E. A. Dasinger1,2, L. A. Graham1,2, T. S. Wahl1,2, S. J. Baker1,2, M. T. Hawn3, T. Hernandez-Boussard3, K. Desai3, J. S. Richman1,2, K. M. Itani4, G. L. Telford5, S. J. Knight1,2, M. S. Morris1,2 1University Of Alabama At Birmingham,Birmingham, AL, USA 2Birmingham VA Medical Center,Birmingham, AL, USA 3VA Palo Alto Healthcare Systems,Palo Alto, CA, USA 4VA Boston Healthcare System,West Roxbury, MA, USA 5Clement J Zablocki Veterans Affairs Medical Center,Milwaukee, WI, USA
Introduction: The number of patients taking opioids has risen drastically over the last decade and recent literature suggests that preoperative opioid use is associated with higher readmission rates. Although opioids are widely used to manage acute postsurgical pain, some patients remain opioid dependent following surgery. This study examines the relationship between opioid use and surgical readmissions and evaluates the incidence of new persistent opioid use in a veteran population.
Methods: We performed a national retrospective cohort study of general, orthopedic, and peripheral vascular inpatient surgeries occurring in the VA Healthcare System between October 2007 and September 2014. Pharmacy outpatient data within the VA Corporate Data Warehouse was used to calculate the proportion of days covered (PDC) for opioid medications in the six months prior to surgery and six months post-discharge. Patients were stratified into four groups defining preoperative opioid usage: no use, infrequent use (< 2 prescription fills or < 30 days of supply), frequent but not daily use (≥ 3 prescription fills with < 80% PDC), and daily use (≥ 3 prescription fills with ≥ 80% PDC). Our primary outcome of interest was unplanned 30 day readmission rates. Univariate and bivariate statistics along with adjusted logistic models were used to examine odds of pain-related readmission.
Results: A total of 237,441 patients were included in the analysis. In the six months prior to surgery, 59.8% showed no evidence of opioid use, 18.5% were considered infrequent users, 7.9% were frequent users, and 13.8% were considered daily opioid users. The adjusted odds of pain-related readmission within 30 days of discharge were higher for those with opioids on hand at admission and for the three groups with exposure to opioids within the six months prior to surgery as compared to the opioid naïve group: opioids on hand at admission (OR 1.17; 95% CI 1.05-1.31), infrequent (OR 1.12; 95% CI 1.02-1.23), frequent (OR 1.24; 95% CI 1.08-1.42), and daily (OR 1.40; 95% CI 1.23-1.59). Overall, patients who filled opioids at discharge had higher odds of pain related readmission within 30 days of discharge (adjusted OR 1.51; 95% CI 1.29-1.78). Of the previously opioid naïve patients, 6.8% became frequent users and 2.8% became daily opioid users at six months post-surgery. Of the previously infrequent opioid users, 19.4% became frequent users and 7.5% became daily opioid users at six months post-surgery.
Conclusion: Preoperative and postoperative opioid use is associated with an increased risk for readmission. Decreasing the use of opioids before and after surgery may improve surgical quality and lead to better outcomes. Persistent opioid use following surgery is common and providers should minimize the amount of opioids prescribed and consider alternative pain management strategies to prevent patients from moving beyond acute opioid use.