Material and method
The material comprises data obtained from all electronic reimbursement claims submitted by doctors on out-of-hours duty in Norway in the period 2008–18, previously used to compile the annual statistics for the emergency primary health care service (1). Anonymised computer files were obtained from the KUHR database (Control and Payment of Reimbursement to Health Service Providers) held by the Norwegian Health Economics Administration.
Consultations and home visits (tariff codes 2ad, 2ak, 2fk, 11ad and 11ak, hereafter referred to as 'consultations') were recorded for Christmas Eve, New Year's Eve and the last Saturday of January (3). Saturday was chosen because it is followed by a non-working day, the same as Christmas Eve and New Year's Eve. GP surgeries are therefore closed and patients directed to the emergency primary health care unit instead of waiting for GP surgeries to open the next morning. The last Saturday of January was chosen because this is a period with no special celebratory events, and in order to enable comparisons to be made for the same time of year.
The variables used were the time of contacting the service, diagnostic codes (ICPC-2) and tariff codes. Times of contacting the service were grouped in the categories of daytime (12.00–17.59), evening (18.00–23.59) and night-time (00.00–05.59). In order to enable valid comparisons to be made, mornings before 12 noon were excluded, because many GP surgeries are open until noon on Christmas Eve and New Year's Eve, but not on Saturdays. Night-time was defined as the hours after 12 midnight on 24 or 31 December, or the early hours of Sunday morning. In the interest of securing robust and stable figures, numbers for the years 2008–2018 were added up to form a total. Injuries are defined and categorised in accordance with the Norwegian Institute of Public Health's report on accidents and injuries in Norway (4).
The annual statistics for the emergency primary health care service were assessed by the Norwegian Labour and Welfare Administration's data protection officer and by the Data Protection Official for Research (1). Because it is impossible to identify individuals on the basis of the material, be it directly or indirectly, the project is not subject to the notification requirements imposed by the Norwegian Personal Data Act.
Because the material comprises all the electronic reimbursement claims rather than a sample, the differences demonstrated are real and not encumbered by statistical uncertainty. The data are therefore presented without confidence intervals, and no statistical tests have been conducted.