ID: 208410 Development of Machine Learning-Based Models to Predict Short Term Success in Substance Use Disorder treatment

Prescription opioids have resulted in large part to a substance use disorder (SUD) epidemic. Relapse rates for patients with SUD within 30 days of leaving an inpatient drug and alcohol treatment center are reported at 40-60%. When determining which therapy (e.g., opioids v. neuromodulation) is ideal for a person with chronic pain, understanding the factors that lead to treatment refractory SUD is essential. We aim to show proof of principle for machine learning (ML) and predictive analytics on short-term success or failure with SUD treatment.