Improving postoperative opioid stewardship in cancer-related surgery: Impact of a pill-dispensing system on disposal and refill rates.

486 Background: Postoperative opioid misuse contributes to the opioid crisis, driven by overprescribing and accumulation of unused medications. A pill-dispensing technology (Addinex) addresses excess opioid use by controlling opioid access and enabling return of unused pills. A pilot study of this technology showed most patients returned the device, resulting in substantial unused pill disposal. This study expands the prior work to include symptom monitoring and explores impacts on opioid refills. Methods: We conducted a single-center prospective cohort study of adult English and Spanish-speaking patients undergoing cancer-related surgery, who received post-op oxycodone prescriptions (5-30 pills). Opioids were directly delivered in a password-protected Addinex device linked to a smartphone app to guide use and tapering. A prepaid mailer enabled secure return of unused pills. The app collected data on pill use and pain scores daily. Opioid refill rates were assessed via electronic medical records (EMRs) and state prescription monitoring programs. Primary outcomes included device return and disposal rates. Secondary outcomes included opioid use, refills, pain control, and system usability, assessed through app data, EMRs, and patient surveys. Results: To date, 82 patients have been enrolled with complete follow-up. 72 were female (88%). Surgeries included breast (55%), gynecologic (28%), and other (17%; soft tissue, urologic, and plastic). The cohort was racially and ethnically diverse (51% White, 18% Black, 10% Asian, 22% Hispanic). Of 82 patients, 57 (70%) returned the Addinex device after surgery with a median return time of 20 days (interquartile range 15-32). 50 patients (61%) obtained opioids from the device and 32 (39%) did not. Among the 50 device users, the median number of doses dispensed was 8.5 (range 1-20). In total, 1,527 opioids were prescribed; 501 were used (33%) and 1,026 (67%) remained unused. Approximately 702 pills (68% of unused) were returned by mail, whereas 259 (25%) were retained. Out of 66 patients who reported daily pain scores, 39 (59%) had pain for ≥7 days. At three months post-op, 15 out of 82 patients (18%) received a prescription refill. Of the 15 patients who obtained refills, 5 (33%) were device non-users. Conclusions: Use of the Addinex system promoted disposal of unused opioids, as the majority of excess pills were returned. High device return rates suggest the practicality of this intervention in clinical practice. Most patients used far fewer opioids than prescribed, and about a third reported pain after day 6, suggesting shorter opioid courses may suffice. While 18% of patients required a refill, one-third of those refills were among non-users, suggesting that lack of engagement with the Addinex device may be associated with continued opioid needs and may help identify high-risk patients for future interventions to curb use.