Abstract
OBJECTIVE: Postoperative opioid use is commonly determined by patient-report; however, the accuracy of this measure is unclear. We aimed to determine the accuracy and reliability of patient-reported daily opioid consumption up to day 7 and over weeks 1, 2, and 3 after hospital discharge. METHODS: This was a secondary analysis of a randomised trial which tested the comparative effectiveness of 2 opioid regimens. Participants' daily opioid consumption was recorded by patient self-report and the quantity of returned pills in blister packs during the treatment period. Intraclass correlation coefficient (ICC) and Bland-Altman plots for repeated measures were used to assess the agreement between each collection method. RESULTS: Records with complete patient-reported daily opioid consumption and returned blister packs were included in the analysis (81 of 120). For day 1 postdischarge, the ICC between patient-reported opioid consumption and consumption determined by blister-pack countback was 0.730. From days 2 to 7, the ICC was consistently greater than 0.8, and the mean difference was consistently less than half a tablet. For week 1 overall, the ICC was 0.904, and the mean difference was -0.4 tablets. For weeks 2 and weeks 3, the ICC was 0.953 and 0.884, respectively. CONCLUSION: Patient-reported consumption was a reliable and accurate measure of opioid consumption after hospital discharge. Daily consumption demonstrated less reliability and accuracy during the first 24 hours (day 1) which improved overtime. Although patients tended to under-report use, this difference was not seen as clinically meaningful.