Evaluating the outcomes of problem-based learning in postgraduate medical education: a systematic review and meta-analysis.

Background: Educators have recently been compelled to incorporate more active instructional formats into medical education, such as problem-based learning (PBL). In view of the mixed outcome data on the use of PBL in postgraduate medical education (PGME), there is a need to synthesize the data to inform the application of PBL in PGME contexts.

Objective: The aim of this systematic review and meta-analysis was to synthesize learning outcomes of PBL in PGME contexts.

Methods: The authors systematically searched MEDLINE, Embase, APA PsycINFO, AMED, CINAHL, Web of Science, ERIC, and Cochrane databases from January 1, 1950, to July 1, 2022 for original studies that reported Kirkpatrick outcomes of PBL in PGME contexts. Outcomes data were extracted. Quantitative data relating to learning outcomes were meta-analyzed using a random-effects model to generate weighted mean differences.

Results: Of 4310 abstracts screened, the authors included 21 studies encompassing anesthesia, family medicine, internal medicine, occupational medicine, pediatrics, psychiatry, public health and surgical residency programs. The studies reported reaction (n = 12), learning (n = 15), behavioural (n = 6) and/or results outcomes (n = 4). Meta-analysis of the three eligible articles demonstrated no significant difference after PBL in pre- and post-test results (pooled mean difference=0.13%, 95% CI, -6.74-7.00). There were observed improvements in satisfaction levels and self-reported behavioural outcomes following PBL.

Conclusions: Although similar learning outcomes were observed using PBL and the usual teaching in PGME, PBL was associated with benefits in trainee satisfaction and behavioural changes that contribute to learning and performance. PGME programs should consider incorporating PBL into curricula.