Substance Use Prevention, Treatment and Recovery

MP24 - Novel Predictors of Short-term Retention in Methadone Maintenance Treatment

Monday, April 15, 2024

12:30 PM – 1:00 PM CT

Location: Exhibit Hall, Hall 3, Level 1, ACCC

Poster Presenter(s)

Krista Noam, PhD, MSc

Assistant Vice President of Analytics and Innovation
Carelon Behavioral Health
ROCKY HILL, Connecticut
BH

Bonni Hopkins, PhD

Senior Vice President Quality, Analytics & Innovation
Carelon Behavioral Health Connecticut
Rocky Hill, Connecticut

This poster describes the analysis performed to identify factors that predict 90-day retention in Methadone Maintenance Treatment (MMT). The analysis was performed on behalf of the Connecticut Behavioral Health Partnership (CT BHP) to support the providers providing MMT. The main predictors found were distance between a HUSKY Health member’s ZIP Code and that of their provider, receiving a take-home dose during the first 30 days for MMT, and previous MMT episodes. Other predicting factors were related to medical and substance use disorder diagnoses, and behavioral health service utilization.

Learning Objectives:

Clarify modifiable variables that contribute to improved short-term methadone maintenance retention among adults with opioid use disorder.
Test multiple models to predict retention, including logistic regression, decision tree, and Least Absolute Shrinkage and Selection Operator (LASSO) regression.
Share ideas about how similar projects could inform clinical decision support tools for providers or value-based reimbursement programs.