From claim to evidence: a scoping review of the use of claims data for health research in Switzerland

Abstract Background Real-world data on health care utilization are increasingly important for health research and policy. While countries with national public health care systems can rely on comprehensive administrative data, such data are mostly missing in countries with decentralized health care systems such as Switzerland. Health insurance claims data may serve as a valuable alternative data source. A comprehensive overview of claims data applications in Swiss health research is lacking. Methods We conducted a scoping review mapping claims data use in Swiss health research over the last 25 years (2000–2024). We searched five databases (Medline, Scopus, CINAHL, Business Source Premier, and EconLit) and conducted a manual search of gray literature. Sources were screened independently by two researchers using CADIMA. Extraction was conducted via a standardized charting form developed in REDCap. Eligibility criteria included source language, publication type, health-related research, and Swiss claims data use. Results Of the 1922 individual records identified, 256 were included in the review. The use of Swiss claims data in health research increased by a factor of 22, from five publications (2000–2004) to 110 publications (2020–2024). Among those published in journals, 90.0% were published in international journals. Health services research (42.6%) and health economics (29.7%) were the dominant fields using Swiss claims data. Claims data served as the main data source in 74.2% of studies, with 44.9% of studies enriching the claims data with additional data sources. Patient-level data were used in 81.6% of studies. 78.9% of studies received claims data directly from health insurances, with 64.9% of those being received from Helsana Group. The most frequently mentioned limitations of the use of claims data were lack of diagnostic information (30.9%), limited population coverage (28.1%), and missing invoices paid directly by the patient (22.7%). Conclusions Claims data use in Swiss health research has grown substantially, evolving from a rarely used approach to a popular data source across multiple disciplines. Major limitations include missing diagnostic information and limited population coverage when single-insurer data are used. Ongoing efforts to address these limitations may push the boundaries of health-related research in Switzerland.