A simple checkbox reminder may influence postoperative opioid prescribing behaviour in acute pain service clinicians

Opioid over-prescribing following surgery is associated with diversion, misuse and overdose [1]. Our acute pain service implemented patient-specific opioid tapers based on inpatient consumption and reduced opioid prescriptions following surgery [2, 3]. We then introduced inpatient opioid de-escalation intended to more accurately predict patient requirements at discharge. A simple reminder (in the form of a checkbox) was added to our paper-based patient tracking tool in order to prompt prescribers to discontinue intravenous and scheduled oral opioids within 48 h of surgery, rather than automatic discontinuation. Before this intervention, our acute pain service did not have a standardised opioid de-escalation protocol. We hypothesised that the checkbox reminder would reduce inpatient opioid consumption in patients undergoing total knee arthroplasty (TKA). This retrospective cohort study was approved by the institutional review board (Stanford, CA, USA) and Veterans Affairs (VA) Research Committee (Palo Alto, CA, USA) and patient consent waived. The study took place at a tertiary care university-affiliated VA hospital with a peri-operative surgical home model [4]. All patients who underwent primary TKA between May 2018 and April 2020 were identified through a local quality improvement database. Patients were excluded if they underwent any other surgery; were taking long-acting opioids, buprenorphine or > 100 morphine mg equivalents (MME); or hadanopioid allergyorpostoperative complication requiring departure from the clinical pathway. Patient records were examined for demographic data; medical history; anaesthetic technique; nerve block technique; inpatient opioid administration; and lengthof stay. Analgesia for all patients undergoing joint replacement is managed by the acute pain service and patients receive standardised multimodal analgesia, which has been described previously [2]. Briefly, the analgesic regimen includes an adductor canal continuous peripheral nerve block, regular paracetamol, celecoxib and oxycodone, and oxycodone and intravenous hydromorphone when required. Starting in May 2019, a checkbox reminder was added to our paper-based tracking sheet, which is used to document daily analgesic requirements and medication changes. The checkbox reminderswere added to postoperative days 1 and 2 of the tracking sheet in order to prompt prescribers to discontinue intravenous hydromorphone on postoperative day 1 and scheduled oxycodone on or before postoperative day 2 if clinically appropriate. The primary outcome was 2-day inpatient postoperative opioid use (in morphine milligram equivalent, MME) 1 year before (pre) and 1 year after (post) intervention. Secondary outcomes included time to cessation of intravenous and scheduled oral opioids and total MME administered per day. Statistical analyses were performed using NCSS-PASS software (Kaysville, UT, USA). Normality of distribution was determined using the Kolmogorov–Smirnov test. For normally distributed data, single comparisons were performed using Student’s t-test; the Mann–Whitney U test was used for continuous data in non-normal distributions. The chi-square test or Fisher’s exact test (n < 5 in any field) was used for comparisons of categorical data. Times to cessation were plotted on Kaplan–Meier survival curves and analysed using the log-rank test. Pairwise comparisons were conducted using log-rank tests with reporting of Mantel–Haenszel probability levels. Ap value< 0.05was considered statistically significant. From a total of 360 consecutive patients who underwent primary TKA between 1 May 2018 and 1 April 2020, 15 were excluded: long-acting opioids or buprenorphine (n = 12); high MME (n = 1); postoperative complications (n = 1); and opioid allergy (n = 1). Data for 345 patients (203 preand 142 post-introduction) were analysed. There were no significant differences in baseline characteristics or anaesthetic techniquesbetween the groups (Table 1). Two-day median (IQR [range]) postoperative MME decreased from 67.5 (37.5–117.9 [0–423.9]) mg pre to 59.7 (15.0–114.0 [0–270.9]) mg post, p = 0.038. Median (IQR [range]) postoperative day-1 MME was no different between preand post-introduction, 45.0 (30.0–74.4 [0– 250.0]) vs. 37.5 (15.0–67.5 [0–178.8]), respectively, p = 0.090. Median (IQR [range]) postoperative day-2 MME was lower for post than pre, 20.0 (0.0–44.9 [0–118.8]) vs. 22.5 (7.5–45.0 [0–247.5]) respectively, p = 0.021 (Fig. 1). Median (IQR [range]) time to cessation for intravenous opioids was 1 (1–1 [0–4]) days post compared with 1 (1–1