Health centres in Sweden
An open overview of primary care across the country, built on public data. We bring together patients’ own ratings and objective accessibility into a comparison anyone can check and recompute. No unit can pay for its position.
How we do it, and what we don’t
In short: we only compute on open figures, we always show how many people responded, and we set no score when the basis is too thin. Better an honest “too few responses” than a figure that misleads.
The scores are based on the National Patient Survey 2025 (patients’ ratings per health centre) and the National Board of Health and Welfare’s waiting-time statistics for primary care (objective accessibility). Both are open sources. We combine them with a public method and adjust for how many people responded, so a handful of ratings cannot let a unit top a list by pure chance. Read how the Vårdbetyg score is calculated.
We do not measure everything. The statistics say something about patient experience and accessibility, not about the medical quality of each individual visit. The waiting-time data is also preliminary. We never promise more than the figures actually carry.
Coverage in figures
- 1 209health centres in the catalogue
- 1 200with a Vårdbetyg score
- 280municipalities
- 20regions
- 108 901questionnaires answered
The 2025 survey covers 20 regions (Sörmland is not included). 9 health centres are shown without a Vårdbetyg score because too few patients responded – see “Too few responses” below.
National averages per dimension
Patients’ answers are summarised in several dimensions, each with a score from 0 to 100. Here is the equal-weighted national average across all scored health centres in 2025. The average is also the reference each unit’s score is drawn toward when few have responded.
- Care and treatment
- 79,3
- Respect and empathy
- 83,9
- Involvement
- 80,4
- Information
- 74,8
- Continuity
- 70,4
- Accessibility
- 80,2
What “too few responses” means
If fewer than 10 patients responded at a health centre, we set no Vårdbetyg score at all. The unit is then shown as “Too few responses”. This says nothing about the quality of care, only that the basis is too thin to say anything honest. On each unit’s page we show the number of responses, the response rate and the confidence interval, so you can see for yourself how reliable each figure is.
Explore further
- Find your health centreSearch in your municipality and rank by what you prioritise.
- Compare health centresPut units side by side, dimension by dimension.
- Sweden’s bestThe top-ranked health centres in the country and the best in each county.
- MethodExactly how the Vårdbetyg score is calculated, step by step.
Download the data
The underlying data is open. You may use and share it freely, including commercially, as long as you credit Vårdbetyg as the source (licence CC BY 4.0). The source data comes from the National Patient Survey and the National Board of Health and Welfare’s waiting times and is subject to its own terms.
Two files: one row per health centre (score, dimensions, response data) and one aggregate per municipality. Dot as the decimal separator, UTF-8. The columns are described in the data dictionary that ships with the export.
The data is also available on GitHub (vardbetyg-open-data), with version history and data documentation.
Source: National Patient Survey 2025 (SKR). Open data at resultat.patientenkat.se.