Inter-observer and intra-observer agreement in drug-induced sedation endoscopy — a systematic approach

Abstract Objective To evaluate the feasibility of systematic observer training in drug-induced sedation endoscopy (DISE) interpretation. Methods Fifty DISE videos were randomly selected from a group 200 cohort patients with OSA and/or snoring. The assessed blindly and independently using modified VOTE classification by an expert two novice observers starting their training. A approach was initiated. Clusters 10 scored individually each then re-evaluated as clarified decision-making plenum. Kappa coefficient (κ) calculated measure agreement. Results intra-observer variation for total agreement kappa values ranged between moderate to substantial classification, whereas novices varied “less than chance agreement” “moderate agreement.” inter-observer showed increased day 1 2 both except at velum level C. site also improved compared results 1, seemed be more difficult evaluate. learning curve during study course upper airways. Conclusion This shows that intensive is feasible, although not acquired after 100 evaluations. but differed trainees. Level evidence