The research team, led by Robin Schepers and colleagues, linked national Norwegian registry data covering the period from 2008 to 2021. Rather than following a random sample of the general population, they focused on adults who were assigned a new GP: the closure or reduction of a GP’s patient list, or immigration.

For each GP, the researchers constructed two separate measures of diagnostic tendency. One measure captured how often that doctor diagnosed depression or anxiety among patients seen before a given index patient was assigned to them, and the other captured the same rate among patients seen afterwards. The team then used ordinary least squares regressions with fixed effects to check that these tendency measures genuinely reflected variation in diagnostic behaviour, before estimating how a GP’s tendency related to psychiatric and somatic healthcare use and to sickness absence over a three-year follow-up period.

The first finding was that diagnostic tendency varied moderately across GPs. This was not a marginal effect buried in statistical noise. When the researchers compared GPs one standard deviation above versus below the average tendency, the current measure corresponded to roughly 27 additional depression or anxiety diagnoses per 1,000 newly assigned patients, while the prior measure corresponded to about 13 additional diagnoses per 1,000 patients. In regression terms, a one-percentage-point increase in a GP’s diagnostic tendency was associated with a 0.47 percentage-point increase in the probability that a patient received a diagnosis using the current measure, with a 95 percent confidence interval of 0.44 to 0.49, and a 0.32 percentage-point increase using the prior measure, with a confidence interval of 0.29 to 0.34. The fact that both the before and after measures pointed in the same direction strengthened the case that the variation reflects stable tendencies in how individual GPs diagnose, rather than one-off quirks of particular patient encounters.

The second and arguably more consequential finding was that this diagnostic variation was linked to what happened to patients afterwards. A five-percentage-point increase in a GP’s diagnostic tendency was associated with a 0.4 percentage-point increase in the use of non-acute psychiatric outpatient care, with a confidence interval of 0.2 to 0.5, and a 0.3 percentage-point increase in repeated sickness absence, with a confidence interval of 0.1 to 0.4. Translated into absolute numbers, this corresponds to roughly two to six additional psychiatric outpatient care cases and three to four additional repeated sickness absence cases per 1,000 patients per year. The researchers report that these associations were consistent across time and across the different tendency instruments they used.

Not every outcome moved in lockstep with diagnostic tendency. Associations with other measures of healthcare utilisation, including somatic care, were smaller and statistically imprecise, meaning the data could not rule out effects of zero for those outcomes. The pattern suggests that the downstream consequences of a GP’s diagnostic style are concentrated specifically in mental health-related services and in sickness absence, rather than reflecting a general shift in how often diagnosed patients engage with the healthcare system overall.

The study’s authors are careful about what these numbers can and cannot tell us. The design links a GP’s tendency to outcomes among newly assigned patients, and the fixed-effects approach helps control for stable differences between patient groups. Even so, the associations reported are observational in nature, and the authors themselves state that whether the higher rates of diagnosis, psychiatric outpatient care and sick leave associated with more diagnosis-prone GPs are beneficial or harmful for patients remains unclear.

What the study does establish is that the gatekeeping role of general practitioners produces measurable, patient-relevant consequences. Two adults with similar symptoms who are assigned to different GPs in Norway face different probabilities of receiving a depression or anxiety diagnosis, different likelihoods of entering non-acute psychiatric outpatient care, and different chances of repeated sickness absence over the following three years. The researchers conclude that GPs vary in the extent to which they diagnose these conditions and that this variation affects patients’ likelihood of psychiatric care use and sick leave, while calling for further investigation to determine whether the observed patterns ultimately help or harm the people behind the registry entries.

Subject of Research: Variation in general practitioners’ mental health diagnostic tendencies and subsequent patient care: a Norwegian registry study

Article Title: Variation in general practitioners’ mental health diagnostic tendencies and subsequent patient care: a Norwegian registry study

Article References: Schepers, R., Bennett, S. M., Anthun, K. S., Asheim, A., Bjerkeset, O., Bjørnelv, G. W., Bjørngaard, J. H., & Edwards, C. H. (2026). Variation in general practitioners’ mental health diagnostic tendencies and subsequent patient care: a Norwegian registry study. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15516-9

Image Credits: AI Generated

DOI: 10.1186/s12913-026-15516-9

Keywords: Variation, general, practitioners, mental, health, diagnostic, tendencies, subsequent, patient, care, Norwegian, registry