543

Abstract

Introduction: Older trauma patients have a threefold increased risk of death compared to younger patients, in part due to higher comorbidity burden. However, the relative contribution of multiple medications intake (polypharmacy) on trauma outcomes remains unclear. The Comorbidity-Polypharmacy Score (CPS), which assigns a point for each chronic health condition and medication, has recently been proposed as an alternative to the Charlson method that does not include medications. Methods: This was a retrospective analysis of trauma patients aged ≥ 55 between January 1 and December 31, 2010 admitted to a level 1 trauma center. Older patients identified in the Trauma Registry were included, with complementary data recorded through medical records review. State Department of Health records were obtained for 1-year mortality data. Kaplan-Meier and logistic regression were used to evaluate mortality risk. The predictive ability of CPS and Charlson were compared by receiver operator curves. Results: A total of 667 patients were included. Mean age was 71 years, and 60% were male. Of these, 89 died in the hospital and 53 additional patients died within 1 year of injury. CPS is an independent risk factor for 1-year mortality, after adjusting for age, injury severity, ED hypotension and the presence of traumatic brain injury (Adj. OR 1.05, 95% CI 1.01-1.08 for each point assigned). However, multivariate models with Charlson score are better predictors of 1-year mortality as compared to similar models that include CPS (AUC 0.77 for Charlson score compared to 0.75 for CPS, p=0.02). Conclusions: Whereas higher CPS correlates with 1-year mortality, the inclusion of polypharmacy is not superior to Charlson scoring in predicting late trauma deaths in older adults.