Using Machine Learning to Predict Treatment Adherence in Patients on Medication for Opioid Use Disorder.

OBJECTIVE: Patients receiving medication for opioid use disorder (MOUD) may continue using nonprescribed drugs or have trouble with medication adherence, and it is difficult to predict which patients will continue to do so. In this study, we develop and validate an automated risk-modeling framework to predict opioid abstinence and medication adherence at a patient's next attended appointment and evaluate the predictive performance of machine-learning algorithms versus logistic regression. METHODS: Urine drug screen and attendance records from 40,005 appointments drawn from 2742 patients at a multilocation office-based MOUD program were used to train logistic regression, logistic ridge regression, and XGBoost models to predict a composite indicator of treatment adherence (opioid-negative and norbuprenorphine-positive urine, no evidence of urine adulteration) at next attended appointment. RESULTS: The XGBoost model had similar accuracy and discriminative ability (accuracy, 88%; area under the receiver operating curve, 0.87) to the two logistic regression models (accuracy, 88%; area under the receiver operating curve, 0.87). The XGBoost model had nearly perfect calibration in independent validation data; the logistic and ridge regression models slightly overestimated adherence likelihood. Historical treatment adherence, attendance rate, and fentanyl-positive urine at current appointment were the strongest contributors to treatment adherence at next attended appointment. DISCUSSION: There is a need for risk prediction tools to improve delivery of MOUD. This study presents an automated and portable risk-modeling framework to predict treatment adherence at each patient's next attended appointment. The XGBoost algorithm appears to provide similar classification accuracy to logistic regression models; however, XGBoost may offer improved calibration of risk estimates compared with logistic regression.

Total hip arthroplasty in patients younger than 21 years: a minimum, 10-year follow-up.

INTRODUCTION: To evaluate the function, radiographic results and implant survival in patients younger than 21 years at the time of operation and with at least a 10-year follow-up, we reviewed the course of young adults who underwent total hip arthroplasty at McGill University Health Centre. PATIENTS AND METHODS: Twelve patients (16 hips) met the criteria for inclusion in this study. They ranged in age from 10-20 years (median 16.5 yr). One patient (1 hip) died 5 years after the index surgery and was excluded from the study. Functional results were measured by the Harris hip score, and radiographs of surviving implants were assessed for radiolucencies, osteolysis and eccentric polyethylene wear. RESULTS: Eleven arthroplasties were cementless, 2 were hybrid and 2 were cemented. At follow-up ranging from 10 to 25 years (mean 13.6 yr), the Harris hip score ranged from 34.2-97.2 (mean 64.5). Four (26%) acetabular components were revised and (at the time of writing) 1 required revision. Only 1 (7%) femoral stem was revised. Radiographically, there was a high prevalence of eccentric polyethylene wear, but none of the surviving implants were found to be loose. Overall, 67% (10 of 15) of the total hip arthroplasties continued to function well at a mean of 13.6 years postoperatively. CONCLUSIONS: Total hip arthroplasty remains a reasonable option in adolescence and early adulthood in selected patients with severe debility from multiple joint disease. Polyethylene wear is a particularly significant problem in this group of active young people.