Comparison of AR, ITS, CBT, and Didactic Training and Evaluation of Retinopathy-Based Diagnosis

To test the relative efficacy of adaptive and fixed simulator training vs. adaptive and fixed computer-based training (CBT) vs. traditional lecture.

Ninety-two board certified Family Physicians were given a pretest, followed by training under one of five randomly assigned training conditions: (i) adaptive CBT; (ii) non-adaptive CBT; (iii) adaptive simulator; (iv) non-adaptive simulator; and (v) lecture. Subjects were given a post-test immediately after training and again in 6 weeks. In the adaptive groups, the content presented to subjects was a function of the subjects' knowledge and performance, under control of a MOODLE LMS. In the lecture component, a physician projected images identical to the images in the CBT component of the study.

No significant differences (p = 0.89 for ANOVA) in baseline knowledge between the five groups, based on pretest scores; In each of the five groups, our intervention resulted in immediate changes in knowledge (95% CI), based on analysis of pretest and PostTest1 scores; and, with the exception of subjects with non-adaptive, simulator intervention, all groups had significant decay in knowledge between the first and second post-tests (95% CI).

Periodic simulator intervention could result in significant knowledge retention over extended periods of time.