Predictors of Attrition in an Interdisciplinary Pain Management Program

Purpose/Objective: This study sought to investigate the extent to which demographic and clinical characteristics predict which patients drop out of an interdisciplinary pain management program (IPP). Research Method/Design: Participants (N = 178 outpatients, 18–75 years of age) received treatment for various chronic pain conditions in an IPP (including biopsychosocial assessment, cognitive–behavioral, and physical therapies). Separate logistic regression analyses identified the demographic and clinical variables most predictive of attrition across five domains: (a) demographics, (b) number of medical and non/psychiatric diagnoses, (c) opioid use (yes versus no)/risk of misuse, (d) pain-related cognition and behavior, and (e) physical, social, and mental well-being. Significant predictors from the five domains were integrated in a final multivariable logistic regression model. Results: Among patients exposed to a 4-week IPP, 34% dropped out. In the final model, significant predictors of higher odds of attrition included younger age or being unemployed. Also, patients on opioids at preintervention had higher odds of completing the IPP than patients not on opioids at preintervention. Follow-up analyses revealed 24 of 37 completers (65%) on opioids at preintervention reduced or eliminated use over the course of the IPP. Conclusions/Implications: Because findings are limited by sample and design characteristics, they require replication yet offer novel hypotheses for identifying patients at risk of attrition. Specifically, patients with preintervention opioid use (contrasted with opioid dependence) may particularly benefit from an IPP.