Training methods in optical diagnosis and characterization of colorectal polyps: a systematic review and meta-analysis

Samuel C L Smith, Keith Siau, Rosanna Cannatelli, Giulio Antonelli, Uday N Shivaji, Subrata Ghosh, John R Saltzman, Cesare Hassan, Marietta Iacucci

Research output: Contribution to journalArticlepeer-review

93 Downloads (Pure)


Background and study aims  Correct optical diagnosis of colorectal polyps is crucial to implement a resect and discard strategy. Training methods have been proposed to reach recommended optical diagnosis thresholds. The aim of our study was to present a systematic review and meta-analysis on optical diagnosis training. Methods  PubMed/Medline and Cochrane databases were searched between 1980 and October 2019 for studies reporting outcomes on optical diagnosis training of colorectal polyps. The primary outcome was optical diagnosis accuracy compared to histological analysis pre-training and post-training intervention. Subgroup analyses of experienced/trainee endoscopists, training methods, and small/diminutive polyps were included. Results  Overall, 16 studies met inclusion criteria, analyzing the impact of training on 179 endoscopists. Pre-training accuracy was 70.3 % (6416/9131 correct diagnoses) whereas post-training accuracy was 81.6 % (7416/9213 correct diagnoses) (risk ratio [RR] 1.17; 95 % confidence interval [CI]: 1.09-1.24, P  < 0.001). In experienced endoscopists, accuracy improved from 69.8 % (3771/5403 correct diagnoses) to 82.4 % (4521/5485 correct diagnoses) (RR 1.20; 95 % CI: 1.11-1.29, P  < 0.001). Among trainees, accuracy improved from 69.6 % (2645/3803 correct diagnoses) to 78.8 % (2995/3803 correct diagnoses) (RR 1.14; 95 % CI 1.06-1.24, P  < 0.001). In the small/diminutive polyp subgroup, accuracy improved from 68.1 % (3549/5214 correct diagnoses) to 77.1 % (4022/5214 correct diagnoses) in (RR 1.16 95 % CI 1.08-1.24 P  < 0.001). On meta-regression analysis, the improvement in accuracy did not differ between computerized vs. didactic training approaches for experienced ( P  = 0.792) and trainee endoscopists ( P  = 0.312). Conclusions  Optical diagnosis training is effective in improving accuracy of histology prediction in colorectal polyps. Didactic and computer-based training show comparable effectiveness in improving diagnostic accuracy.

Original languageEnglish
Pages (from-to)E716-E726
JournalEndoscopy International Open
Issue number5
Early online date22 Apr 2021
Publication statusPublished - May 2021

Bibliographical note

The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (


Dive into the research topics of 'Training methods in optical diagnosis and characterization of colorectal polyps: a systematic review and meta-analysis'. Together they form a unique fingerprint.

Cite this