Effectiveness of Deprescribing Interventions in Older Adults: An Overview of Systematic Reviews

Thesis (Master's)--University of Washington, 2021Importance: Potentially inappropriate medications and unnecessary polypharmacy are associated with poor outcomes. Despite numerous systematic reviews examining deprescribing in recent years, rigorous evidence to guide deprescribing is limited, and priorities for deprescribing practice are still unclear.Objective: Synthesize results from systematic reviews (SRs) addressing the effectiveness of deprescribing interventions in older adults. Evidence Review: 11 databases were searched without language restrictions from January 2005 to October 2020. Inclusion criteria were developed a priori, using a conceptual framework and included SRs comparing deprescribing interventions to a control group, reporting on 18 outcomes of interest. Article selection, data extraction and quality assessment of included articles using a validated tool were conducted in duplicate, and conflicts resolved by consensus. Data synthesis was performed at three levels – at Level 1, we report SRs’ authors’ conclusions; at Level 2, results of SRs not including narrative of primary studies not of relevance and at Level 3 relevant primary studies. Findings: 2335 unique citations were retrieved, of which 93 full-text SRs were assessed for eligibility and 21 SRs and 13 meta-analyses included. SRs scored low or critically low in quality assessment. At Level 1, all SRs reported on benefits, while ten made conclusions regarding harm, only two found harm associated with deprescribing. At Level 2, majority of SRs found that deprescribing can reduce the number of medications: 16 (52%) found positive effects and nine (29%) found mixed positive effects. The majority of studies reporting on mortality and patient-centered outcomes reported mixed positive or no effects. Only seven SRs reported effects that were considered harmful across all outcome themes. Few SRs reported on cost outcomes or results according to pre-specified subgroups of interest. At Level 3, regardless of intervention type, deprescribing interventions were associated with positive or no effect on outcomes. Conclusions and Relevance: In older adults, deprescribing interventions can successfully result in medication withdrawal, and appear safe, but have demonstrated little impact on mortality, patient-centered outcomes or adverse effects. Few SRs examined utilization and cost outcomes, or focused on subgroups. Future research should focus on special populations