Derivation And Validation Of Risk Prediction For Posttraumatic Chronic Pain

Document Type

Conference Proceeding

Publication Date


Publication Title

The Journal of Pain


Chronic pain is common following traumatic stress exposure (TSE). Identification of individuals with chronic pain in the early aftermath of TSE is important to enable targeted preventive interventions. In this study we used baseline survey data from two prospective cohort studies to identify the most influential predictors of moderate-to-severe chronic pain. Self-identifying black and white American women and men (n=1,600) presenting to one of sixteen emergency departments (EDs) within 24 hours of motor vehicle collision (MVC) TSE were enrolled. Individuals with moderate-to-severe pain (≥4, 0-10 numeric rating scale) six months after MVC were identified via follow-up questionnaire. Sociodemographic, pain, general health, event, and psychological/cognitive characteristics were collected in the ED and used in prediction modeling. Ensemble learning methods and Monte Carlo cross-validation were used for feature selection and to determine prediction accuracy. External validation was performed on a hold-out sample (30% of total sample). Fifty-three percent of individuals reported moderate-to-severe pain six months following MVC. Regularized linear regression was the top performing learning method. The top thirty factors together showed good reliability in predicting chronic pain in the external sample (AUC=0.74±0.006). Top predictors included acute pain, opioid administration in the ED, high distress, depressive symptoms, and a high number of reported somatic symptoms. These analyses add to a growing literature indicating that influential predictors of pain can be identified and risk for future pain estimated from characteristics easily available/assessable at the time of ED presentation following TSE.





First Page