The Impact of Deprescribing Interventions on the Drug Burden Index and Other Outcomes: A Systematic Review.

Objectives: The Drug Burden Index (DBI) calculates a person's exposure to anticholinergic and sedative medications. We aimed to review randomized controlled trials (RCTs) of deprescribing interventions that reported the DBI as an outcome, their characteristics, effectiveness in reducing the DBI, and impact on other outcomes.

Design: Systematic review with meta-analysis.

Setting And Participants: RCTs of deprescribing interventions where the DBI was measured as a primary or secondary outcome in humans within any setting were included.

Methods: Electronic databases, citation indexes, and gray literature were searched from April 1, 2007, to September 1, 2023. Quality was assessed using the Cochrane risk-of-bias tool.

Results: Of 1721 records identified, 9 met the inclusion criteria. Six interventions were delivered by pharmacists and 3 were delivered by pharmacists/nurses or pharmacists/geriatricians. All interventions required at least intermediate-level skills and involved multiple components and target groups. Studies were conducted in the community (n = 5), nursing homes (n = 2), and hospitals (n = 2). The mean or median age was ≥75 years and most participants were women in all studies. Most (n = 6) studies were underpowered. The follow-up period ranged from 3 to 12 months. Three studies reported a lower DBI in the intervention group compared with control: 1 pharmacist independent prescriber-delivered in nursing homes (adjusted rate ratio, 0.83; 95% CI, 0.74 to 0.92), 1 pharmacist/nurse practitioner-delivered in hospital (adjusted mean difference (MD), -0.28; 95% CI, -0.51 to -0.04), and 1 geriatrician/pharmacist-delivered in hospital (MD, -0.28; 95% CI, -0.52 to -0.04). Meta-analysis showed no difference in the change in DBI between control and intervention groups in the community including nursing homes (MD, -0.03; 95% CI, -0.08 to 0.01) or hospital setting (MD, -0.19; 95% CI, -0.45 to 0.06). Interventions had inconsistent effects on cognition and no effect on other reported outcomes.

Conclusions And Implications: RCTs of deprescribing interventions had no significant impact on reducing DBI or improving outcomes. Further suitably powered studies are required.