This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Evidence-based medicine (EBM) is an important element of medical education. However, nationwide data on EBM education in Korean medical schools are limited. This study aimed to describe the organizational structure, content, methods, and assessment of EBM education, together with barriers and support needs, and to explore factors associated with variation across schools.
Methods
A 26-item questionnaire was administered to EBM education leaders or course directors at 21 Korean medical schools. Data were analyzed descriptively, and differences by structural and exogenous characteristics were explored using cross-tabulation.
Results
Twenty-one of 40 medical schools responded. All responding institutions offered EBM as a mandatory subject, but 42.9% had no dedicated organizing unit. Coverage of classic EBM steps was high (formulating clinical questions: 95.2%; literature searching: 95.2%; critical appraisal: 76.2%), whereas AI-assisted evidence summarization was included in only 19.0%. Only 19.0% reported systematic theory–clinical integration, and 57.1% left bedside EBM to preceptor discretion. Demand for a standardized curriculum guide (90.5%) and faculty development (76.2%) was high, and 100% were willing to use externally developed materials. Variation in key outcomes was associated with organizational governance but was not explained by exogenous structural characteristics such as class size, clinical infrastructure, establishment type, or region.
Conclusions
EBM education in Korean medical schools is universally mandatory but uneven in organization, theory–clinical integration, faculty capacity, assessment, and AI integration. These differences appeared to reflect governance more than resource size, suggesting that society-level support should include organizational models alongside standardized curricula, assessment tools, and faculty development.
Evidence-based medicine (EBM) is an approach to clinical decision-making that integrates the best research evidence with clinical expertise and patient values, and since the 1990s it has become a core competency in medical education [1,2]. EBM education is typically structured around five steps: formulating clinical questions (PICO), acquiring evidence (literature searching), critical appraisal, application, and evaluation of performance. There is international consensus that it should not be confined to lectures but linked to clinical settings and reinforced longitudinally across the curriculum [3].
Korean medical education is undergoing a major transition. The February 2024 amendment to the Enforcement Decree of the Higher Education Act removed the distinction between a 2-year premedical and a 4-year medical phase, allowing each school to autonomously design an integrated 6-year curriculum [4]. The accreditation standards of the Korean Institute of Medical Education and Evaluation, ASK2026, explicitly require evidence-based reasoning and decision-making competency (K.2.2.1), basic–clinical vertical integration (K.2.3.2), and modular/spiral design (K.2.1.1) [5]. Within this context, it has been argued that EBM should be positioned not as knowledge confined to a single course but as a longitudinal competency running through the 6-year curriculum [6,7].
Despite these institutional demands, systematic nationwide data on how Korean medical schools actually organize, deliver, and assess EBM are scarce. Although individual curricular reform cases have been reported [8,9], surveys encompassing the responsible body, timing and hours, inclusion of each step, theory–clinical integration, assessment, and the barriers and support needs perceived in the field are rare.
Therefore, this study surveyed Korean medical schools to describe (1) the organizational structure and mode of delivery, (2) content and methods by step, (3) theory–clinical integration and assessment, (4) barriers and support needs of EBM education, and (5) to explore which structural factors and characteristics were associated with variation in outcomes, in order to provide baseline data for future standardization and curricular integration.
Methods
Study Design and Participants
This was a cross-sectional survey to assess the status of EBM education in Korean medical schools. With the cooperation of the Korean Society for Evidence-Based Medicine and the Korean Society of Medical Education, the survey was distributed to all 40 medical schools nationwide and administered via Google Forms.
Survey Instrument
The questionnaire comprised 26 items covering (1) institutional and respondent characteristics; (2) the body governing EBM education and the mode, timing, and hours of delivery; (3) inclusion of content for each EBM step (formulating clinical questions, literature searching, critical appraisal, clinical practice guidelines, levels of evidence/strength of recommendation, and AI-based evidence summarization); (4) teaching methods and use of digital/AI tools; (5) theory–clinical integration and involvement of specialists; (6) assessment methods and reflection in graduation requirements; (7) operational difficulties and barriers; and (8) demand for society-level support materials and willingness to use external materials. Items used a mix of single-choice, multiple-choice, and free-text formats. To develop the instrument, a preliminary status review was conducted with faculty members responsible for EBM education at several medical schools, and the content validity of the items was reviewed by the board of directors of the Korean Society for Evidence-Based Medicine
Data Collection and Analysis
Data were collected via an online survey between April 1 and April 30, 2026. To improve the response rate, an official request was made through the Korean Society of Medical Education. Given the small sample size, analyses were primarily descriptive (frequencies and percentages), and 95% confidence intervals for key proportions were calculated using the Wilson method. For multiple-choice items, the proportion selecting each option was calculated using the number of responding institutions (n=21) as the denominator. To explore factors associated with variation in outcomes, EBM governance was operationalized as “having a dedicated organizing unit” and “an integrated methodology–clinical design,” and was cross-tabulated with outcome indicators and exogenous structural characteristics (class size, establishment type, region, and number of tertiary hospitals). For this, the 2026 admission quota (based on the reversion to the pre-expansion level of 3,058) and the Ministry of Health and Welfare’s 5th-cycle (2024–2026) designation of tertiary hospitals were used as external data, and a sensitivity analysis weighted by admission quota was performed in parallel. Fisher’s exact test was applied to selected bivariate relationships; however, because no correction for multiple comparisons was applied and statistical power was limited, p-values were interpreted only for hypothesis generation rather than confirmation. Owing to constraints on sample and event numbers, multivariable analysis was not performed. Free-text responses were summarized qualitatively.
Results
Organizational Structure of EBM Education
The respondents were one person per institution responsible for or involved in EBM education at each of the 21 schools. Eight respondents (38.1%) were medical education leaders (e.g., associate dean for education), 7 (33.3%) were course directors (e.g., EBM/clinical epidemiology), 5 (23.8%) were ‘other’, and one (4.8%) was a clinical clerkship director (Table 1). As of the 2026 academic year, there are 40 medical schools in Korea (admission quota 3,058); the 21 responding schools represent 52.5% of institutions and 56.8% (1,738/3,058) of the admission quota. Response rates were relatively higher among national universities (7/10) and schools in the Seoul metropolitan area (9/13). All responding institutions (21/21, 100%) provided EBM as mandatory education (Table 2). However, the most common arrangement was having no dedicated organizing unit (9 schools, 42.9%), followed by a specific department (e.g., preventive medicine, health policy and management, biostatistics) in 7 (33.3%), an associate-dean-led structure in 4 (19.0%), and a department of medical education in 1 (4.8%) (Table 1). The most common mode of delivery was as part of an integrated course (10 schools, 47.6%), followed by integrated methodology–clinical teaching (7, 33.3%) and clinical-department-led teaching (3, 14.3%). EBM was most often taught in medical years 1–2 (57.1%) and years 3–4 (38.1%), with early exposure in the premedical phase in only 14.3%. Total instructional hours were ≥21 hours and 11–20 hours in 33.3% each, whereas ≤10 hours also accounted for 33.3%, indicating substantial variation across institutions.
Educational Content
Coverage by EBM step was highest for formulating clinical questions (PICO, 95.2%) and literature searching (95.2%), followed by interpreting levels of evidence/strength of recommendation (81.0%), critical appraisal of research (76.2%), and appraisal/use of clinical practice guidelines (61.9%), indicating generally high coverage of the classic EBM steps (Table 3). In contrast, only 19.0% included education on AI-assisted evidence summarization and search efficiency, while 47.6% did not include it and 33.3% were planning or improving its adoption.
Teaching Methods and Assessment
Lecture was the predominant method (90.5%), followed by clinical-practice-based EBM activities (52.4%), team-based learning (33.3%), and problem-based learning (28.6%), with journal clubs used by only 9.5% (Table 4). AI/digital tools were not used in the EBM process at 57.1% of institutions. Assessment methods (multiple response) were assignments (CAT/appraisal, 81.0%), presentations (71.4%), and written examinations (47.6%); however, EBM assessment results were reflected in graduation requirements or grades at only 38.1%, and 33.3% did not reflect them.
Theory–Clinical Integration and Personnel
Bedside EBM learning based on actual patient questions during clerkship was most often left to preceptor discretion (57.1%), with regular or partial implementation in 19.0% each (Table 4). Linkage between EBM theory faculty and clinical preceptors was based on individual instructors’ efforts in 47.6% and was reported as not achieved in 33.3%, with only 19.0% reporting systematic integration within the curriculum. Statisticians or information specialists participated partially in 81.0% of institutions but always in only 9.5%, and 47.6% reported that dedicated EBM personnel were insufficient or very insufficient.
Barriers and Support Needs
The greatest operational difficulty was a shortage of instructors (76.2%), followed by lack of educational materials (28.6%), lack of instructional time (23.8%), and lack of assessment methods (23.8%) (Table 5). The most frequently cited barriers were EBM being deprioritized relative to the heavy volume of medical knowledge (52.4%) and low awareness of the necessity of EBM within the curriculum (52.4%), followed by insufficient clinical faculty EBM teaching capacity (lack of role modeling, 42.9%). Materials considered helpful if provided by a society included a standardized EBM curriculum guide (90.5%), faculty development programs (train-the-trainer [TTT], 76.2%), shared online modules (66.7%), CAT templates/cases (57.1%), and assessment rubrics (52.4%); willingness to use externally developed materials was 100%.
Governance-Centered Exploration of Structural Factors (Exploratory)
Because the small sample precluded confirmatory testing and causal inference, the relationships between structural factors and outcomes were analyzed descriptively and exploratorily. The following results reflect co-occurrence and should be interpreted as hypothesis-generating.
Operationalizing governance. EBM governance was operationalized as “having a dedicated unit” (Table 1) and “an integrated methodology–clinical design” (Table 2); institutions with at least one of these were classified as having strong governance (15 schools), and those with neither as weak governance (6 schools).
Outcomes differed by the presence of governance. Key outcome indicators differed by strong versus weak governance (Table 6). Systematic theory–clinical integration (27% vs. 0%), inclusion of clinical practice guidelines (80% vs. 17%), reflection in graduation/grades (47% vs. 17%), and EBM conducted during clerkship (47% vs. 17%) were all higher in the strong-governance group, as was mean content coverage (4.60 vs. 2.83). Inclusion of the most basic step (PICO) was high in both groups (100% vs. 83%, near a ceiling effect), whereas higher-order steps such as critical appraisal (87% vs. 50%) and clinical practice guidelines were higher in the strong-governance group. Adding clinical faculty involvement as an additional axis (reverse-coded from the perceived lack of modeling) did not clearly differentiate outcomes; however, because this proxy is an indirect, subjective measure, whether its weak discrimination reflects a true null or a measurement limitation cannot be determined.
Governance was not explained by exogenous structural characteristics. We explored whether the distribution of governance was associated with schools’ exogenous structural characteristics (Table 7). Admission quota (small 86%, medium 67%, large 62%), establishment type (national 57%, private 79%), region (metropolitan 78%, non-metropolitan 67%), or number of tertiary hospitals (0: 50%, 1: 67%, ≥2: 86%) did not showed a consistent association with the proportion having strong governance. If anything, class size was weakly inverse, with a lower mean admission quota in the strong-governance group (78.8) than in the weak-governance group (92.7). In a quota-weighted sensitivity analysis of responding institutions, the distribution of structural factors was likewise independent of school size (weighted vs. unweighted differences mostly within ±2 percentage points). The number of tertiary hospitals was weakly associated with some clinical-application indicators such as clinical practice guidelines but was unrelated to systematic integration.
In summary, variation in EBM education appeared more related to governance (organization)—a dedicated unit and an integrated design—than to educational content or a school’s outward size, clinical infrastructure, establishment type, or region. However, these are exploratory, correlational observations from a small cross-sectional dataset, and the question of what shapes governance cannot be resolved with these data.
Discussion
This cross-sectional survey of 21 Korean medical schools found that, although EBM education is universally mandatory, substantial unevenness exists in operational infrastructure, theory–clinical integration, faculty capacity, assessment, and AI integration.
First, EBM education showed a “universal but structurally fragile” pattern. All institutions offered EBM as mandatory and coverage of basic steps such as PICO and literature searching was high, yet 42.9% operated without a dedicated unit and delivery was often integrated and dispersed. This suggests that although EBM has become established as content, the governance responsible for coordinating it remains immature. Given that ASK2026 requires a designated integrated-education lead faculty member and an education committee as the operating body of the integrated curriculum [5], clarifying the organizational responsibility for EBM is a priority.
Second, vertical integration linking theory and clinical practice emerged as the weakest point. That bedside EBM depended on preceptor discretion (57.1%) and that only 19.0% reported systematic theory–clinical integration indicate a disconnect in which EBM remains as early-year methodology teaching and fails to extend to the clinical setting. This reveals a gap between practice and both ASK2026’s requirement for basic–clinical vertical integration (K.2.3.2) and proposals to place EBM as a longitudinal spiral track in the integrated 6-year curriculum [6,7].
Third, the absence of faculty capacity and role modeling was a key barrier. A shortage of instructors was the greatest operational difficulty (76.2%), insufficient clinical faculty EBM teaching capacity (42.9%) was cited as a major barrier, and demand for faculty development programs (TTT) was high (76.2%). Given the importance of faculty who practice EBM at the bedside as role models for students’ EBM attitudes, developing EBM teaching skills for clinical faculty is needed.
Fourth, integration of AI/digital tools remained at an early stage. More than half (57.1%) did not use AI tools and AI-based evidence summarization was included in only 19.0%, although 33.3% reported planning or improving its adoption. As the use of large language models in evidence retrieval and summarization spreads, education should address both the efficiency of use and a critical stance toward the verification, accountability, and limitations of evidence.
Fifth, demand for society-level standardization support was virtually unanimous. Demand for a standardized curriculum guide (90.5%) and willingness to use external materials (100%) were especially high, indicating that, rather than relying on individual schools’ capacity, provision by relevant societies of standardized curricula, assessment rubrics, shared online modules, and faculty development programs would meet field needs.
Sixth, when structural factors were reanalyzed with a focus on governance, key outcomes (systematic integration, clinical practice guidelines, reflection in assessment, and clerkship EBM) differed by the presence of governance operationalized as a dedicated unit and an integrated design. In contrast, exogenous structural characteristics such as class size, number of tertiary hospitals, establishment type, and region showed no clear association with the distribution of governance or with outcomes, and class size was even weakly inverse [10]. This suggests that variation in EBM education may be more related to organizing EBM under clear responsibility and an integrated design than to educational content or outward resource size; that is, larger or more clinically resourced institutions did not necessarily have better-organized EBM education. However, these are exploratory, correlational observations from a small cross-sectional dataset and cannot serve as a basis for causal inference or for identifying the determinants of governance.
Limitations
This study has several limitations. First, 21 of the 40 Korean medical schools responded, covering 56.8% of the national admission quota; however, because national and Seoul metropolitan area schools responded at relatively higher rates, non-response bias cannot be excluded. Second, the small sample and subgroup sizes (e.g., only 6 schools in the weak-governance group) yield wide confidence intervals and limited statistical power; therefore, the governance and exogenous-characteristic analyses were exploratory and hypothesis-generating, without correction for multiple comparisons or multivariable modeling. Third, governance and clinical faculty involvement were indirectly operationalized from questionnaire items and thus have measurement limitations, and the determinants of governance could not be identified. Fourth, the study relied on a single self-reported respondent per institution, and its cross-sectional design precludes causal or temporal inference. Future multi-institution, multi-respondent studies combining curricular document analysis and student outcome measures are needed.
EBM education in Korean medical schools is largely established in terms of content, being universally mandatory with high coverage of the classic core steps. However, substantial variation exists across institutions in dedicated organization, theory–clinical vertical integration, faculty capacity, meaningful reflection in assessment, and AI integration. In exploratory analysis, this variation was associated with the presence of governance (a dedicated unit and an integrated design) and was not explained by exogenous structural characteristics such as class size, clinical infrastructure, establishment type, or region. This suggests that the quality of EBM education may be more related to organization than to resource size. During the transition to the integrated 6-year curriculum and ASK2026, it will be important to redesign EBM as a longitudinal competency running through the 6-year curriculum and to provide, at the society level, support for organization (dedicated responsibility structures and integrated design) alongside standardized curricula, assessment tools, and faculty development. However, as a small cross-sectional survey, this study cannot establish causality.
Notes
Conflict of Interest
Hyun Jung Kim has been a editor of Journal of Evidence-Based Practice since 2025, and Soo Young Kim has been a editor of Journal of Evidence-Based Practice since 2025. However, they were not involved in the peer reviewer selection, evaluation, or decision process of this article. No other potential conflicts of interest relevant to this article were reported.
Strong governance = having a dedicated unit or an integrated methodology–clinical design (15 schools); weak = neither (6 schools). Values are the number of institutions meeting the indicator / total within the subgroup (%). Because of the small sample, confidence intervals are wide; results are exploratory and cannot be interpreted causally.
Table 7.
Proportion with Strong Governance by Exogenous Structural Characteristics (N/N, %) — Exploratory Descriptive Analysis
Exogenous characteristic
Category
Strong governance n/N (%)
Admission quota (2026)
Small (≤49)
6/7 (86)
Medium (50–99)
4/6 (67)
Large (≥100)
5/8 (62)
Establishment type
National
4/7 (57)
Private
11/14 (79)
Region
Seoul Metropolitan Area
7/9 (78)
Non-metropolitan
8/12 (67)
No. of tertiary hospitals
None (0)
1/2 (50)
1
8/12 (67)
≥2
6/7 (86)
Mean admission quota was 78.8 in the strong-governance group vs. 92.7 in the weak-governance group. Admission quota is based on the 2026 academic year (reverted to 3,058); tertiary (advanced general) hospitals are based on the MOHW 5th-cycle (2024–2026) designation. No exogenous characteristic showed a consistent association with the proportion having strong governance (exploratory, not causal).
References
1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn't. BMJ 1996; 312: 71-2.
3. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, Hopayian K, et al. Sicily statement on evidence-based practice. BMC Med Educ 2005; 5: 1-7.
4. Ministry of Education. Enforcement Decree of the Higher Education Act [Internet]. Sejong: Ministry of Education; 2023 [cited 2026 08 15]. Available from https://www.law.go.kr/
7. Lee SH, Ahn J, Lee MJ, Ha CW. Development and operation of a learner-centered integrated 6-year medical program. Korean Med Educ Rev 2025; 27: 132-43.
8. Kim K. A case of student-centered integrated 6-year medical curriculum development: early implementation and challenges. Korean Med Educ Rev 2025; 27: 154-61.
9. Yoo DM, Kang WS. A practical approach: reform of the 6-year medical curriculum at the Catholic University of Korea. Korean Med Educ Rev 2025; 27: 162-8.
10. Ministry of Health and Welfare. Designation of 5th-cycle (2024-2026) tertiary (advanced general) hospitals [Internet]. Sejong: Ministry of Health and Welfare; 2025 [cited 2026 08 15]. Available from https://www.mohw.go.kr/
Strong governance = having a dedicated unit or an integrated methodology–clinical design (15 schools); weak = neither (6 schools). Values are the number of institutions meeting the indicator / total within the subgroup (%). Because of the small sample, confidence intervals are wide; results are exploratory and cannot be interpreted causally.
Table 7. Proportion with Strong Governance by Exogenous Structural Characteristics (N/N, %) — Exploratory Descriptive Analysis
Mean admission quota was 78.8 in the strong-governance group vs. 92.7 in the weak-governance group. Admission quota is based on the 2026 academic year (reverted to 3,058); tertiary (advanced general) hospitals are based on the MOHW 5th-cycle (2024–2026) designation. No exogenous characteristic showed a consistent association with the proportion having strong governance (exploratory, not causal).
