Statistical processing

Source data

The primary source is LUNA, which is the IT system used by the regions to process reimbursements for health insurance services. Additionally, there are supplementary sources regarding services from the tariff folders.

Internal Statistics Denmark sources:

• The register of population statistics (ancestry)
• The register of income statistics (level of income) for the previous year
• The Employment Classification Module (SOCIO13) as of December 31st of the previous year – Education (BUE): Highest completed level of education as of 30 September of the previous year.
• The Relatives Register (resident parents, relatives, partner)

Frequency of data collection

Yearly.

Data collection

The data collection is based directly on administrative registers as well as information from fee schedules and collective agreements available on ok-portalen, which consolidates collective agreements and contracts within the healthcare sector.

Data validation

The data received are compared with data from the previous year, and any major fluctuations examined to reassure quality. For the purpose of statistical production data are analyzed thoroughly.

Data compilation

Danmarks Statistics determines, based on the tariff catalogues, whether a reimbursed service should be counted as a contact. Services that are not specified as supplementary services in the tariff catalogues are registered as contacts. For services that do not appear in the tariff catalogues, the assessment is based on whether the service can only be provided once per contact with the provider and whether it precludes other services that can be registered as contacts from being provided simultaneously. For specialties 49 (dental hygienist) and 50 (dental care), consultations and examinations, including preventive treatment, are categorised as contacts. The number of contacts is calculated as the sum of the variable “number of services” for the services categorised as contacts. A different method is used for diagnostic radiology (05) and chiropody/podiatry (54, 55, 59, and 60). Here, the service with the lowest service number (the last four digits of the service code) per person per treatment date is registered as the contact. A person can therefore only have one contact per day within these specialties in the statistics. This method was chosen because it is considered to provide the best estimate of the number of contacts. However, this means that a service within these specialties may be categorised both as a contact and as a non-contact.

Starting from 2006, gender and age imputation has been performed for the smaller group of children registered with child marking, and the person_id is set to unknown. Basic data (SSSY) is generated from the above data, and basic data is created, where the number of contacts (SSKO) and gross fees (SSHO) are aggregated at the person and service level. In these data, spec2 for contacts with general practitioners is further divided into the following categories: daytime consultation, evening consultation, daytime telephone consultation, evening telephone consultation, daytime visit, evening visit, e-communication (including with municipal care staff), other services, and prevention, etc. Additionally, an indicator for the basic fee is calculated. This is calculated as the sum of the calculated basic and practice cost fees distributed among group 1 insured persons who have received services from general practitioners (excluding persons marked as children). For SSSY, a variable spec80 has been introduced from 2023, which indicates the above division. However, "Other services and prevention, etc." does not appear in SSSY, but the algorithm can be obtained by contacting the statistical responsible party. Likewise, the basic fee is not specified in SSSY.

Before the data is uploaded to the Statbank Denmark, further data processing takes place. Individuals with invalid CPR numbers, individuals with child marking, and corrections (negative values for the service) that cannot be associated with a registration in the respective fiscal year are deleted. This means that corrections (negative entries) that do not match any registrations on the variables person_id, 'date of treatment' (date of service), 'specialty' (type of service), and "ydeltid" (time of service) are excluded. Registrations where the variable 'SIKGRUP' (Health Insurance Group) has the value 9 (Deceased) are deleted. All deleted observations are assigned a value of 0 for the variable statpop in SSSY. The first registered gender and the first registered age are used for individuals who appear in the data with multiple CPR numbers. Health insurance data is linked to other data on family relations, origin, education, labour market attachment, and income.

Adjustment

From 2005, the register was cleansed of observations for which there are no reimbursements via the public health insurance (the gross fee equals 0). This applies primarily to physiotherapy and dental treatment. Accordingly, data is assessed for 2005 both by the old method of assessment by which data is not cleansed, and the new assessment method by which data is cleansed. The chronic care honoraria (0130, 0131, 0132, 0133) are not deleted for specialty 80 (general practice). As of 1 January 2024, the structured care pathway for patients with multiple sclerosis was implemented under specialty 62. The new pathway services have the service codes 0211, 0212, 0213, 0214, and 0215.As of 1 January 2025, structured care pathways for osteoarthritis of the knee/hip have also been implemented, with service codes 0221, 0222, 0223, 0224 and 0225. These services appear in the data with a gross fee equal to 0. None of these services are deleted.

There is a very small number of records where the contacts are negative. In 2025, there are 393.474 negative records (equivalent to 0.4pct. of all records). This is due to billing-related corrections in the registry, meaning corrections that are not made by Danmarks Statistik. Starting from 2021, corrections that cannot be linked to a registration in the respective year will be deleted before the compilation of the statistics bank.