Research Summary

Electronic Decision Support Boosts Deprescribing in Long-Term Care Homes

Edited by:
Ashton L. Stahl

Key Highlights

  • Electronic decision support increased deprescribing of at least 1 potentially inappropriate medication from 12.7% to 36.4%.
  • The number needed to treat was 4 to achieve deprescribing of 1 or more potentially inappropriate medications.
  • Falls were significantly higher in the intervention group.

pupcTake Your Understanding of Irritable Bowel Syndrome to the Next Level at the Practical Updates in Primary Care Virtual Meeting

Practical Updates in Primary Care provides expert-led education on irritable bowel syndrome and related clinical challenges. Join clinicians nationwide for practical insights, emerging research, and evidence-based strategies.

Register for Practical Updates in Primary Care today to deepen your clinical expertise and improve patient management.

An electronic decision support tool integrated into routine medication reviews significantly increased deprescribing of potentially inappropriate medications (PIMs) among older adults living in long-term care (LTC) facilities, according to a stepped-wedge cluster-randomized trial published in JAMA Network Open.

The study, conducted across 5 LTC homes in New Brunswick, Canada, found that residents exposed to electronically generated deprescribing reports were significantly more likely to have at least 1 PIM discontinued than those in usual care. Potentially inappropriate prescribing is common in LTC settings, where polypharmacy and complex comorbidities contribute to elevated risks of adverse drug events.

Methods

Investigators conducted a stepped-wedge cluster randomized trial from August 1, 2021, to October 31, 2022, during the COVID-19 pandemic. The study included older adults residing in 1 of 5 LTC homes who were prescribed at least 1 PIM. The homes were divided into 3 clusters; each cluster transitioned from control to intervention at 3-month intervals. Data analysis occurred from October 15, 2023, to March 24, 2025.

The intervention consisted of electronically generated, individualized reports identifying prioritized deprescribing opportunities. These reports were paired with preexisting quarterly medication reviews and accessed through a secure viewer. The primary outcome was the proportion of residents with at least 1 PIM deprescribed during control versus intervention phases, assessed every 3 months. Adjusted odds ratios (AORs) were calculated using a generalized linear model with a logit link, controlling for the number of PIMs, age, sex, language, and period as fixed effects, with participants nested within sites as random effects. Analyses were conducted according to the intention-to-treat principle.

Study Findings

A total of 1,228 residents were screened; 725 (59.0%) had at least 1 PIM and were included in the analysis. The median age was 84 years (IQR, 76-90), 65.9% were women, and the median number of medications was 10 (IQR, 7-13). Residents had a median of 3 PIMs (IQR, 2-4). The most commonly used PIMs at baseline were proton pump inhibitors (55.6%), benzodiazepines or sedative hypnotics (39.0%), and anticoagulants (37.1%).

During the control phase, 92 of 725 residents (12.7%) had at least 1 PIM deprescribed compared with 226 of 621 residents (36.4%) during the intervention phase. This result represented a risk difference of 23.7% (95% CI, 19.2%-28.2%) and a number needed to treat of 4. The adjusted odds ratio for deprescribing during the intervention was 1.58 (95% CI, 1.07-2.34).

The most frequently deprescribed medication classes during the intervention were opioids, antipsychotics, docusate, and benzodiazepines or sedative hypnotics. Falls were more frequent in the intervention group (20.6% vs 17.1%; AOR, 1.77; 95% CI, 1.15-2.71). Use of restraints was numerically higher but not statistically significant after adjustment (AOR, 1.95; 95% CI, 0.52-7.37). Delirium was uncommon in both groups.

Clinical Implications

According to the study authors, electronic decision support, when paired with existing clinical workflows, may offer a scalable and effective approach to deprescribing in LTC homes. They suggest that medication reviews should incorporate deprescribing as part of usual care.

The authors noted that the study was conducted during the COVID-19 pandemic, when changes in staffing and resident conditions may have influenced safety outcomes, such as falls. They also reported that although tapering instructions were provided, adherence to tapering was not monitored.

Expert Commentary

“This study found that electronic decision support paired with the usual workflow could render the deprescribing process scalable and effective,” the researchers concluded. “Incorporating deprescribing should be encouraged as best practice where time and resources permit.”

Reference
McDonald EG, Estey JL, Davenport C, et al. Electronic decision support for deprescribing in older adults living in long-term care: a stepped-wedge cluster randomized trial. JAMA Netw Open. 2025;8(5):e2512931. doi:10.1001/jamanetworkopen.2025.12931