A computerized pharmaceutical intervention is shown to reduce reconciliation errors in the context of a high incidence of such errors.

Primary Care computerised medication records, although of undoubted interest, are not be reliable enough to be used as the sole source of information on patient chronic medications when admitted to hospital.

BackgroundThe thrombin inhibitor dabigatran (D) is the first new oral anticoagulant approved in Europe for the prevention of non-valvular atrial fibrillation; its advantage is that it has less interactions that antagonists of vitamin K.PurposeThe aim of the study was to determine the prevalence and type of potential drug interactions (PDI) in the treatment of patients with D in a health area, and to analyse the possible clinical relevance of these.Material and methodsThe study was performed in a health area serving 194 737 inhabitants for 6 months (July–December 2014). We included all patients treated with D and recorded demographic data and the full treatment prescribed for each patient to identify PDI, which were obtained from programs prescribing and dispensing primary care (ADN and Agoraplus) and managing medication dispensed in hospital (SAVAC). We considered PDI as those described in the technical data and classified according to the mechanism and recommendation indicated. Finally, we estimated the potential clinical relevance of the presence of PDI based on: visits to the emergency department (per patients and average/patient), hospitalisations and diagnoses in emergencies related to an adverse effect to D.ResultsWe included 206 patients treated with D (56% women, mean age 76.8 ± 8.6 years). 128 PDI were recorded in 50.5% of patients, with an average per patient of 1.24 ± 0.53 (75.3% for 1 interaction, 18.6% for 2, 6.2% for >2). 25.8% were pharmacokinetic and 74.2% were pharmacodynamics. In 11 interactions (8.6%), co-administration was contraindicated, in 86 (67.2%) it was necessary to monitor and in 31 (24.2%) the dosage was reduced and track performed. The drug groups involved in the PDI were: 7.8% NSAIDs; 25.8% inhibitors of P-glycoprotein (IGP-P), dronedarone, amiodarone, verapamil, etc; 30.5% antiplatelet drugs; 28,9% SSRI/SNRI; and 7.1% anticoagulants. We did not find significant differences in any of the relevant clinical variables studied between patients with and without PDI.ConclusionA considerable proportion of patients (50.5%) presented PDI in treatment, but without apparent clinical relevance to serious adverse events.The majority of PDI were pharmacodynamic and could be sought to improve the therapeutic effect. However, the significant percentage of PDI with SSRIs suggests that they may be unknown by some prescribers; there is a need to monitor their use along with inhibitors of IGP-P which are often prescribed to these patients.References and/or AcknowledgementsThanks to the documentation department.No conflict of interest.

No abstract