Main measures
The data were collected from April to August 2017. Before data collection, two study coordinators (L.M.C.S. and C.C.S.) trained a team of research assistants (A.S.D., G.A.C.C., L.A.M., R.O.S.S. and T.S.A.) to properly apply the data collection form and clarified their questions. The research assistants were pharmacists and researchers, members of the same research group. All medical records were assessed by the team. Cases of divergence regarding the communication failure were resolved by consensus among them.
All data were manually reported in paper medical records which contained all medications prescribed during the hospitalization. The clinical notes made by pharmacists, physicians and nurses, and healthcare professionals involved in the use of medications, were examined at three transition points of care (hospital admission, during the hospital stay, and hospital discharge).
At hospital admission, data were collected from admission forms or, when they were not present, from the first medical note in the patient medical record. Pharmacists and nurses’ notes from the first day of hospitalization were also considered as sources of information for hospital admission as well the physician’s notes. We considered admission as the first day of hospitalization, and hospital stay from the following days. During the hospital stay, data were extracted from medications prescriptions, pre-anesthetic evaluations, forms requesting other expert opinions and the pharmacists’, physicians’, and nurses’ notes. At hospital discharge, only information presented in the hospital discharge form was evaluated or, when it was not present, the last registered clinical note of a pharmacist, physician, or nurse was considered.