Endoscopist quality is important after a positive stool test


Higher detection rates of endoscopist polyps are significantly associated with fewer negative colonoscopies after positive stool tests, indicating a need for new detection benchmarks, researchers say.


  • Researchers evaluated the extent to which positive stool tests followed by negative colonoscopy (“false-positive” stool tests) vary depending on the quality of the endoscopist.
  • They analyzed data from the New Hampshire Colonoscopy Registry of 864 patients with a positive multitarget stool DNA test (mt-sDNA) and 497 with a positive fecal immunochemical test (FIT) who underwent follow-up colonoscopy.
  • They compared the frequency of ‘false’ negative colonoscopies and polyp detection across four quartiles of the detection rate of endoscopist adenoma (ADR) and the clinically significant detection rate of serrated polyps (CSSDR).


  • Negative colonoscopies were significantly less common among endoscopists with higher ADR and CSSDR, especially in the top two quartiles.
  • After a positive mt-sDNA test, the detection rate of any adenoma was 62.8% for endoscopists in the top quartile versus 48.7% in the bottom quartile (P <.001). For CSSDR, detection rates were 66.7% in the top quartile versus 46.9% in the bottom quartile (P <.001).
  • The results were similar after a positive FIT test. Detection of any adenoma was 63.3% in the top quartile versus 35.8% in the bottom quartile (P <.001). For CSSDR, detection rates were 54.6% in the top quartile versus 37.3% in the bottom quartile (P <.001).
  • Significant differences were also observed in the detection of sessile serrated lesions (SSL) after a positive stool test, with higher detection rates in the upper quartile than in the lowest quartile. Among endoscopists in the upper quartile of CSSDR, SSLs were found in 29.2% of examinations after a positive mt-sDNA test and in 13.5% of examinations after a positive FIT test.


Based on their findings from high-performing endoscopists, the researchers suggested “benchmarks of at least 40% (with 60% aspiration detection) for adenoma detection after positive mt-sDNA or FIT, benchmarks of 20% (with 30% aspiration detection) for sessile serrated lesions after positive mt-sDNA, and a benchmark of 15% for detection of serrated polyps after a positive FIT.”


The study, with first author Lynn F. Butterly, MD, of the Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, was published online in The American Journal of Gastroenterology.


The study was limited to a relatively racially homogeneous population in New Hampshire, which may have affected generalizability. The authors recognized the need for further validation of the proposed benchmarks in other populations.


This research was supported by a grant to the New Hampshire Colonoscopy Registry from Exact Sciences. One author is an employee of Physical Sciences.