Inter- and Intra-Observer Reliability for Colon Capsule CLEansing Assessment and Report

A validated scoring system for reporting of colonic cleanliness at colon capsule endoscopy (CCE) is not routinely used to date and great inter-observer variability exists. To ensure appropriate and standardised bowel cleansing reporting, a validated scoring system is of great importance. This would improve the quality of colon capsule reporting and facilitate optimisation of bowel preparation protocols by supporting research reliability. A novel system named Colon Capsule CLEansing Assessment and Report (CC-Clear) has shown superior inter- and intra-observer agreement to the Leighton-Rex scale, an earlier scoring system from 2011. This was a retrospective, single-centre study assessing the intra- and inter-observer reliability of CC-Clear. Materials and methods Studies were identified retrospectively from the capsule database of a tertiary referral centre. 30 consecutive complete CCE were included. Three expert capsule readers reviewed and reported each examination twice. Results and conclusion The inter-observer agreement was reported as Kendall’s W and was 0.8, p< 0.01. An intra-rater reliability analysis was performed between the first and second readings of the three readers. The weighted Cohen's Kappa showed that there was moderate agreement between the readings of Reader A with κ= 0.52 and p <0.001. For Reader B, the weighted Cohen's Kappa showed moderate agreement between first and second readings, κ= 0.5 and p<.001. For Reader C, the weighted Cohen's Kappa revealed almost perfect agreement between reading with κ= 0.8 and p<.001. CC-Clear is a simple tool for the CCE reader to utilise, and has been shown again here to have excellent inter- and intra-observer reliability.