[Medication reconciliation at hospital admission: Results and identification of target patients]

Objective: To quantify and to classify the discrepancies between the admission treatment and the usual patient treatment. To determine the variables that predict those patients that will have more benefit from medication reconciliation.

Material and methods: A prospective medication reconciliation study was conducted in the Vascular Surgery Unit from March 2014 to December 2014. When the patients were admitted to the Vascular Surgery Unit, they were informed about the study and asked to prepare information about their chronic treatment. The pharmacist then checked their clinical records, outpatient prescriptions, and also interviewed the patient, obtaining the best pharmacotherapeutic history available. The discrepancies with the admission treatment were written into the patient electronic clinical records. Finally, the physician classified the discrepancies, and changed the treatment, if needed. The statistical analysis included a comparison between patients with and without a non-justified discrepancy (NJD). The statistically different characteristics were used to plot Receiver Operating Characteristic curves, in order to determine the sensitivity and the specificity of these variables to select patients with discrepancies.

Results: A total of 380 patients were included. There were 845 non-justified, 600 justified non-documented, and 439 justified documented discrepancies. At least one NJD was identified in 293 patients (77%), with 65 patients (17%) having only justified discrepancies, and 22 patients (6%) having no discrepancies. NJD were: different dose, route or schedule (51%), omission (39%), wrong drug (8%) and commission (2%). The variables associated with discrepancies were number of chronic medications drugs and provider of information.

Conclusions: In most studies, omission is the most frequent error. In contrast, in our study the most frequent error is different dose, route, or schedule. The variable that allows selecting patients at higher risk of discrepancies is the number of chronic drugs. This risk is also increased if the patients are not the manager of their own medication.