[Observational study of factors predictive of medication errors and of prioritization criteria in geriatric medication reconciliation].

Objectives: The geriatric population, often polymedicated, is exposed to the risk of adverse drug events. Medication reconciliation (MR), which is an interactive and pluriprofessional process, helps ensure continuity of care. The objective of this study was to analyze and to define relevant prioritization criteria for MR in older patients in order to avoid a maximum of medication errors.

Methods: A clinical audit of MR at the transition points of patient admission and discharge was conducted prospectively for 10 months. Patients were selected on the basis of a prioritization procedure already established in our structure, that is the presence of at least one of the three following criteria: originating from an hospital department, severe renal failure and prescription of at-risk drugs.

Results: The cohort of patients reconciled at admission included 136 patients. A total of 63 unintentional discrepancies (UDs) were identified, the majority of which (76.2%) involved drug omissions. Three criteria were identified as independent predictors of UDs risk on the entry prescription compared to the optimized drug assessment: rheumatological history, originating from an hospital department and hyponatremia. Hyponatremia was found in the present study to be the most relevant criterion that significantly increased the risk of having an UD on the patient's prescription, particularly a risk of treatment omission at admission.

Conclusion: This study will allow to improve the prioritization criteria on the healthcare establishment's procedure and to implement MR in geriatric day hospitalization in order to strengthen the city-hospital link.