Non-indicated vitamin B- and D-testing among Dutch hospital clinicians: a cross-sectional analysis in data registries.

Objectives: To assess the extent of non-indicated vitamin B- and D-testing among Dutch clinicians and its variation among hospitals.

Design: Cross-sectional study using registration data from 2015 to 2019.

Participants: Patients aged between 18 and 70 years who received a vitamin B- or D-test.

Primary And Secondary Outcome Measures: The proportion of non-indicated vitamin B- and D-testing among Dutch clinicians and its variation between hospitals (n=68) over 2015-2019.

Results: Between 2015 and 2019, at least 79.0% of all vitamin B-tests and 82.0% of vitamin D-tests lacked a clear indication. The number of vitamin B-tests increased by 2.0% over the examined period, while the number of D-tests increased by 12.2%. The proportion of the unexplained variation in non-indicated vitamin B- and D-tests that can be ascribed to differences between hospitals remained low. Intraclass correlation coefficients ranged between 0.072 and 0.085 and 0.081 and 0.096 for non-indicated vitamin B- and D-tests, respectively. The included casemix variables patient age, gender, socioeconomic status and hospital size only accounted for a small part of the unexplained variation in non-indicated testing. Additionally, a significant correlation was observed in non-indicated vitamin B- and D-testing among the included hospitals.

Conclusion: Hospital clinicians order vitamin B- and D-tests without a clear indication on a large scale. Only a small proportion of the unexplained variation could be attributed to differences between hospitals.