Clinical Question

Is deprescribing long-term medication safe and effective?

Bottom Line

The limited rigorous data on deprescribing suggest that many patients can safely stop unnecessary medication, but symptom relapse is significant. (Level of Evidence = 1a–)

Synopsis

The authors searched PubMed and EMBASE for randomized trials that compared deprescribing (i.e., the process of withdrawing unnecessary medications) with placebo or usual care. Two authors independently assessed the inclusion of studies and the risk of bias for each study. Although they reviewed the reference lists of the included studies, the authors do not describe a formal search or formal assessment of the potential of publication bias. They included 27 studies, each of which included between 20 and 2,471 patients. Sixteen of the studies used placebo and 11 used usual care as the comparator. The studies evaluated a wide range of drug classes, including antihypertensives, antipsychotics, corticosteroids, and so forth. The authors reasonably decided against pooling data because of the marked variability in the target drugs, target group (mean age varied between 50 and 89 years of age), and follow-up duration (four weeks to five years). Only 10 of the studies were of low risk of bias. The rate of successful deprescribing varied from 20% to 100%; in 19 of the studies the rate of successful deprescribing exceeded 50%. Sixteen of the studies reported on symptom relapse or resumption of deprescribed medications (range 0% to 80%). Among the nine placebo-controlled studies reporting on relapse, five found significantly greater relapse in the intervention groups (rate difference ranged from 14% to 50%). The included studies found that adverse events were infrequent.

Study design: Systematic review

Funding source: Self-funded or unfunded

Setting: Outpatient (any)

Reference: Thio SL, Nam J, van Driel ML, Dirven T, Blom JW. Effects of discontinuation of chronic medication in primary care: a systematic review of deprescribing trials. Br J Gen Pract. 2018;68(675):e663–e672.