Randomised controlled trials (RCTs), quasi-RCTs and before-and-after studies that assessed interventions to improve medication reconciliation in community-dwelling adults in primary care, ambulatory settings or transition into or out of hospital were eligible for inclusion. Studies of patients who were living in or being transferred to long-term care or retirement homes were excluded from the review. The primary outcome was the number of discrepancies in name, dose and frequency between medications reported in the medical record and patient-reported medications. Secondary outcomes were clinical relevance of medication discrepancies and change in level of patient knowledge of their medication. Studies were required to report complete results.
Included studies took place in ambulatory settings or post hospital discharge situations. Half of the studies assessed multi-faceted interventions and half assessed interventions described as pharmacist-mediated but using multiple data-sources. All studies included only adult patients; characteristics of patients varied considerably between studies. There were differences in methods and timing of outcome evaluation.
One reviewer assessed abstracts and subsequently two reviewers independently assessed identified full papers for inclusion. Disagreements were resolved through consensus.