Background: The importance of ensuring medical students are equipped with the skills to be able to practice evidence-based medicine (EBM) has been increasingly recognized in recent years. However, there is limited information on an effective EBM curriculum for undergraduate medical schools. This study aims to test the feasibility of integrating a multifaceted EBM curriculum in the early years of an undergraduate medical school. This was subsequently evaluated using the validated Fresno test and students' self-reported knowledge and attitudes as they progressed through the curriculum. Methods: EBM was integrated horizontally and vertically into the curriculum into the first 2 years of undergraduate medical school. First year medical students were recruited to participate in the study. The 212-point Fresno test was administered along with a locally developed questionnaire at baseline before EBM teaching in year one and at the end of EBM teaching in year two. Results: Thirty-one students participated at baseline and 55 students participated at the end of second year EBM teaching. For the 18 students who completed the Fresno at both time points, the average score increased by 38.7 marks (p < 0.001) after EBM teaching. Students felt confident in formulating clinical questions and in critically appraising journal articles after EBM teaching. EBM was perceived to be important to their future practice as a doctor and for improving patient outcomes at both time points. Conclusions: It has been feasible to integrate a multifaceted, EBM curriculum from the first year of an undergraduate medical program. Early evaluation of the curriculum using the Fresno test has shown a significant increase in students' EBM knowledge. The curriculum also demonstrated an increase in students' perceptions of the clinical relevance of EBM in their developing practice.

Conclusions EBM+ is shown to provide a promising programme for making better use of existing evidence when assessing interventions and exposures.Concerns about infeasibility, bias and subjectivity can be allayed, although some attempts to tackle these concerns are unsuccessful.

No abstract