Overprescription of Opioids at Discharge for Surgical Patients Varies by Specialty and Provider Types
*Erin M White, *Holly N Blackburn, *Ruo Zhu Chen, Alexander S Chiu, Peter S Yoo
Yale University, New Haven, CT
Objective: Current guidelines suggest that post-operative inpatients requiring no opioids in the last 24 hours of admission should not be prescribed opioids at discharge. We sought to understand adherence to this recommendation and contributory factors.
Design: Retrospective medical record review.
Setting: Urban, university-affiliated healthcare system.
Patients: Post-operative inpatient admissions with length of stay >24 hours discharged to home from October 2017 to September 2019.
Interventions: None
Main Outcome Measures: Potential overprescription was defined as any patient who received no opioids in the 24-hour pre-discharge period but received an opioid prescription at discharge. Frequency of potential overprescription was analyzed using multiple binary logistic regression.
Results: During 7,637 (32.8%) of 23,289 eligible admissions, no opioids were given in the pre-discharge period. However, opioids were prescribed at time of discharge for 4,537 (59.4%) of these cases. Controlling for covariates (figure 1), potential overprescription was more likely to be observed among residents than advanced practice providers(OR=0.61, 95%CI 0.53-0.71) or attending physicians(OR=0.22, 95%CI 0.18-0.26). Patient factors including male gender(OR=1.22, 95%CI 1.07-1.29), Asian race/ethnicity(OR=1.47, 95%CI 1.09-1.98), and shorter length of stay(β=-0.081, p<.001) were also associated with increased odds. Significant variation was seen between procedural specialties. Potential overprescription was highest in Thoracic(n=188/223, 84.3%), Orthopedics(n=675/826, 81.7%) and General Surgery(n=1289/1673, 77%), and lowest in Cardiology(n=11/312, 3.5%), Pediatrics(n=25/140, 17.9%), and Gastroenterology(n=7/37, 18.9%).
Conclusions: Among post-operative inpatients, potential overprescription of opioids at discharge occurs frequently. It may be beneficial to focus future interventions on resident trainees and specialty services among which the odds of potential overprescription are highest. 
Back to 2020 Posters