Abstract
Preoperative narcotic use is associated with poor postoperative pain management and worse outcomes after total joint arthroplasty (TJA). Therefore, identifying controlled substance use preoperatively is necessary. Electronic medical records (EMRs), prescription monitoring programs (PMP), or urine toxicology screening (UTS) are most commonly used. This study aims to compare the accuracy of EMR and PMP versus UTS to determine whether UTS should be implemented as standard of care in TJA preoperative assessment. Preoperative UTS was performed for primary or revision TJA from November 1, 2018, to March 31, 2019. Patient demographics, medical history, prescription history, and UTS results were retrospectively recorded. Prescription monitoring program and EMR were queried for prescription history in the past 2 years. The accuracy of EMR and PMP compared with UTS was calculated. Multivariable logistic regression analysis was performed to identify patient predictors associated with UTS+. Thirty of 148 patients had UTS+. Positive urine toxicology screening was more common in patients younger than 58 years, White race, and undergoing revision surgery. Electronic medical record and PMP documentation had the highest sensitivity (73.3%), specificity (92.4%), positive predictive value (71.0%), and negative predictive value (93.2%). Patients with higher odds of UTS+ include current/former smokers, those with a history of alcohol abuse, drug abuse, hepatitis C diagnosis, and mental illness. For patients without any risk factors for having a UTS+, the use of EMR and PMP may be sufficient to evaluate for controlled substance use; however, UTS should be considered in patients who present with one of the risk factors for UTS+.