07 Routine data from primary care practices serve to improve the healthcare system in Switzerland

New GP practices were acquired for participation in the project and the necessary technical interfaces were developed. Furthermore, classification systems for the recording of 50 chronic diseases and a semi-automatic language processing algorithm for the categorisation of laboratory values were implemented. In addition, the requirements for data entry were developed based on a literature analysis and discussions with GPs. Subsequently, user training was conducted in the GP practices and its effectiveness evaluated. After the implementation of the system, the participants were divided into two groups that differed in the amount of compensation they received for participating in the project. By comparing the two groups, the optimal compensation for systematic data collection in GP practices was determined.

Routine data are an important basis for investigating questions about the health status and treatment of patients. However, the data basis in the outpatient sector in Switzerland is weak, fragmented or inaccessible for research. The most comprehensive database in the general practitioner (GP) sector in Switzerland - FIRE - has existed for around ten years and had 132 participating GP practices in 13 cantons at the start of the study in 2017.

The aim of the study was to recruit further GP practices for the standardised collection of routine data in the FIRE database and to determine the optimal level for their financial compensation. In addition, the quality of the data was to be checked and improved if necessary. In this way, a representative database was to be created in the GP sector in Switzerland.

The FIRE project established the first database in Switzerland with routine clinical data on outpatient care. In the course of the project, the number of participating doctors tripled: Around 700 GPs from over 200 German-speaking practices exported anonymised clinical routine data from their electronic medical records to the FIRE database. This corresponds to about ten percent of the GPs working in this field in Switzerland. Currently, the FIRE database contains over nine million consultation records with administrative information, diagnosis codes according to the International Classification of Primary Care 2nd edition (ICPC-2), laboratory and vital sign measures as well as medication plans and treatment regimens. This semi-automated process increases the quality of the available data and keeps the database constantly up-to-date.

Significance of the results for research:

Thanks to the expansion of the FIRE database and the improvement of data quality and availability, a sound and reliable database for health services research has been created. Routine medical data can be increasingly used secondarily and numerous health services research projects can now be realised.

Significance of the results for practice:

The FIRE database can further be used for quality assessments or the development of electronic decisions support tools, for example.

Establishing a nationwide collection of medical routine data in primary care – expanding the FIRE project