Despite widespread use following orthopedic surgery, opioid utilization is rarely monitored in real time. Investigators often rely on patient self-report or dispensing records; leaving a critical gap in understanding actual day-to-day usage.1–3 Medication event monitoring systems (MEMS), Bluetooth enabled pill bottle caps that record time and date of openings, are the gold standard for measuring medication utilization.4 While utilized extensively in chronic disease symptom management to assess medication adherence, studies examining opioid utilization with MEMS are limited to cancer pain or non-orthopaedic postoperative settings.1,2 Opioid use carries risks and can impair aspects of postoperative recovery, including sleep.5 Objective monitoring with MEMS offers a novel approach to capture time-stamped medication utilization data. This has the potential to reduce reliance on self-report, which can be prone to recall bias, or prescription dispensing records that do not capture daily use. MEMS enables detection of daily patterns of medication utilization and can be leveraged to assess the effect of care interventions aimed at improving postoperative recovery.
Linking opioid use to postoperative recovery metrics, such as sleep, offers insights into the interplay between pain control, functional recovery, and potential opioid misuse. Sleep architecture is predictive of both short- and long-term postoperative pain outcomes.6,7 Specifically, poor sleep can increase pain sensitivity and impair healing. Understanding how postoperative sleep is affected by pain treatments, such as opioids, can provide critical insight into recovery trajectories and inform approaches to optimize both analgesic strategies and functional outcomes. While sleep disturbances are common following orthopaedic surgery, sleep is rarely measured objectively, and its relationship with opioid use is seldom explored objectively outside of inpatient or lab settings.6,8 Using daily, rather than aggregated, medication utilization and actigraphy data allows better understanding of how opioid use and symptoms, like sleep, change over time. This study assessed the feasibility of utilizing MEMS in postoperative settings to measure medication use and combine objective monitoring of opioid use with actigraphy to examine how daily opioid consumption is linked to postoperative sleep architecture.
Methods
Adults undergoing scheduled upper extremity orthopedic surgery due to arthritis (e.g., arthroplasty and arthroscopy to shoulders, elbow, wrist) without sleep apnea, insomnia, other chronic widespread bodily pain conditions (e.g., fibromyalgia), cancer, or autoimmune diseases were eligible. Participants were discharged the day of surgery. Individuals taking opioids or steroids 30 days preoperatively or undergoing revisions were ineligible. Of 329 patients screened between March 2022 and April 2023, 44 enrolled. To assess sleep, study coordinators provided participants with an actigraphy device (Actigraph wGT3X-BT) to wear on their non-surgical wrist for one week postoperatively. Participants placed opioid medication into a MEMS bottle (AARDEX Group), which recorded each bottle opening thereby assuming one dose was taken as other investigators previously reported.4 Each opening was converted to compute total morphine milligram equivalents (MME) over the week and MME daily (MED). Average pain scores were collected preoperatively and 2 weeks postoperatively. Mixed effect models examined relationships between opioid utilization and actigraphy measures, accounting for days from surgery and change in pre- to postoperative pain scores. The Institutional Review Board approved the study.
Results
Of 44 participants enrolled, 28 (63.6%) participants wore actigraphy and opened the MEMs cap at least once and were included in analysis. The mean age was 62.00 (±12.02) with the majority being female (67.90%) and white (75.00%) (Table 1). Participants wore devices for an average of 3.78 days (±2.06) with an average of 5 hours and 17 minutes of sleep per day with 6.51 awakenings. Sleep fragmentation, movement, and awakenings increased with each subsequent postoperative day. Participants opened the MEMS a median of 4 times over the week and utilized an average of 8.44 (±11.92) MME throughout the week postoperatively. However, when examining just days when medication was taken per subject, the average MED was 14.78 (±8.01). Opioid utilization peaked on Day 2 with an average MME use of 13.42 (±16.12). Postoperatively, average pain scores improved by 2.75 points. Mixed effects models indicated increased opioid utilization resulted in fewer awakenings (β=−0.09; p=.031), reduced sleep fragmentation (β=−0.31; p=.019), and less movement (β=−0.12; p=.043) (Figure 1). No association was observed between MED and other actigraphy measures.
Table 1:
Sample Characteristics (n=28)
| Variable | Sample Measures |
|---|---|
| Age, mean (SD) | 62.00 (12.02) |
| Sex | |
| Male | 9 (32.1%) |
| Female | 19 (67.9%) |
| Race | |
| White | 21 (75.0%) |
| Black or African American | 7 (25.0%) |
| Surgery Type | |
| Arthroplasty | 16 (57.1%) |
| Other | 12 (42.9%) |
| Regional anesthesia, n (%) | 21 (75.0%) |
| Total openings per participant, median [IQR] | 4 [0,16] |
| Morphine Milligram Equivalents, mean (SD) | 8.44 (11.92) |
| Morphine Milligram Equivalents per day (MED), mean (SD) | 14.78 (8.01) |
| Days of wear, mean (SD) | 3.78 (2.06) |
| Number of Awakenings, mean (SD) | 6.51 (5.65) |
| Efficiency, mean (SD) | 92.97 (6.22) |
| Total Sleep Time, mean (SD) | 317.36 (200.26) |
| Length of Awakenings, mean (SD) | 2.50 (1.82) |
| Movement Index, mean (SD) | 11.78 (8.01) |
| Fragmentation Index, mean (SD) | 9.80 (12.03) |
| Sleep Fragmentation Index, mean (SD) | 21.58 (17.12) |
| Past 24 hour Average Pain Score at Baseline, mean (SD) | 6.39 (2.30) |
| Past 24 hour Average Pain Score at 2-Weeks, mean (SD) | 3.64 (3.29) |
Figure 1:
Linear mixed effects models estimating actigraphy metrics predicted by daily MME utilization. When accounting for days since surgery and change in pain scores, Figure A illustrates that increased opioid utilization, measured in MME, was significantly associated with a decrease in awakenings (β=−0.09; p=.031). Similarly, Figure B illustrates the mixed effects model showing greater MME was linked to lower sleep fragmentation index (β=−0.31; p=.019). Figure C illustrates that in the adjusted mixed-effects model increased opioid utilization, measured in MME, was significantly associated with reduced movement during sleep (β=−0.12; p=.043)
Discussion
This study found that opioid use the week following surgery was associated with improved aspects of sleep architecture, specifically fewer awakenings and reduced movement yet overall sleep time was less than expert group recommendations. These findings reflect those in meta-analyses, showing reduced movement following opioid utilization.9 However, decreased movement does not translate to improved sleep quality as opioid use alters overall sleep architecture.9 While patients often report better subjective sleep when taking opioids, likely due to sedative effects or pain relief, studies using objective sleep measures alongside opioid use remain limited.5 This study is among the first to prospectively assess orthopaedic patients’ postoperative sleep outcomes and pain medication use through objective monitoring rather than relying on self-reports or records. This study indicates MEMS are feasible to use in prospective pain-focused studies, evident by a 63.6% compliance rate aligning with previous studies in differing patient populations and settings.10 MEMS can elucidate insights into adherence patterns and their relationship to recovery outcomes. Clinically, this integrated monitoring approach can inform personalized pain management strategies and optimize interventions to enhance postoperative recovery.
This study’s small sample size and follow-up period may limit generalizability. MEMS studies assume one dose is taken per opening, which may not be accurate for all subjects.4 Research linking medication consumption using MEMS to patient-reported medication utilization is warranted to better discern accuracy of this technology in measuring postoperative opioid use. Additional studies collecting multi-modal pain management approaches utilized by subjects, pre- and postoperative sleep quality, functional outcomes, as well as usability data and participant feedback on devices are needed to better investigate the role of postoperative opioid use captured with MEMS. The pairing of daily MEMS and actigraphy data in future longitudinal studies can inform when and how to strategically implement interventions to enhance postoperative recovery. By moving beyond self-report or chart review, this work advances the methodological toolkit for researchers and opens future lines of inquiry examining precision recovery strategies.
Funding:
Dr. Giordano received funding for this work, in part, by the Emory University Synergy Fund and the National Institute on Drug Abuse (K23DA057415). The funders did not have any influence on the outcome of the study.
Footnotes
Conflicts of Interest: The authors declare no conflicts of interest.
References
- 1.Lawal OD, Gold J, Murthy A, et al. Rate and Risk Factors Associated With Prolonged Opioid Use After Surgery: A Systematic Review and Meta-analysis. JAMA Netw Open. 2020;3(6):e207367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Schirle LM, Dietrich MS, Lam L, Stone AL, Bruehl S, Osmundson SS. Accuracy of patient-reported versus real-time electronic postoperative opioid use outcomes. Am J Obstet Gynecol MFM. 2021;3(4):100347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Daoust R, Paquet J, Williamson D, et al. Accuracy of a self-report prescription opioid use diary for patients discharge from the emergency department with acute pain: a multicentre prospective cohort study. BMJ Open. 2022;12(10):e062984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Giordano NA, Riman KA, French R, et al. Comparing medication adherence using a smartphone application and electronic monitoring among patients with acute coronary syndrome. Appl Nurs Res. 2021;60:151448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Puterflam JM, Comis JJ, Lan Q, et al. The short-term effects of opioid and non-opioid pharmacotherapies on sleep in people with chronic low back pain: A systematic review and meta-analysis of randomized controlled trials. Sleep Med Rev. 2022;65:101672. [DOI] [PubMed] [Google Scholar]
- 6.Chouchou F, Khoury S, Chauny JM, Denis R, Lavigne GJ. Postoperative sleep disruptions: a potential catalyst of acute pain? Sleep Med Rev. 2014;18(3):273–282. [DOI] [PubMed] [Google Scholar]
- 7.Cheah JW, Freshman RD, Mah CD, et al. Orthopedic sleep and novel analgesia pathway: a prospective randomized controlled trial to advance recovery after shoulder arthroplasty. J Shoulder Elbow Surg. 2022;31(6s):S143–s151. [DOI] [PubMed] [Google Scholar]
- 8.Driesman AS, Montgomery WC, Kleeman-Forsthuber LT, Johnson RM, Dennis DA, Jennings JM. Perioperative Sleep Quality Disturbances in Total Joint Arthroplasty is Multifactorial. J Arthroplasty. 2024;39(6):1474–1479. [DOI] [PubMed] [Google Scholar]
- 9.Tang NKY, Stella MT, Banks PDW, Sandhu HK, Berna C. The effect of opioid therapy on sleep quality in patients with chronic non-malignant pain: A systematic review and exploratory meta-analysis. Sleep Med Rev. 2019;45:105–126. [DOI] [PubMed] [Google Scholar]
- 10.Hartman L, Lems WF, Boers M. Outcome measures for adherence data from a medication event monitoring system: A literature review. Journal of Clinical Pharmacy and Therapeutics. 2019;44(1):1–5. [DOI] [PMC free article] [PubMed] [Google Scholar]