at multiple time-points based on a patient's evolving history. Most relevant treatment strategies for ARDS are DTRs. DTRs may be evaluated in randomized trials, however it is infeasible to conduct randomized trials testing all DTRs of interest. This grant proposes `target trial emulation' from observational data using `g-methods' for confounding adjustment to address multiple gaps in our knowledge about ARDS care. Target trial emulation with g-methods is essentially the gold standard for causal inference about DTRs from observational data, but the approach is underutilized in the critical care setting. We will also explore generating personalized DTRs based on machine learning derived phenotypes. The investigation will utilize two large datasets—including the Medical Information Mart for Intensive Care (MIMIC) IV database and the eICU collaborative research database—representing a wide geographic and demographic spectrum. We will specifically address questions surrounding initiation, duration and dosing of steroids among patients with ARDS in the following clinical aims. Using target trial emulations and g-methods, we will: 1) estimate effects of early and sustained steroid use compared with delayed, abbreviated, or no-steroids regimes across a range of doses in patients with ARDS, 2) estimate the effects of dynamic strategies for steroid initiation based on evolving markers of disease severity in ARDS patients, and 3) identify the effects of steroid strategies across cohorts defined by joint ARDS and sepsis status. We will utilize multiple datasets to assess stability of findings across centers. The project represents a collaborative effort between experts in critical care medicine (with a specialty in mechanical ventilation), critical care data science, and causal inference. Our results will address important gaps in clinical knowledge about treatment of ARDS and inform the design of future randomized trials. Our study designs, code, and constructed cohorts will also provide valuable templates for other researchers to use in future observational studies, which we hope will broadly improve the quality of evidence from observational data in critical care.