Evaluating and Optimizing Care for Opioid Use Disorder using a Structured Data-Science Approach

Opioid use disorder (OUD) is a public health challenge that affects millions of people worldwide. Despite growing evidence supporting the effectiveness of medications for opioid use disorder (MOUD, including methadone, buprenorphine and extended-release naltrexone), access to these medications is still insufficient, with most patients remaining untreated. Therefore, enhancing knowledge about real-world effectiveness and guiding optimal use of MOUD in the clinical care of patients is of critical importance. This study, funded by NIH/NIDA K99/R00 award, leverages electronic heath records from the Veterans Affairs, state-of-the-art data models, machine learning algorithms and causal inference methods, to close these knowledge gaps and answer a set of timely questions centering around OUD care.