Statistical processing

Source data

The primary data sources are the national Catalogue of Health Organisations, the register of primary public health services statistics and the register of population statistics. The sources can be divided into internal Statistics Denmark sources and external sources:

The National Catalogue of Health Organisations (SOR) is an organizational register, that amongst others include information on organizational and address data in the health sector, hereunder general practitioners. Data is collected manually from The National Health Data Authority.

Internal Statistics Denmark sources include: - The register of population statistics - The register of primary health services statistics - The statistics on Urban Areas

Frequency of data collection

Annually

Data collection

Data are collected directly from administrative registers

Data validation

The national Catalogue of Health Organisations (SOR)

When first collecting SOR a validation is made to check the coverage of provider ids of general practitioners.

1. Provider ids valid in the year of measurement valid in SOR is checked against a list of active providers collected from MEDCOM.

2. Valid provider ids in the year of measurement in SOR is checked to see if they actually provided services in the national health services statistics registry. A few cases has been made where a non-valid provider id is providing services. In those cases it was found that the provider id was closed in SOR the year before the year of measurement, and services occur in the year of measurement due to financial reasons. In those cases the closed provider id is added to the list of valid provider ids.

3. It is checked if the provider id, holds multiple addresses in SOR. In the special case where provider id is attached to more than one address, the address in the list of valid provider ids from MEDCOM is chosen.

4. It is checked if the provider address is valid according to the source data from the Urban Areas statistics. In cases where an address is not valid it is corrected manually. Cases of spelling mistakes where found in addresses in SOR, and corrected by Statistics Denmark in order to obtain the correct addresses.

Valid home address

From address information knowledge of invalid addresses are collected and used to identify persons living on invalid addresses.

Data compilation

The National Catalogues of Health Organisations is delimited to hold information on active providers in the year of measurement.

On the basis of the data validation, described in chapter 3.4 a number of adjustments are made: - A few cases has been made where a provider id, that is in-active in the measurement year in SOR, is found providing services in the National Health Insurance registry. This might occur due to financial reasons. In these cases the closed provider id is added to the list of providers in the measurement year. - In the special case where a provider id is attached to more than one address, the address in the list of valid provider ids from MEDCOM is chosen. - I cases where an address (due to misspelling) is not valid it is corrected manually. Cases of spelling mistakes where found in SOR, and corrected by Statistics Denmark.

National health insurance: From the statistics of National health insurance only services provided by a general practitioner is selected. The selection varies from that of the statistics "Health Insurance Statistics" and as a result counts may vary

Population: These statistics contains the entire population as of January 1st in the measurement year. However there's a number of circumstances in which a person might not have a valid address and for whom it is not possible to calculate distance to general practitioner. Expatriates with the right to be registered as citizen being one, and people being homeless an other.

Coordinates to distance calculation: To ensure uniformity in coordinates between general practitioners and living addresses data from the Urban area Statistics are used. This also ensures that those addresses manually corrected get the correct addresses.

Coordinates are also found in SOR, and has proved to be identical with the coordinates from the back ground data from the Urban area Statistics.

Coordinates used are projection in numbers and displayed as integers.

Connecting inhabited addresses with addresses of general practitioners When the different data sources have been collected, cleaned and coordinates has been attached, all inhabited addresses are crossed with addresses of general practitioners in the region of living. Hereafter the bird's eye view distances are calculated.

Calculating the distance: To calculate the precise distance between inhabited address and general practitioner a road map and the GIS-program ArcGis is used. In this way it is taken into account that the closest general practitioner might not be that having the shortest bird's eye view distance. This method of calculation also takes into account that the shortest distance might include bridges and ferries. In the calculation of distance measured in roadmap only the 3 closest general practitioners measured in bird's eye view are included. In some circumstances the 3 closest general practitioners measured in birds eye view are not the closest measured in the roadmap. However in some circumstances - especially around people living on islands the closest general practitioner measured in birds eye view with in the region is not the general practitioner measured in road distance. For this reason the 3 closest doctors within the municipality are included in the calculation if they are not already included by the first condition.

When the distances have been calculated they are merged from residential addresses to the population, in order to have the distance for every person in the population.

The distance between general practitioner and persons with invalid residential address (as described above) addresses is not calculated. Further no distances are calculated for persons living on the island Christiansø. Lastly distances are not calculated in the event the residential address is on an islands with out bridge- and ferry-connection. In the measurement year 2024 the number of persons for which no calculations are made is app. 15.000 out of 5,9 million inhabitants. The group for whom distances are not calculated are placed in a combined municipality group. The total number of persons on a municipality level in this statistics are hence not comparable to that of the populations statistics on municipality level, eg. FOLK1AM.

Average distance: When calculating the average distance only the shortest distance, measured in roadmap, for an individual to general practitioner is selected.

It's important to bear in mind that the general practitioner closest to an individual might not be the same general practitioner the individual actually consult. The experienced distance for an individual might be longer than that used in these statistics.

** Individuals with contact to general practitioner: ** Using the National Health registry with the selections described earlier in this section the number of visits to general practitioner is calculated and merged onto the population. Hereafter the population is devided into two groups. A group with and a group without contact to a general practitioner within the measurement year.

Adjustment

No other corrections are made than that described in data collection and data validation.