Validation study reveals moderate sensitivity for colonoscopy registration in screened individuals, highlighting major underreporting among nurse-led procedures.
Background Accurate registration of colonoscopies in registers is essential for register-based research of colonoscopy use and outcomes. We aimed to assess the sensitivity of colonoscopy registration in the Swedish National Patient Register (NPR) compared to colonoscopy data from the SCREEning of Swedish COlons (SCREESCO) study.Material and methods All individuals who underwent a colonoscopy through SCREESCO were included in the current analysis at the date of their first SCREESCO colonoscopy. Colonoscopies registered in the NPR on the same date as the SCREESCO were identified and we estimated the sensitivity of finding a SCREESCO colonoscopy in the NPR overall and according to calendar year, profession (gastroenterologists, surgeons, and endoscopy nurses) and health care regions, with 95% confidence intervals (CIs).Results The overall sensitivity was 64% (95% CI: 63-65%, n = 10,638/N = 16,555), more than twice as high among gastroenterologists (76%, 95% CI: 76-77%, n = 7240/N = 9473) and surgeons (87%, 95% CI: 86-89%, n = 1739/N = 1990) than among nurse endoscopists (32%, 95% CI: 30-33%, n = 1619/N = 5092), and varied across regions and calendar time.Conclusion The trend of increasing number of colonoscopies performed by nurse endoscopists may cause major underreporting in the NPR.
No takes yet. Share an insight, caveat, or question.
Westerberg et al. (2026) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: