Self-reported knowledge, skills, and attitudes toward evidence-based medicine among medical interns in a Latin American multicenter teaching-hospital network: a cross-sectional study from Bolivia

Background Evidence-based medicine (EBM) is an important component of contemporary medical education, but evidence describing self-reported EBM knowledge, skills, and attitudes among medical interns in Bolivia is limited. We aimed to characterize these self-reported dimensions and examine their statistical associations with selected individual and institutional variables. Methods We conducted a multicenter cross-sectional study between January and July 2025 in eight teaching hospitals in La Paz and El Alto. Of 393 interns in the source cohort, 233 completed the questionnaire. Because the recruitment was consecutive and non-probabilistic, 233/393 is reported as a cohort participation proportion (59.3%), not as a response rate unless the full cohort is confirmed to have been invited. The questionnaire included 36 knowledge, 11 skills, and 11 attitude items, all scored on 5-point response scales. Internal consistency was assessed with Cronbach’s alpha, and exploratory factor analysis (EFA) was used as structural evidence. Group comparisons, rank-based institutional comparisons, MANOVA, multiple linear regression, and Spearman correlations were performed in R 4.5.1. Results Mean scores were 2.39 ± 0.47 for self-reported knowledge, 2.65 ± 0.64 for self-reported skills, and 3.76 ± 0.52 for attitudes. Prior Research Methodology training was associated with higher self-reported knowledge (2.50 ± 0.45 vs. 2.23 ± 0.45; p < 0.001; Cohen’s d = 0.60) and skills (2.75 ± 0.62 vs. 2.58 ± 0.64; p ≈ 0.043 based on the reported group summaries). Institutional variations in self-reported skills were observed; however, subgroup comparisons involving institutions with sparse sample sizes ( n ≤ 3) were restricted to maintain analytical stability. Conclusions Medical interns reported substantially more favorable attitudes toward evidence-based medicine than self-reported knowledge or practical skills. Prior Research Methodology training was positively associated with higher self-reported knowledge and skills in bivariate analyses. Institutional variations across teaching hospitals and universities warrant cautious interpretation due to self-reported metrics, institutional clustering, and sparse subgroup constraints (n ≤ 3). Objective performance assessments and cluster-aware longitudinal studies are required to guide regional educational strategies.