Meta-analytic Review of Flipped Classrooms in Health Professions Education

Flipped classes are used in health professions education (HPE). However, their effectiveness varies considerably between studies. We conducted a meta-analysis to identify factors that influenced flipped class effectiveness in HPE.

We included studies that compared learning outcomes between flipped and lecture sections in undergraduate or graduate HPE programs. We excluded studies not reported in English. Our search yielded 3,549 unique records. Screening related systematic reviews yielded another 40 records. Risk of bias was assessed using the RoB-2 and ROBINS-I tools. Exam scores (extracted as percentage points [%pt]) and student-rated learning (extracted as 5-point Likert scale ratings) were synthesized using random-effects models.

Very low-quality evidence showed that exam scores were 5.44 percentage points (%pt) higher in flipped sections than lecture sections (CI 95 = 4.53 to 6.35 %pt, 124 studies, 23,326 observations). Low-quality evidence showed that student-rated learning was 0.47 Likert scale points higher in flipped sections than lecture sections (CI 95 = 0.29 to 0.66 points, 22 studies, 3,790 observations). In subgroup analyses, flipped sections' effectiveness was influenced by: (1) additional retrieval practice (low-to moderate-quality evidence), (2) retrieval practice format (low-quality evidence), and (3) control group type (very low- to low-quality evidence).

Providing additional retrieval practice improves flipped class effectiveness across multiple HPE fields. Flipped class effects remained heterogeneous even after subgroup analyses, suggesting additional factors remain unidentified.