Use of Electronic Health Records in the PRIMEGE PACA project. Main reasons of encounter, diagnoses and prescriptions in primary care

Background. In France, there is no public information system associating reasons for consultation, acts performed, or decisions made during the session. This situation impelled us to create the data collection PRIMEGE PACA (Plateforme Regionale d'Information en MEdecine GEnerale en Provence-Alpes-Cote d'Azur). Its purposes are the improvement of existing professional practices and academic research. Objectives. To assess the feasibility of creating a database from the Electronic Health records (EHRs). To provide examples of the use of this database by describing the main reasons for consultation, diagnoses and drug prescriptions. Methods. EHRs of 11 general practitioners were collected and imported into a MySQL (R) database. To overcome the low coding rate of the physicians, a procedure allowing translation free text into ICPC-2 was implemented. Results. The medical records of 11 general practitioners over 4 years (2012 to 2015) were extracted representing: 205 343 consultations. The automatic coding procedure allowed us to associate main reasons for consultation and diagnoses with an ICPC-2 code (about 97% of each). The main reasons were respiratory (23.5%), osteoarticular (17.7%), general (14.1%) and digestive (11.9%). Diagnoses were mainly osteoarticular (19.1%), respiratory (16.9%), digestive (9.5%) and cardiovascular (9.2%). Prescriptions concerning the nervous system (22.2%), the digestive system and metabolism (15.8%), the cardiovascular system (12.5%) and the respiratory system (12.2%) were the most numerous. Conclusion. This study has shown that it is possible to create a database from electronic health records. New software is being studied to be integrated into the database and will allow inclusion of new physicians. This regional model could be configured in a network of regional observatories providing a representative panel of general practitioners ``Producers'' of data