Two-round Delphi technique for the consensual design of a paediatric pharmaceutical care model.

The main goal of clinical pharmacists is to improve patient care quality by providing individualised care. This is achieved by validating prescriptions and performing pharmacotherapeutic follow-up, leading to effective, safe and efficient drug therapy. However, there is no specific model for medication order validation in paediatrics addressing the distinct issues of paediatric drug use or the greater susceptibility of medication errors occurring in this patient group. For this reason we intended to design and reach consensus on a paediatric pharmaceutical care model by applying a two-round Delphi technique. In order to define the levels of complexity in reaching consensus for the model, three variables were taken into account: level of access to patient medical and drug records (partial vs. total), access to medical and nursing staff attending the patient (on-site vs. off-site) and the pharmacist available time (limited vs. adequate). In order to describe the minimum items to be included in the validation process at each of the three levels of complexity, we analysed the safety profile in the medication order prescription previously defined in an epidemiological study in eight hospitals with a total capacity of 1565 paediatric beds. A panel of experts was recruited (50 experts from 20 different hospitals) and the questionnaire was completed. Consensus was established at 70% of agreement by experts for an item. Following debate on the items that did not obtain consensus, a second round was performed, after which the final consensus model was defined. After two rounds of consultation, consensus was obtained for 39 out of the 41 items that were surveyed. Of these, 17 were then used for the basic validation model (e.g. weight/age check, dose/weight check), 13 were implemented in the intermediate level (e.g. identification of adverse effects from excipients) and 9 were incorporated at advanced level (e.g. reconciliation at discharge for patients at risk). By applying the model, based on homogenous criteria for action, the clinical pharmacist's role will improve, and in turn, this will doubtlessly reduce drug errors through medication order validation.