Defining and Applying Locally Relevant Benchmarks for the Adenoma Detection Rate. (August 2019)

• Record Type: Journal Article
• Authors: Hilsden, Robert J.; Rose, Sarah M.; Dube, Catherine; Rostom, Alaa; Bridges, Ronald; McGregor, S. Elizabeth; Brenner, Darren R.; Heitman, Steven J.
• Is Part Of: American journal of gastroenterology. Volume 114:Number 8(2019)
• Publication Date: 2019-08
• Subjects: Stomach -- Diseases -- Periodicals; Intestines -- Diseases -- Periodicals; Gastroenterology -- Periodicals; Gastrointestinal Diseases -- Periodicals; Electronic journals; Periodicals; 616.33
• DOI: 10.14309/ajg.0000000000000120
• Languages: English
• ISSNs: 0002-9270

Abstract

Abstract : INTRODUCTION: The adenoma detection rate (ADR) is the best validated colonoscopy performance quality indicator. The ASGE/ACG Task Force on Colonoscopy Quality set an ADR benchmark of ≥25% in a mixed male/female population. We propose a novel means for defining locally relevant ADR benchmarks using data from the population of interest and for applying ADR benchmarks using 95% confidence intervals (CIs) of an endoscopist's ADR. We further propose that ADR benchmarks should be raised to reflect what can be achieved by high-performing endoscopists. METHODS: We used endoscopists' performance in a baseline year to develop and apply benchmarks in an assessment year. We defined assessment year benchmarks (Minimally Acceptable, Standard of Care, and Aspirational) based on the average ADR of performance groups defined by baseline year ADR quartiles. We demonstrated the use of these benchmarks in endoscopists performing screening colonoscopies by determining if the upper bound of the 95% CI of the endoscopist's ADR included the ADR benchmark. RESULTS: The study included 8, 492 colonoscopies (mean ADR 29%) in 2014 and 5, 193 colonoscopies (mean ADR 32%) in 2015, completed at a regional screening center in Calgary, Canada. The Minimally Acceptable, Standard of Care, and Aspirational benchmarks for 2015 were 25%, 30%, and 39%, respectively. The 95% CI of the ADR of 1 (3%), 3 (10%), and 12 (39%) endoscopists did not include the benchmark. DISCUSSION: We have proposed methods for defining and applying benchmarks for ADR in average-risk patients that go beyond the "minimally acceptable" threshold currently recommended.