Potenciais interações medicamentosas graves no ambiente hospitalar: Validação de mensagens de alerta para uso em software de prescrição eletrônica

Objective: The present study has as its main objective the validation of alert messages, for prescribers, about possible serious drug interactions. Methods: This is a descriptive documentary study, using a hospital’s medication list. The first step was identifying and analyzingpotential serious drug interactions through the database on the website Drugs.com among the drugs on this list. The interactions identified were categorized into Medicines; Mechanism; Recommendations, and Action. The second step was the validation of the selected interactions. The validation was carried out using the Delphi Technique, in which the interaction bank was sent with a questionnaire to assess content and clinical utility for Brazilian professional specialists with expertise in the area. In the end, he tried to obtain at least a Content Validity Coefficient (CVCc) of at least 0.8 of agreement between the judges. Results: After the analysis, 330 possible serious drug interactions were selected in which they were validated. Regarding evaluating the interaction alerts content, all the questions presented CVCc equal to or greater than 0.8 during the study. All 37 judges responded and participated in the validation process, who suggested adequacy of language, insertion of new interactions and standardization of messages. In addition, the jury was composed of a group of heterogeneous people, with several training areas, the highest proportion composed of masters (43.24%) and PhD (18.91%). Conclusion: The database was validated with a CVCc of at least 0.8 of agreement between the judges, considering the high prevalence of possible serious drug interactions, both in the hospital and in the outpatient setting. It can be implemented in electronic prescription software, helping to reduce the prevalence of these potential serious drug interactions and contributing to patient safety.