OBJECTIVES: To evaluate the impact of pharmacist-led medication review on hospital readmission and length of stay through a meta-analysis. Additionally, the definitions of ‘patient hospitalization’ were explored.

METHODS: A systematic review mapping the meta-analyses on pharmacist services was previously performed. Meta-analyses assessing effect of pharmacist-led medication review reporting hospital readmission were selected. All primary studies included in the meta-analyses were extracted. Duplicates removal, screening and full-text appraisal of these primary studies were performed. Finally, primary studies of interventional design evaluating hospitalization had their data extracted by two independent reviewers (e.g. authors, source of hospitalization data, rates of admission). Discrepancies were solved by a third researcher. We performed a meta-analysis of intervention vs. usual care for hospital readmission using Peto method and random-effect model (CMA v.2). Results were reported as odds ratio (OR) with 95% confidence intervals (CI). Heterogeneity between-trials was measure using I2 index.

RESULTS: Overall 109 meta-analyses were gathered, of which nine were on pharmacist-led medication review, with six of them providing hospitalization data. A total of 35 randomized or cluster-controlled trials were included in the qualitative synthesis and 28 in the meta-analysis (n=15,388 patients). The meta-analysis revealed no significant difference in the impact between pharmacist-led medication review and usual care for overall hospital readmission (OR= 0.909 [95% CI 0.791-1.044], p=0.177). A moderate to high heterogeneity among trials (I2=56.9%) was found. The median length of stay in the intervention group was of 6.0 days (IQR 3-12) and of 7.97 days (IQR 6.9-11) for usual care (n=8 studies). Only one study (2.85%) provided a definition for the outcome of hospitalization as ‘admission for an in-hospital stay’.

CONCLUSIONS: The available evidence on the effect of pharmacist-led medication review on hospital readmission is insufficient and weak. No clear definition of hospitalization was found. Further well-designed and well-reported clinical trials are needed to provide additional conclusions.