Mo1646 IS ENDOSCOPIST PERFORMANCE STABLE ACROSS COLONOSCOPY INDICATIONS AND TYPES OF COLORECTAL CANCER (CRC) PRECURSORS? IMPLICATIONS FOR QUALITY ASSURANCE

Abstract

The adenoma detection rate (ADR) during first-time screening colonoscopy is inversely correlated with the interval CRC rate, and it has therefore become the key colonoscopy quality performance metric. However, first-time screening is only one of the colonoscopy indications related to CRC prevention and control, a large procedure volume is needed for a precise ADR estimate, and ADR does not measure advanced or sessile serrated polyp (SSP) detection. To explore correlations across detection rates for any and advanced adenomatous polyps and SSPs, and the range of colonoscopy indications related to CRC control. The Stanford Colonoscopy Quality Assurance Program database was queried for Oct,2017–Aug,2019. The database relies on standardized procedure report documentation and standardized pathology result letters. Data for 10 endoscopists whose metrics were not reliable because of known documentation deficiencies were excluded. Correlations (Pearson correlation coefficient, r) were explored between 1) adenoma detection rates during colonoscopies for first-time screening and colonoscopies for: a) screening not first-time, b) post-polypectomy surveillance, and c) family history of CRC; and 2) detection rates at colonoscopies for all these indications combined between any adenoma and: a) SSP, b) advanced adenoma, and c) advanced SSP. A minimum of 10 procedures per category were required for inclusion in a particular analysis. The Table shows the correlation coefficients and p-values. Adenoma detection at colonoscopy for first-time screening was not significantly correlated with adenoma detection at colonoscopy for screening not first-time, surveillance or family history of CRC (Table). When combining colonoscopies with any of the four indications, adenoma detection was correlated with SSP detection (r=0.65, p<0.0001, Figure), but not advanced adenoma or advanced SSP detection (Table). In the current era, ADR over broadened indications beyond first-time screening may also reflect SSP detection, but not necessarily lesion detection during screening not first-time, surveillance or screening for family history of CRC, or advanced lesion detection. The contributions of chance, patient characteristics, or endoscopist performance to these results remain to be determined, including whether endoscopist performance is affected by particular vigilance to ADR during first-time screening. More comprehensive monitoring across the range of colonoscopy indications and lesion types could improve the assessment of endoscopist performance, and possibly improve CRC control efforts.Significant correlation between adenoma and sessile serrated polyp detection rates during colonoscopy for four screening and surveillance indicationsView Large Image Figure ViewerDownload Hi-res image Download (PPT)