TOPLINE:

An analysis of over 61,000 imaging requests in German academic hospitals found that the odds of an inappropriate imaging request were 22% higher among women than among men, highlighting a small but consistent sex-based disparity in diagnostic decision-making.

METHODOLOGY:

• Researchers analysed data from the MIDAS trial, a prospective, multicentre observational study conducted across 26 departments in three German academic hospitals.
• They analysed all imaging requests submitted through computerised physician order entry systems, and each imaging request was scored for appropriateness using the European Society of Radiology (ESR) iGuide system; 61,220 requests received valid scores.
• The iGuide system then categorised these 61,220 requests as appropriate, conditionally appropriate, or inappropriate; physicians were blinded to appropriateness scores during the baseline phase.
• The primary outcome was the proportion of inappropriate imaging requests among women and men; odds ratios (ORs) were calculated with 99% CIs.
• Stratified analyses examined sex-based differences across age groups and imaging modalities, including CT, MRI, x-ray, and ultrasound.

TAKEAWAY:

• Inappropriate imaging requests comprised 6.70% of total requests, with women showing a significantly higher rate than men (7.32% vs 6.08%; OR, 1.22; 99% CI, 1.12-1.33; P < .001).
• MRI and the age group of 25-65 years demonstrated statistically significant differences in inappropriate imaging request rates between women and men (P < .001).
• The percentage of women was lower among unscored requests than among scored requests (47.9% vs 50.7%; P < .001), and the mean age was higher (63 vs 61 years; P < .001).
• All imaging modalities except CT showed higher proportions of inappropriate imaging requests among women, although not reaching the stringent statistical significance level.

IN PRACTICE:

"This study found that women were more likely than men to have inappropriate imaging ordered for them, highlighting a small but consistent sex-based disparity. Addressing these gaps will require further research and more sex- and gender-sensitive approaches in clinical decision-making and guideline development to ensure equitable imaging practices," the authors wrote.

Acknowledging that their analysis was limited to binary sex, they added, "future work should aim to incorporate more inclusive data collection and explore how patient and provider characteristics intersect to shape diagnostic pathways."

SOURCE:

This study was led by Stijntje Willemijn Dijk, Erasmus MC University Medical Center, Rotterdam, Netherlands. It was published online on November 06, 2025, in European Radiology.

LIMITATIONS:

Nearly half of all requested examinations remained unscored by the ESR iGuide system, which represented a limitation of the clinical decision support system used. The analysis was limited to binary sex data, preventing the exploration of disparities for gender-diverse individuals, and did not capture the sex or gender identity of the requesting physician. The study evaluated each imaging request as a discrete event, meaning that it could not account for the cumulative context of the trajectory of the patient or assess potential imaging underuse.

DISCLOSURES:

This study received funding from the German Innovation Fund. Several authors reported receiving research funding, honoraria, reimbursements of expenses, and royalties from various bodies, including the Gordon and Betty Moore Foundation, the Bavarian Centre for Cancer Research, the national Network University Medicine, Cambridge University Press, the American Diabetes Association, and Siemens Healthineers. Full details are provided in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.