Background: The medical education of French family medicine residents involves active, socioconstructivist-inspired small-group courses useful for skill acquisition. This is challenged by the increasing gap between the growing number of residents and the limited number of teachers. Blended courses have the potential to address this issue by reducing the duration of face-to-face sessions while preserving small-group courses.

Objective: This study aimed to compare the effects of blended vs traditional, face-to-face, active, socioconstructivist learning on the acquisition of knowledge and skills by family medicine residents.

Methods: We conducted a randomized controlled trial to compare a blended course and a traditional course. The blended course involved 2.5 hours of asynchronous e-learning and a 3-hour face-to-face session. The traditional course involved 5.5 hours of face-to-face teaching. Both courses were grounded in socioconstructivist principles and actively engaged residents. The primary outcome was residents' self-assessment of knowledge and skills. Secondary outcomes included satisfaction with knowledge- or skill-related learning objectives and academic achievement at 6 months.

Results: We included 155 family medicine residents (n=78, 50.3% in the blended course and n=77, 49.7% in the traditional course). There was no significant difference between groups regarding the primary outcome (mean difference 0.40 [maximum mean difference 20] points, 95% CI -0.21 to 1.02; P=.19; Cohen d=0.21). No significant differences were observed for the secondary outcomes except for knowledge self-assessment, which was higher in the blended course but not educationally meaningful (mean difference 0.40 [maximum possible 10] points, 95% CI 0.07-0.71; P=.02; Cohen d=0.39).

Conclusions: Blended courses can help sustain socioconstructivist small-group teaching methods while accommodating a growing family medicine resident population, with no deleterious impact on knowledge and skill self-assessments.