Introduction Capsule endoscopy (CE) is now the standard investigation for suspected small bowel disease, yet training opportunities remain limited. Unlike flexible endoscopy training, the lesion recognition skills required for CE are ideally suited to web-based e-learning, and with the advent of the e-learning for Health (Elf-H) endoscopy module, could be made available to a wide audience. We have previously developed an e-learning module for lesion recognition skills and validated them under research conditions. A topic-based teaching module provides multimedia learning on normal anatomy, anatomical variations/incidental findings, and pathological abnormalities. An assessment module consists of 50 de-identified video clips with multiple choice questions that can be used as a formative assessment. The aim of this study was to investigate the training efficacy of these modules in a real-life web-based learning environment.

Methods The assessment module questions that had previously demonstrated construct validity were stratified into two 22 question papers (A and B) which were used in a pre/post test design, with the programme randomly allocating the order of completion (AB or BA) to each individual. The programme automatically records and emails the results from the assessments. Both the training and assessment modules were developed into a downloadable package and placed on the Wolfson Unit website. Invitations to test the module were sent to all gastroenterology trainees in England and placed on several training-related websites. Participants expressing an interest were given access to download the package and could complete the module in their own time.

Results 99 individuals have so far expressed an interest in completing the module, with full results available for 27. Logistic regression analysis was used to compare the difference between the pre and post training performances on the test modules. There was a significant improvement in test performance after training (p=0.037), although this improvement was slightly smaller than in our previous validation study (8.9% vs 14.9%). Feedback was extremely positive regarding both the training and test modules, with median (range) scores out of 10 for their usefulness of 8.4 (6–10) and 7.2 (4–10), respectively. Trainees found it enjoyable, comprehensive, interactive and easy to use.

Conclusion Validated web-based e-learning is an effective tool for improving CE lesion recognition skills under “real-life” learning conditions. It can allow rapid dissemination of quality training material and best practice to a wide audience. This approach could help address the current issues surrounding provision of CE training in the UK and elsewhere.