Association between admission medication reconciliation by pharmacists on the accuracy of hospital discharge medication lists

Introduction Medication errors upon hospital discharge can lead to patient harm. Medication reconciliation during the admission process can potentially impact discharge medication lists. This study aimed to determine the extent to which a pharmacist medication reconciliation on hospital admission affects the accuracy of discharge medication lists. Methods A prospective observational study was conducted in a major metropolitan tertiary hospital in Sydney, Australia. Patients in the hospital discharge lounge were recruited to evaluate the accuracy of their medication list in the discharge letter. Patients were categorized into two groups: (a) those who received a pharmacist admission medication history (PHARM) and (b) those who did not (non-PHARM). The primary outcomes were the proportion of patients with medication errors in their discharge letter. Results There were 102 patients included in the study (51 in each group). Patients in the PHARM group were less likely to have one or more medication errors in the discharge letter compared to the no-PHARM group (70.6% [n = 36/51] versus 92.2% [n = 47/51] [P = .010]). When restricted to prescription only errors (i.e. excluding over the counter [OTC] or herbal medications) the proportion with errors was 51.0% (n = 26/51) in the PHARM group versus 84.3% (n = 43/51) in the no-PHARM group (P = .001). Patients in the PHARM group were less likely to have errors that were of moderate or greater risk (31.4% [n = 16/51] versus 68.6% [n = 35/51] [P < .001]). After adjusting for confounders in the Poisson regression analysis, patients in the PHARM group had a lower error rate compared to the non-PHARM group (incidence rate ratio 0.51, 95% confidence interval 0.38-0.69, P < .001). Conclusion A pharmacist-completed medication history and reconciliation on hospital admission improves the accuracy of discharge medication lists. This study highlights that medication errors during the admission process may be propagated to hospital discharge.