RESUMO
AIMS: To assess the prevalence and characteristics of medication errors at hospital admission and discharge in people with Type 1 and Type 2 diabetes, and identify potential risk factors for these errors. METHODS: This prospective observational study included all people with Type 1 (n = 163) and Type 2 diabetes (n = 508) admitted to the Diabetology-Department of the University Hospital of Montpellier, France, between 2013 and 2015. Pharmacists conducted medication reconciliation within 24 h of admission and at hospital discharge. Medication history collected from different sources (patient/family interviews, prescriptions/medical records, contact with community pharmacies/general practitioners/nurses) was compared with admission and discharge prescriptions to detect unintentional discrepancies in medication indicating involuntary medication changes. Medication errors were defined as unintentional medication discrepancies corrected by physicians. Risk factors for medication errors and serious errors (i.e. errors that may cause harm) were assessed using logistic regression. RESULTS: A total of 322 medication errors were identified and were mainly omissions. Prevalence of medication errors in Type 1 and Type 2 diabetes was 21.5% and 22.2% respectively at admission, and 9.0% and 12.2% at discharge. After adjusting for age and number of treatments, people with Type 1 diabetes had nearly a twofold higher odds of having medication errors (odds ratio (OR) 1.72, 95% confidence interval (CI) 1.02-2.94) and serious errors (OR 2.17, 95% CI 1.02-4.76) at admission compared with those with Type 2 diabetes. CONCLUSIONS: Medication reconciliation identified medication errors in one third of individuals. Clinical pharmacists should focus on poly-medicated individuals, but also on other high-risk people, for example, those with Type 1 diabetes.