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

Smith, Samuel C L, Siau, Keith, Cannatelli, Rosanna, Antonelli, Giulio, Shivaji, Uday N, Ghosh, Subrata, Saltzman, John R, Hassan, Cesare and Iacucci, Marietta (2021) Training methods in optical diagnosis and characterization of colorectal polyps: a systematic review and meta-analysis. Endoscopy international open, 9 (5). E716-E726. ISSN 2364-3722. This article is available to all UHB staff and students via ASK Discovery tool http://tinyurl.com/z795c8c by using their UHB Athens login IDs

[img]
Preview
Text (PDF file format)
10-1055-a-1381-7181.pdf - Published Version
Available under License Creative Commons Attribution 4.0.

Download (911kB) | Preview
Official URL: https://doaj.org/toc/2364-3722

Abstract

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. 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. 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,  < 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,  < 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,  < 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  < 0.001). On meta-regression analysis, the improvement in accuracy did not differ between computerized vs. didactic training approaches for experienced (  = 0.792) and trainee endoscopists (  = 0.312). 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.

Item Type: Article
Additional Information: This article is available to all UHB staff and students via ASK Discovery tool http://tinyurl.com/z795c8c by using their UHB Athens login IDs
Subjects: QW Microbiology. Immunology
WI Digestive system. Gastroenterology
Divisions: Planned IP Care > Gastroentrology
Related URLs:
Depositing User: Jamie Edgar
Date Deposited: 13 May 2021 09:55
Last Modified: 13 May 2021 09:55
URI: http://www.repository.uhblibrary.co.uk/id/eprint/4312

Actions (login required)

View Item View Item