Background Medicines errors are a major cause of adverse events in hospitalised elderly patients and increase morbidity, mortality and healthcare costs.

Purpose To improve the reconciliation process in these patients, to establish the degree of risk of the discrepancies, to analyse potentially inappropriate prescriptions (STOPP criteria) and to identify drug interactions.

Materials and methods Retrospective and descriptive study conducted at a general hospital from January to December 2012 on patients aged over 75. The patient’s usual medicines were recorded by HORUS (software application of outpatient clinic medicines records), medical history and interview with the patient. The patient’s chronic medicines were compared with the prescribed at admission to identify discrepancies classified according to the ‘Consensus Document on Terminology and Classification in Medication Reconciliation’. The potential risk of reconciliation errors (REs) was evaluated based on the NCCMERP index. We reviewed potentially inappropriate prescriptions (STOPP criteria) and drug interactions.

Results Medicines reconciliation was performed in 1,530 patients, 59.71% were women. 13,117 drugs were evaluated (8.64/patient) and 2,722 discrepancies were detected (1.78/patient). More frequently justified discrepancies were not to prescribe a drug due to clinical and medical decisions (33.73%), and change of dose or route of administration of a drug based on new clinical situation (28.04%). Most common causes of REs were: omission of chronic medicines (73.53%) and incorrect dose, route or frequency (17.35%). The risk associated with REs was category C (71.76%), category D (25%), and category E (2.35%). There were 80 inappropriate prescriptions according to STOPP criteria (6.92% of patients). 187 clinically significant drug interactions were found (15.56% of patients).

Conclusions The incorporation of the reconciliation process in the hospital has enabled us to detect and intercept REs. Before any prescriptions are written it is necessary to consider all aspects of elderly patients’ conditions that may affect the efficacy, safety and success of pharmacotherapy.

No conflict of interest.