The variability in interpretation of colonic codes in CT colonography reporting: a single-centre experience

Although standardised summary codes to classify colonic findings (C-codes) on computed tomography colonography (CTC) have been used for several years, there is no clear guidance on how these codes should be interpreted. The aims of this study were to (1) establish CTC C-code demographics and reporting practice at our hospital and (2) determine the agreement between CTC reporters when using C-codes.

Waiving ethical approval, this online questionnaire study invited all radiologists, reporting radiographers and radiology trainees who reported CTC at our hospital between 22/02/2023 and 05/03/2023. In total 20 questions were developed with 9 questions on demographics and reporting practice followed by 11 case scenarios. Agreement between participants for the case scenarios was calculated using Fleiss kappa and mean pairwise agreement.

18/21 (85.7%) of participants completed the questionnaire. The majority of respondents reported using C-codes "always" (17/18, 94.4%). Overall agreement for the 11 case scenarios was fair [0.39 (95% CI 0.38-0.41)] with a mean pairwise agreement of 46.9%. Agreement was significantly greater for reporters with ≤ 1000 than > 1000 CTC experience (p < 0.001), those who reported diminutive polyps than those who did not (p < 0.001), and adequate than inadequate case scenarios (p < 0.001).

This questionnaire study demonstrates variation with how C-codes are interpreted at our institution. We suggest a national survey to determine whether this is a widespread issue and to inform development of formal implementation guidance within the UK Bowel Cancer Screening Programme.