A systematic review of outcomes reported in studies to optimise the medication use of patients at hospital discharge.

introductionCare transitions, specifically hospital discharge, hold a risk for drug-related problems and medication errors. Effective interventions that optimise medication use during and after transitions are needed, yet there is no standardisation of the outcomes. This literature review aimed at collecting outcomes from studies investigating how to optimise medication use of patients following hospital discharge, and to categorise them, as a first step in the development of a core outcome set.

methodsWe systematically reviewed quantitative and qualitative literature using Embase, PubMed, CINAHL and the EU Clinical Trial Register databases. Studies investigating the optimisation of medication use following hospital discharge were eligible. The quantitative literature review specifically included trials and protocols that evaluated the effect of an intervention for patients ≥ 65 years or multimorbid / polypharmacy patients, as they are at high risk of drug-related problems. The qualitative literature review focused on the patients' and healthcare professionals' views. Outcomes were summarised into unique outcome terms and categorised using an adapted version of the OMERACT filter 2.0.

resultsThe review included 75 quantitative and 20 qualitative studies. The interventions investigated in the quantitative literature mostly had multiple components performed either pre- or post-discharge. Sixty percent of the qualitative studies addressed the views of healthcare professionals, 40% the views of patients, and only one study addressed both. A median of 5 outcomes (range 1-17) were reported in the quantitative studies. In total, 91 unique outcomes were identified from the quantitative or qualitative literature, or both (73, 12 and 6 outcomes, respectively). Outcomes were categorised into five domains: 'medication' (n = 32 outcomes), 'economic impact/resource use' (n = 26), 'life impact' (n = 16), 'pathophysiological manifestations' (n = 15) and 'death' (n = 2). The top 5 most frequently measured outcomes in quantitative studies were number of readmissions (n = 54/75, 72%), mortality (n = 30/75, 40%), number of emergency department visits (n = 26/75, 35%), number of outpatient physician visits (n = 12/75, 16%), and medication adherence (n = 12/75, 16%). DISCUSSION AND

conclusionThis study identified a large number of different outcomes, especially in the domains medication and economic impact/resource use. This heterogeneity impedes the identification of effective interventions and confirms the need for a core outcome set.