Predictors of Dropout from an Outpatient Treatment Program for Substance Use Disorders in India: a Retrospective Cohort Study of Patients Registered over a 10-Year Period (2009–2018)

This was a retrospective cohort study to determine the predictors of dropout in the first year from an out-patient treatment program for substance use disorders. Study period was from January 1, 2009, to December 31, 2018. Patients registered for detailed evaluation in the outpatient treatment program and followed up at least once were included. The primary end-point was dropout from treatment, defined as no follow-up at four time-points. We used Cox proportional hazard regression model to determine the predictors of dropout. Dropout rate at the end of 12 months: 80.37%. Dispensing free medicines (hazard ratio [HR]: 1.271), and self-treatment referral (HR: 1.116) increased risk of dropout. Dual diagnosis (HR: 0.906) and prescription of maintenance medication (HR: 0.734) reduced the risks of dropout. A priori knowledge of predictors could guide clinicians, service planners, and policymakers to provide tailor-made treatment and reorganize service delivery.