Using medications for prediction of revision after total joint arthroplasty

Abstract

This study evaluated the ability of a pharmacy based co-morbidity measure (RxRisk-V) to predict odds of one and five years revision in total hip arthroplasty (THA) and total knee arthroplasty (TKA) and compared its performance to the more commonly used co-morbidity measures in orthopaedics (Charlson and Elixhauser). 11,848 patients with THAs and 18,972 with TKAs performed between 2001 and 2012 were evaluated. Using a combination of conditions, identified by both the pharmacy and diagnoses based coding algorithms, models with acceptable predictive ability of THA and TKA revision were developed. These findings suggest prescription based co-morbidity measures can positively contribute to case-mix adjustment and outcome prediction in this patient population.