Statistical processing

Source data

The data basis for the statistics is based on extracts from the Municipal Sickness Benefit System, where all information from a case is sent to Statistics Denmark if there have been changes since the previous data delivery.

Frequency of data collection

Monthly

Data collection

Electronic transfer of administrative data.

Data validation

When the data is received, all fields are checked by machine and the number of records received is compared at entity level with corresponding figures from a control table that is part of the data delivery. It is checked that there is data from all municipalities.

Data compilation

Data is received directly from the administrative sickness benefit system's data model, which means that a data delivery consists of 32 tables used to update as many registers called source registers. Every quarter, data is extracted from the source registers and merged together to form the integrated sickness benefit register, which as of May 2025 consists of just under 44 million observations and 55 variables plus various metadata variables. The purpose of the integrated register is to make it easy to perform counts on sickness benefit data, which is why all information, regardless of type, in the registers is embedded in a standardised way so that all records have the same structure. In this connection, it has proved necessary to make corrections in cases where different information for the same case has overlapping periods. Holidays that fall on one of the first five days of the week (public holidays) are artificially counted as sick days. In cases where there is no information on the number of hours of sick leave for a given person in a given week, the hours are calculated based on the amount paid out, and if this is not possible, the hours are imputed from other observations from the case in question.

Adjustment

Nothing to note