Impact of Deprescribing Interventions in Older Hospitalised Patients on Prescribing and Clinical Outcomes: A Systematic Review of Randomised Trials

Janani Thillainadesan, Danijela Gnjidic, Sarah Green, Sarah N. Hilmer

Polypharmacy and potentially inappropriate medications (PIMs) are prevalent in older adults in hospital, and are associated with negative outcomes including adverse drug reactions, falls, confusion, hospitalisation and death. Deprescribing may reduce inappropriate polypharmacy and use of inappropriate medications. The aim of this systematic review was to investigate the efficacy of deprescribing interventions in older inpatients to reduce PIMs and impact on clinical outcomes. Ovid MEDLINE, Embase, Informit, International Pharmaceutical Abstracts, Scopus, PsycINFO, the Cochrane Central Register of Controlled Trials (CENTRAL) and CINAHL were searched for randomised controlled trials (RCTs) from 1996 to April 2017. RCTs reporting on deprescribing interventions to reduce PIMs in older hospitalised adults were eligible. Data were extracted, and study quality assessed. The primary outcome was reduction in PIMs. Where available, clinically relevant outcomes were assessed. Nine RCTs (n = 2522 subjects) met the inclusion criteria. Deprescribing interventions were either pharmacist-led (n = 4), physician-led (n = 4) or multidisciplinary team-led (n = 1). Seven of the nine studies reported a statistically significant reduction in PIMs in the intervention group. There was no change in one study where there were zero PIMs on admission and discharge, and in the other study a reduction in PIMs that was not statistically significant was observed. There was significant heterogeneity in outcome measures and reporting. Few studies reported on the impact of deprescribing interventions on clinical outcomes. Reported clinical outcomes included drug-related problems (n = 3), quality of life (n = 2), mortality (n = 3), hospital readmissions (n = 4), falls (n = 3) and functional status (n = 2). Most studies reported a benefit in the intervention group that was not statistically significant. No notable harm was observed in the intervention group. There was a high risk of bias in the included studies. The evidence available suggests that deprescribing interventions in hospital are feasible, generally effective at reducing PIMs and safe. However, the current evidence is limited, of low quality and the impact on clinical outcomes is unclear.