Data delivery for researchers

Data requests are assessed according to

• the scientific significance and social added value,
• the appropriateness of the method and strategy for identifying causal relationships
• the time and personnel required internally and
• the risks incurred.

Data delivery is excluded if

• the use is purely commercial,
• the enquiry comes from the field of market research or the pharmaceutical industry,
• data protection cannot be guaranteed
• the data is not available,
• the external project is in thematic competition with CSS Institute projects and/or
• the data is not retrospective, i.e. (multiple) data queries are necessary in the future.

In the Swiss context, health insurers do not have the following data, among others:

• outpatient diagnoses,
• undefined key figures such as the number of visits to the doctor,
• individual income,
• education and training status,
• individual details of gainful employment (employer, place of work, workload, etc.),
• accident costs of persons in employment,
• family structure,
• and nationality/migration status.

The data is particularly suitable for analysing the individual behaviour of the various players in the healthcare market. Due to the many years of insurance coverage in some cases, changes over time can be described well.

As there are other insurers in addition to CSS, the data is not suitable for overall market analyses; in particular, it cannot be used to examine the switching behaviour of insured persons. Due to the lack of diagnoses in the outpatient sector, the data is also not suitable for analysing the burden of disease or the costs of certain illnesses; accordingly, it is also generally not possible to compare the costs of different treatment options.

The data is always provided in an R data format (.rds, .rdata) and without further aggregation. We restrict the data to calendar years; we do not create dynamic time windows during the year around an individual event. Data cleansing and preparation is the responsibility of the data provider and therefore requires experience in processing and analysing very large data sets. Where necessary, data is omitted or summarised in order to ensure the anonymity of individuals.

For projects in which the insured persons give their written consent to the use of data and release CSS from its duty of confidentiality, we recommend involving us in the planning of the project at an early stage. We require a copy of the consent from each insured person and a unique identifier (surname, first name and date of birth are not sufficient). We reserve the right not to provide any data if no unique identifier is available. However, projects with patient consent are generally not our main focus for external data deliveries due to the generally small number of people and the relatively high effort involved.

All persons involved in a project undertake in writing to fully comply with the duty of confidentiality pursuant to Art. 33 ATSG and the provisions of the Data Protection Act and its ordinances (in particular Art. 1, 2 and 3 DSV). In particular, all persons undertake

• to use the data only for the project specified in the data request,
• to inform the CSS Institute of the results of the project,
• to prevent unauthorised persons from gaining access to equipment that can be used to view CSS data,
• not to pass on the data to third parties,
• to comply with data protection regulations and, in particular, not to trace anonymised data back to specific individuals,
• to make the data unreadable, destroy it or return it to CSS on request no later than three months after the end of the project, and
• to enable the CSS Institute and the CSS Data Protection Officer to review the data protection measures taken at any time (Art. 9 DSG).

If data from several insurers is used, the persons involved in the project also undertake

• not to make any comparisons between insurers.

In the event that one or more persons involved in the project fail to fulfil their obligations, the offending persons must pay CSS a contractual penalty of CHF 50,000 per event. Claims for damages remain reserved. Payment of the contractual penalty does not release any person involved in the project from compliance with the data protection declaration.

No, we do not demand any financial compensation for providing the data. However, due to our limited resources, we prioritise young researchers, who generally have more difficult access to other scientific funding due to their young careers. We also refuse to accept co-authorship of the resulting publications for pure data provision.

As the CSS Institute conducts its own scientific research with the data, we can usually deliver the data within a few weeks if we receive a confirmation. We therefore ask you to request data only when the necessary resources (personnel, IT infrastructure, etc.) are available at the data requestor to start analysing the data. With a few exceptions (e.g. applications for SNSF junior researcher funding programmes), we do not make any commitments for future data deliveries for research projects that are still in the planning phase.

In principle, it is possible to request data from several insurers. However, there is currently no standardised process, i.e. the request is processed differently by each insurer. The information provided here can therefore not be generalised. As we have experience with scientific projects, we recommend that you contact us first.