To process research findings and other information, medical doctors need critical appraisal skills for careful and systematic assessment of the trustworthiness, value, and relevance of the available evidence in particular contexts. Medical students must, therefore, be prepared to learn and master these skills, and various approaches have been implemented to that end.^1-3 Within medical education, peer teaching, peer-assisted learning, peer tutoring, peer assessment, and other methods involving the deployment of students or trainees in a teaching role have attracted increasing interest.^4,5 Peer teaching can be defined as “an educational arrangement in which one student teaches one or more fellow students.” When the teacher is a student who is more advanced (by at least one year) in the same curriculum as the learner, this is known as “near-peer teaching,” which is one of the most common forms of peer teaching.⁴ Some studies have also classified recently graduated medical doctors as near-peer teachers by virtue of their proximity in age and experience.^6,7 Peer teaching and near-peer teaching methods have many advantages, which include alleviating faculty teaching burden, providing role models for junior students, enhancing students’ intrinsic motivation, and preparing physicians for their future role as educators.⁵ Although the effectiveness and efficiency of peer-assisted learning have been demonstrated in previous studies, these related mainly to a facilitation of psychomotor skills. Several of the published randomized controlled trials assessing the effectiveness of peer teaching relate to clinical skills such as intravenous (IV) access and bladder catheterization,⁸ neurological examination and lumbar puncture,⁹ resuscitation skills,¹⁰ interpretation of musculoskeletal ultrasound of the shoulder (as part of a sports medicine module),¹¹ and basic clinical examination skills.¹²

Among the limited number of studies comparing student-led tutorials (SLT) and faculty-led tutorials (FLT), Kassab and colleagues¹³ studied the use of problem-based learning (PBL) to teach haematology. Other non-randomized controlled studies of peer teaching of evidence-based medicine include Josephson et al. in the US¹⁴ and Rees and colleagues in the UK.¹⁵ However, both of those aimed solely to demonstrate the feasibility of such programs rather than to provide evidence of the effectiveness of peer teaching interventions. Sabouni and colleagues ¹⁶ also explored the utilization of peer teaching in an evidence-based medicine course in a resource-poor setting. Although the knowledge and skills of course participants were found to have improved significantly after the course, the study did not report any change in attitude. In addition, the pre-post study design is not the most robust means of determining the effectiveness of a given intervention.

Although the literature notes many benefits of peer-assisted learning, these are largely confined to the acquisition of psychomotor skills. In light of the growing need for innovative ways of incorporating EBM into medical school curricula, especially where human resources are limited, the present study aimed to evaluate the effectiveness of recently graduated medical doctors as near-peer tutors as compared to experienced medical staff. Specifically, we hypothesized that near-peer tutors would be as effective as medical staff in teaching critical appraisal skills to medical students as part of a Clinical Epidemiology and Evidence-based Medicine (CE-EBM) program.

The mean + SD age of participating students was 22.1+1.1 years (range 18.9–26.7 years), of whom 89 (36.9%) were male. The median age of near-peer tutors was 24 years (range 23–26 years). The median age of staff tutors was 41.5 years (range 37–55 years).

A mid-evaluation following the first two discussion sessions found no statistically significant difference (t₍₂₄₁₎ = -1.520, p = 0.130) between mean + SD of written test scores for the near-peer tutored group (65.89 + 12.86) and the staff-tutored group (63.18 + 13.69). The groups then exchanged tutors for the next two discussion sessions. At the end of the module, the difference in mean + SD of written test scores for the near-peer tutored group (68.20 + 12.50) and the staff-tutored group (70.61+10.99) was again statistically non-significant (t₍₂₄₁₎ = 1.553, p = 0.122).

At the beginning of the module, student attitude scores were comparable (Table 1). After the module, there were no statistically significant differences in the mean overall attitude score or for the majority of attitude items other than the statement “If evidence-based medicine is valid, then anyone can see patients and do what doctors do.”

The overall score for level of student confidence in Evidence-based Practice (based on the EPIC questionnaire) at the beginning of the module was also comparable for both groups (Table 2). However, a statistically significant lower score was observed for medical staff on two items (4 - Critically appraise the strengths and weaknesses of study methods, and 8 - Determine if evidence from the research literature applies to your patient's or client's situation). No statistically significant difference in mean EPIC scale score was observed either for individual items or overall scores at the end of the module. When the students were asked to evaluate the tutorial process, they assigned more positive scores to the near-peer tutor for the majority of evaluation items (Table 3).

In the present study, near-peer tutored students and staff-tutored students achieved comparable scores on knowledge, skills, attitudes, and confidence in relation to critical appraisal skills. Although confirming what has been reported in earlier studies, these findings must nevertheless be interpreted with a degree of caution. For example, the similarity of these results, especially on the written test, may reflect the fact that all students attended lectures delivered by experts in clinical epidemiology. Another possible explanation (as suggested by ten Cate and colleagues)²⁶ is that peer-teaching may also encourage students to invest more effort in the study because of concerns about inadequate teaching.

Additionally, students scored better on the tutorials led by near-peer tutors than on those led by medical staff tutors. Tolsgaard and colleagues reported a similar result in a study employing second-year medical students as tutors to teach IV access and bladder catheterization to first-year medical students at the University of Copenhagen.⁸ As explained by cognitive congruence theory, a teacher whose knowledge base is similar (i.e., congruent) to the learner’s is likely to be more effective than an expert in the field who has a dissimilar knowledge base (i.e., who is cognitively incongruent or is perceived as more “cognitively distant”). Peer-tutors are also believed to be able to provide a more comfortable and safe learning environment for students, as they come from a similar social environment.⁵

A systematic review by Burgess and colleagues²⁷ indicated that practical issues such as limited faculty resources were most frequently reported to account for the implementation of peer or near-peer teaching. Another justification for the use of near-peer tutors is the need to develop the teaching skills of future doctors. While the Indonesian Medical Council does not explicitly list teaching skills among the competencies of medical doctors, Indonesian medical graduates are expected to be able to provide health education to patients, families, and communities.²⁸ In Tomorrow’s Doctors, the UK General Medical Council states that medical graduates must be able to demonstrate appropriate teaching skills,²⁹ and the Royal College of Physicians and Surgeons of Canada states in its CanMEDS Physicians Competency framework that the physician’s role includes being a “scholar” who are competent to “teach students, residents, the public, and other health care professionals”.³⁰

As well as the benefits for the medical students, the peer tutors also gained, as the teaching experience motivated them to learn more and to deepen their understanding of the subject. This is an expected aspect of EBM skills, which are currently seen as one of the core competencies of medical doctors. Experience in teaching and preparation of teaching materials is also known to build leadership capability and confidence.⁵

Although peer teaching has been widely implemented in medical education, we are not aware of any previous studies of this kind from Indonesia, although in our experience the practice is undoubtedly performed informally. We recognize that as educational interventions are generally complex, it is often difficult to distinguish which element accounts for which effect or to what extent it might be adapted to local circumstances while remaining effective.³¹ Moreover, certain particulars may limit the generalization of the present findings, such as student characteristics and EBM as a distinctive subject. As the oldest and most prominent medical school in Indonesia, FMUI students are a select group, and earlier research found that tutor expertise did not affect FMUI students’ performance during problem-based learning.³² Furthermore, the ethics of medical education require the minimization of any potential disadvantage to students assigned to a near-peer tutor – hence the need for a crossover. As a result, none of the students in this study were assigned only to a near-peer tutor, and this may have led to over-estimation of results, especially in the final test. Future studies should, therefore, extend the utilization of near-peer tutors to all tutorial sessions.

The present findings confirm that near-peer tutors can be as effective as and more readily accepted than staff tutors. This is based on the performance of students in critical appraisal skills in EBM, extending the evidence of the benefits of peer-assisted learning beyond the area of psychomotor skills. This study also provides preliminary evidence that medical school educational policy will be accommodated to the assignment of trained near-peer tutors to teach critical appraisal skills to medical students.