Background: Every physician in Ontario needs to know how to fill out a Form 1 in order to legally hold a person against their will for a psychiatric assessment. These forms are frequently inaccurately filled out, which constitutes wrongful confinement and could lead to fines as large as $25,000. Training people to fill out a Form 1 accurately is a large task, and e-learning (Internet-based training) provides a potentially efficient model for health human resources training on the Form 1. Objective: In this study, we looked at the efficacy of an e-learning module on the Form 1 by comparing baseline knowledge and skills with post-test performance. Methods: 7 medical students and 15 resident physicians were recruited for this study from within an academic health sciences setting in Hamilton, Ontario, Canada (McMaster University). The intervention took place over 1 hour in an educational computing lab and included a pretest (with tests of factual knowledge, clinical reasoning, and demonstration of skill filling out a Form 1), the e-learning module intervention, and a post-test. The primary outcome was the change between pre- and post-test performance. A scoring system for grading the accuracy of the Form 1 was developed and two blinded raters marked forms independently. Participants were randomly assigned to one of two sequences of assessments (A then B vs B then A), with a balanced design determining which test the participants received as either the pretest or post-test. Inter-rater reliability was determined using the Intraclass Correlation. Repeated measures analysis of variance was conducted. Results: The Intraclass Correlation (ICC) as the measure for inter-rater reliability was 0.98. For all outcome measures of knowledge, clinical reasoning, and skill at filling out the Form 1 there was a statistically significant improvement between pretest and post-test performance (knowledge, F(1,21) 54.5, p<0.001; clinical reasoning, F(1,21) 9.39, p=0.006; Form 1 skill, F(1,21) 15.7, p=0.001). Further analysis showed no significant differences or interactions with other variables such as between raters, the order of assessment, or trainee type. Conclusions: Under laboratory conditions, this e-learning module demonstrated substantial efficacy for training medical students and residents on the theory and practice of filling out the Form 1 of the Mental Health Act. E-learning may prove to be an efficient and cost-effective medium for training physicians on this important medico-legal aspect of care. Further research is required to look at the longer-term impact of training and broader implementation strategies across the province for medical trainees and practicing physicians.