Between principle and practice: a qualitative study of perceived relevance and impact of the Ethiopian national medical licensing examination Authors: Lalem Menber Belay, Matthias Siebeck, Tegbar Yigzaw Sendekie, Martin R Fischer, Markus Berndt Published: 2026 ABSTRACT AbstractBackgroundIn 2019, Ethiopia introduced the National Medical Licensing Examination (NMLE) to standardize medical competence, enhance accountability in education, and ensure patient safety. The effectiveness of such high-stakes exams is significantly influenced by the perceptions of key stakeholders, including medical students, faculty, and school deans. This study investigates their views on the relevance and impact of the NMLE.MethodsA qualitative descriptive study was conducted with a stratified sample of public and private medical schools. We conducted eight key informant interviews (KIIs) with deans and eleven focus group discussions (FGDs); five with medical faculty and six with graduating medical students. Interviews and FGDs were audio-recorded, transcribed verbatim, coded, and analysed thematically. Data were coded in MAXQDA24 using pre-identified themes and open coding to include newly identified themes.ResultsThree main themes and eighteen subthemes were identified. The NMLE was generally recognized as an important tool for establishing minimum competencies, standardizing medical education, and building public trust, particularly among faculty and deans. However, students expressed concerns about redundancy with other assessments and relying solely on knowledge-based assessments. Faculty and deans acknowledged the exam’s role in quality assurance and institutional benchmarking but highlighted design flaws and a lack of practical assessments. While faculty expressed trust in the exam’s intent, students expressed scepticism due to perceived imposition and transparency issues. The exam encouraged learning and prompted curricular changes, but also resulted in significant anxiety, stress, and a delayed entry into the workforce. Students felt their institutions provided inadequate support. There is strong consensus among study participants on the need for reform on the exam: adding a practical component, applying a stepwise assessment model, reducing redundancy, and improving transparency.ConclusionWhile the NMLE is recognised for promoting educational quality, ensuring minimum competence and safeguarding patient safety, it is considered inadequate for fully assessing physician competence and redundant with other assessments. The findings highlight gaps between policy intent and implementation, which require collaborative dialogue among stakeholders to co-create meaningful improvements in the assessment and ensure effective policy implementation.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12909-026-09410-5.Keywords: National Medical Licensing Examination, Medical education, Competence assessment, Stakeholder perceptions, Ethiopia Background The National Medical Licensing Examination (NMLE) is a critical gatekeeper that ensures only qualified individuals enter clinical practice. By verifying that graduates possess the essential knowledge, skills, and professional attitudes, these high-stakes assessments fundamentally aim to safeguard public health [1–4]. In low- and middle-income countries (LMICs), NMLEs are used to strengthen educational quality, promote institutional accountability, and standardize training outcomes [5–8]. To fulfil these roles, regulatory bodies use a range of assessment methods. Written formats such as single-best-answer multiple-choice questions provide a reliable and efficient evaluation of knowledge [9], while Objective Structured Clinical Examination (OSCE) is used to evaluate clinical and communication skills in a standardized setting [10]. Other approaches, including computerized adaptive testing and workplace-based assessments, aim to enhance efficiency and authenticity, though each presents limitations in scalability, standardization, or feasibility [11, 12]. Consequently, licensing systems vary across countries, reflecting differences in regulatory priorities and system capacity. Examples from high-income settings, such as the United States Medical Licensing Examination (USMLE), are organized into three steps: Step 1 (basic science), Step 2 (clinical knowledge), and Step 3 (application) [13]. The UK’s Professional and Linguistic Assessments Board (PLAB) test for international graduates separates knowledge testing from clinical skills testing [14]. Canada’s Medical Council of Canada qualifying examination (MCCQE) also has two parts and emphasizes competencies such as health advocacy and communication [15]. Germany employs a state-run, multi-stage assessment fully integrated into undergraduate medical education, culminating in an oral and practical examination [16]. Despite the intended benefits, implementing national licensing examinations remains controversial [17]. Evidence highlights important challenges, including psychometric limitations, substantial resource requirements, and unintended consequences such as curriculum distortion, “teaching to the test,” faculty burden, and increased stress among students [4, 18]. High-stakes exams can influence learning approaches, sometimes at the expense of deeper professional development [15, 19]. These challenges, particularly for resource-intensive formats, raise concerns about feasibility, equity, and scalability, especially in LMICs [20–22]. Most existing evidence on licensing examinations is from high-income countries, raising concerns about its applicability to the LMIC context, where medical education and regulation differ significantly. Many LMICs have rapidly expanded medical schools with limited regulatory oversight and variable educational quality [23, 24]. These contextual differences make findings from high-income countries less applicable to LMIC contexts. However, empirical evidence assessing how licensing examinations function in LMICs remains limited. Emerging studies from LMICs indicate that regulatory systems often face resource constraints and challenges in ensuring quality and accountability [23, 25, 26]. Quantitative evidence from Ethiopia shows variability in examination outcomes, including differences in pass rates across institutions and candidate groups, raising concerns about equity and preparedness [27]. While these studies provide important insights into examination performance, they rely on quantitative approaches and offer a limited understanding of how the NMLE is experienced by key stakeholders. Ethiopia introduced the NMLE in 2019 to ensure minimum competency, improve healthcare quality, and standardize medical education amid the rapid expansion of training institutions [28]. Although early evidence suggests that licensing examinations can positively influence educational quality and clinical performance [6], their broader impact is complex [6, 29, 30]. High-stakes examinations may also lead to unintended consequences, including exam-focused teaching, increased anxiety, and potential misalignment with local health system needs [2, 31, 32]. Since its implementation, however, there has been limited exploration of how the Ethiopian NMLE influences educational practices, student learning, and workforce preparedness from the perspective of those directly involved. This highlights a critical evidence gap in policy implementation. While quantitative studies have examined psychometric properties and performance outcomes, there is a lack of qualitative evidence exploring the perceptions and experiences of key stakeholders, particularly faculty and students. Understanding these perspectives is essential to assessing whether the NMLE is achieving its intended goals or generating unintended challenges within the education and health systems. This study, therefore, examines stakeholders’ perceptions of the relevance and impact of the Ethiopian NMLE on medical education and workforce preparedness. By providing qualitative insights from an underrepresented LMIC context, it aims to inform policy, support alignment with competency-based training, and contribute to a more contextually grounded understanding of licensing examinations. Research questions What do stakeholders perceive as the purpose and relevance of the NMLE in Ethiopia? What do stakeholders perceive as the purpose and relevance of the NMLE in Ethiopia? How does the NMLE shape students’ preparation and competency development? How does the NMLE shape students’ preparation and competency development? How does the NMLE influence the teaching and learning practices in medical schools? How does the NMLE influence the teaching and learning practices in medical schools? What changes do stakeholders recommend for improving the NMLE? What changes do stakeholders recommend for improving the NMLE? Methods Study setting and period This multi-institutional national study was conducted between August and September 2024 across medical schools in Ethiopia to assess stakeholders’ perceptions of the national medical licensing examination (NMLE). Ethiopia adopted a competency-based medical training, a six-year program with institutional qualification examinations (pre-clinical and pre-internship), though implementation varies by institution. At the national level, the Ministry of Health administers licensing exams for health graduates following completion of internship or after graduation. In parallel, the Ministry of Education introduced a national exit examination for all programs in 2023, which is required for graduation. The NMLE is administered three times a year and is mainly a written assessment, although OSCE has been recently added to the Anesthesia licensing exam. The exam is developed by experts through a workshop using a competency-based blueprint derived from task analysis. The blueprint contains competency domains, health problems or tasks, performance levels, task components, and test formats [33]. In Ethiopia, about 2,000 to 2,500 medical doctors graduate annually, though the number varies each year. Pass rates for the NMLE differ by institution and year; in 2024, data from 19 medical schools showed a pass rate of 98.5% [24], with 10 schools achieving a 100% pass rate. Study design A qualitative descriptive study was conducted using FGDs and KIIs to explore stakeholders’ perceptions of the relevance and impact of the NMLE. This design allowed us to get an in-depth and contextually grounded understanding of participants’ experiences and perspectives, while producing findings that are relevant to policy and practice. Research philosophy The study was guided by a pragmatic paradigm, which informed the selection of methods based on their suitability for addressing the research questions and generating actionable insights. Data collection approaches were tailored to participant roles. Focus group discussions were conducted with students and faculty to capture shared experiences and group dynamics, while in-depth interviews were conducted with deans to explore leadership perspectives that may be less openly discussed in group settings. Participant selection and sampling The study included graduating medical students, medical faculty members, and school deans. Recognizing variations in resources, instructional quality, and institutional maturity across medical schools, a stratified sampling approach was applied. Medical schools were first categorized by ownership (public or private) and institutional seniority (established before or after 2000). Established schools typically have better infrastructure and faculty, while new schools often face shortages. Additionally, private institutions may lack sufficient permanent staff and resources. Within selected institutions, purposive sampling was used to ensure diversity and representation. Efforts were made to achieve gender balance among student participants. Faculty participants were selected to reflect a range of professional backgrounds, including biomedical sciences, clinical disciplines, and public health, as well as gender diversity. Potential participants were identified by faculty members who facilitated data collection at the respective medical schools. Initial contact was made through these faculty members, who provided an explanation of the study’s purpose, its voluntary nature, and the expected time commitment. Interested participants received detailed information about the study’s objectives, procedures, and ethical considerations from the FGD facilitator. FGD schedules were coordinated to accommodate participants’ availability. A total of 11 focus group discussions (FGDs) were conducted with 86 participants. Among these, 5 FGDs were conducted with medical faculty members, comprising 38 participants, of whom 11 were female. The remaining 6 FGDs were conducted with graduating medical students, comprising 48 participants, 22 of whom were female. Notably, four of the eleven FGDs were held at a private medical college, comprising 2 FGDs with medical faculty and 2 with graduating medical students. Additionally, 8 key informant interviews (KIIs) were conducted, including 2 from private medical schools (Table 1). All participants were recruited on a voluntary basis and provided informed consent. Table 1Demographic characteristics of the interviewees (n = 8) and focus group discussion participants (n = 86)Interview participants (n = 8)Focus group participants (n = 86)CharacteristicsTotalCharacteristicsTotalGenderMale5Group of FGD participantsFaculty38Female3Students48Positions heldSchool deans8Graduating medical students by genderMale26Female22Graduating medical students by the ownership type of medical schoolsPublic31Key informants by ownership type of institutionsPublic6Private17private2Graduating medical students by seniority of medical schools (Public only)Old16Young15Faculty by ownership type of medical schoolsPublic22Private16Faculty by professionBiomedical7Clinical21Public health10Faculty by seniority of medical schools (public only)Old13Young9 Demographic characteristics of the interviewees (n = 8) and focus group discussion participants (n = 86) Data collection and quality assurance Data were collected using semi-structured FGD and key informant interview guides developed from the literature [6, 34] and piloted at a medical school not part of the study (see the interview guides in the supplementary file). Eight FGDs were conducted in person at the participants’ respective institutions, while three were conducted virtually due to travel and security constraints. All FGDs and interviews were audio-recorded with participants’ consent and supplemented with field notes. The lead author and an experienced qualitative researcher facilitated the focus group discussions using standardized guides and procedures to ensure consistency across sessions. Trained and experienced medical educators conducted in-depth interviews. All interviews and group discussions were conducted in Amharic and translated into English by experienced transcribers. The lead author checked all transcripts against the audio recordings to ensure accuracy. We assigned codes to participants to maintain anonymity. Medical schools were coded as MS1, MS2, etc., and individual identifiers were assigned as S01, S02, S03, etc., for students, and as F01, F02, F03, etc., for faculty in each FGD. By combining school codes and individual identifiers, we generated a unique code for each participant. For instance, MS1S02 refers to the student participant 02 from Medical School 01. Trustworthiness Multiple strategies strengthened the reliability of the study. Credibility was enhanced through prolonged engagement with participants, iterative questioning, probing for clarification, and daily debriefings between the lead author and data collector. Triangulation was achieved through multiple participant groups (students, faculty, deans) and two data collection methods (FGDs and interviews). Transferability was supported by sampling institutions that reflected the diversity of medical schools in ownership type, seniority, and educational context. Dependability and confirmability were strengthened by maintaining audio recordings, documenting analytic decisions, and ensuring that findings reflected participants’ views rather than the researcher’s assumptions. Data analysis Audio recordings were transcribed verbatim and verified for accuracy. We conducted multiple readings of the transcripts to identify themes and subthemes. The lead author (LMB) conducted the coding and analysis using an iterative thematic analysis approach. This involved familiarization with the data, line-by-line coding, combining similar codes, and organizing related codes into categories and subthemes through constant comparison, memo writing, and repeated coding. Axial coding was used to refine relationships among the categories and identify overarching themes, while keeping the underlying theoretical framework in mind. We used both inductive and deductive approaches. The analysis was guided by a mix of pre-identified themes, such as perceived purpose and consequences, and with open coding to capture new insights from participants’ experiences through an iterative, comparative analytic process. The findings were organized into themes and subthemes, supported by representative quotes. The identified themes were reviewed by the research team. The codebook was generated using MAXQDA24. All the authors agreed on the final theme. MAXQDA 24 software was used to facilitate data management and organization. Reflexivity Five male investigators participated in the study. Three of them, MF, MS, and TYS, are seasoned medical doctors with PhDs and extensive experience in medical education, assessment, and research. MB holds a PhD in educational psychology with strong expertise in research and assessment. LMB is a PhD candidate in medical education with a fellowship in health professions education and leadership from the International FAIMER Institute and an MSc in health professions education. Both TYS and LMB are involved in donor-funded national educational quality improvement projects, including supporting the Ministry of Health with licensing examinations and assisting medical schools to enhance the quality of medical education. These positions and experiences were considered during data collection and analysis to minimize potential influences on data interpretation. MF, MS, TYS, and MB are LMB’s supervisors. Results The analysis identified three main themes and eighteen subthemes that capture the perceptions of medical graduates, faculty, and school deans regarding the Ethiopian NMLE (Table 2). Table 2Themes and subthemes emerged in the analysisThemeSubthemes1. Perceived purpose and relevance of the NMLE1.1. Ensuring minimum competence1.2. Quality control and public trust1.3. Benchmarking institutional performance1.4. Conditional future value1.5. Scepticism of relevance 1.5.1. Redundancy and lack of added value 1.5.2. Exam design and format 1.5.3. Legitimacy and governance issues2. Consequences of introducing the NMLE2.1. The intended consequences 2.1.1. Curricular and pedagogical adaptation 2.1.2. Motivation for learning and teaching2.2. The unintended consequences 2.2.1. The psychological burden: anxiety and burnout 2.2.2. The risk of exam-centered learning 2.2.3. A waste of time and workforce delay2.3. Institutional indifference3. Recommendations for reform in the NMLE3.1. System harmonization3.2. Standardized orientation and blueprinting3.3. Comprehensive exam reform3.4. A shift from testing to nurturing3.5. Link to meaningful reward and employment opportunity Themes and subthemes emerged in the analysis Theme 1: Perceived purpose and relevance of the NMLE In this theme, participants viewed the NMLE as a means to ensure a minimum level of competence, thereby protecting public safety and building trust, as well as standardizing medical training outcomes across various institutions. It is regarded as a potentially strong quality assurance tool that can compare institutional performance and establish a shared national standard. However, they emphasized that the exam’s value depends on improvements. Therefore, the NMLE was viewed both as a necessary accountability tool and as an underdeveloped instrument that needed improvement to fully achieve its purpose. Ensuring minimum competence and standardization A predominant view was that the NMLE’s core function is to assess the minimum level of competence across institutions. All groups agreed that the NMLE aims to assess whether graduates meet a national standard of minimum competence. As one participant stated, “Passing the national medical licensing exam means that you are competent enough to work as a physician” (KI06). This concept of a national standard was emphasized by others, with one noting, “I believe that uniformly assessing all students across the country is beneficial” (MS7F01), and another explaining it as “a way to check whether they have common or basic knowledge” (MS7S05). Quality control and public trust Beyond standardization, the NMLE was viewed as essential to ensure patient safety, serving as a gatekeeper and building societal trust. One key informant strongly endorsed this, stating, “Indeed, introducing a national licensing exam is a very helpful step to ensure patient safety… and to build community trust” (KI01). This sentiment was echoed by others who argued, “I also believe that this [exam] helps ensure that services are provided by qualified individuals” (S7S03) and that “This exam will filter competent students and will improve the quality of the health services” (KI04). Benchmarking institutional performance The NMLE was widely viewed as a mechanism for comparing institutional performance, encouraging competition, and driving improvement, although this view was debated. Many believed that exam pass rates directly reflect the quality of education. As one participant noted, “The presence of the licensing exam helps to see where we are compared to other institutions and review our teaching and learning process” (KI03). Another added, “When our students score high, we take that as evidence that our institution is performing well” (MS9F03). However, a strong counterargument challenged this causal claim, suggesting that performance differences could be due to other factors, such as exam difficulty. One participant pointed out internal inconsistencies in exam performance: “Those who took the exam last year achieved good results… but this year’s batch has not performed well. Yet, both groups received training from the same institution…” (MS9S06). However, another participant stated, “If a lot of students do not pass, it may somehow tell the quality of training” (MS8S01). Conditional future value Many participants across all groups recognized the exam’s purpose and believed it could be beneficial if improved. This perspective viewed the NMLE as a good beginning, with one participant noting, “…through time it is possible to add other factors…there has recently started the OSCE in anaesthesia, so it can be standardized over time” (MS8S01). The value was seen as directly linked to reform, with one key informant stating, “If the questions are modified and practical to the health system instead of being scientific and recall-type questions, it will benefit patients and community” (KI07). Scepticism of relevance A key theme that emerged from the data is the perceived effectiveness of the NMLE as a tool for assessing physician competence. Many participants, especially students, view the exam as redundant and theoretical. However, the leaders are more supportive and emphasize the exam’s role in national standardization. Redundancy and lack of added value The NMLE is viewed as redundant after a more rigorous qualification exam. Students strongly criticised and see it as a waste of time and resources. They questioned its purpose, one participant asking, “I don’t see the point of taking this exam after we take the qualification exam” (MS10S04), and another asserting, “The licensing exam cannot further filter out those who passed the qualification” (MS10S02). Participants expressed concerns about the relevance of the examination to physicians’ practical work. One student noted, “I don’t think it assesses what we will practically do at the workplace” (MS9S05). Another student observed, “…the questions were not about the activities we have been doing as general practitioners” (MS9S08). This led to the belief that “… this test has no value… it is wasting time and resources” (MS10S01 & MS9S05). Faculty and academic leaders also noted an overlap with the existing examinations though they were not as critical as students. One key informant said, “In the fifth year, we already confirmed competence through a qualification exam, which allowed students to do their internship, which creates confusion regarding the licensing exam’s purpose”. KI07. Exam design and format A major concern was the exam’s inability to assess the full spectrum of medical competence. All groups noted that the exam is theoretical and cannot adequately measure the core skill domains for the profession. They emphasized that “Medicine is not only a science, but it is both a science and an art, yet this exam assesses only the aspect of science or the clinical knowledge” (MS4S02). This led to strong recommendations for a more comprehensive approach, such as incorporating OSCE: “The anaesthesia exit exam includes OSCE skill assessment, and I recommend the same to medical schools for better assessment” (KI03). Furthermore, the exam was criticised for item quality. One participant said: “I was so sad after the exam… was it prepared by clinicians? I don’t think so, because there were confusing choices” (MS9S02). Others questioned its fundamental validity, stating, “Its ability to assess our knowledge and skill is way less than the qualification exam we take to enter the internship” (MS10S04) and “…is not that much better than the classroom assessment to measure competence; it is not adequate to measure the competency of a person” (MS3F02). Legitimacy and governance issues Students and some faculty express distrust due to a lack of widespread consensus, variability in exam difficulty, security concerns, and a perception of a top-down imposition without stakeholder consultation, demanding equal understanding from all stakeholders. One participant argued that “It should not be practiced simply because it comes from the top of the government.” (MS9S02). A lack of transparency further erodes trust, as one faculty member pointed out, “One of the challenges is that most people don’t know who prepares or sets the questions for the exam.” (MS9F01). Theme 2: Consequences of introducing the NMLE The introduction of the NMLE is perceived to have a mixed impact on the educational system. While it has promoted some positive changes, it has also led to unintended consequences. The intended consequences Curricular and pedagogical adaptation The NMLE was perceived to influence curricular and pedagogical changes. Faculty and key informants reported that the NMLE has encouraged feedback-driven improvements, such as adopting case-based teaching and frequent quizzes. One faculty member noted, “We are now giving mock exams… apply case-based teaching… we have frequent quizzes to prepare students " (MS3F01). Additionally, a key informant shared that “after identifying poor performance in public health domains from previous batches, they implemented tutorial classes and achieved a 100% pass rate.” KI06. A motivation for learning and teaching Faculty acknowledged that the examination serves as a motivation for both learning and teaching. One faculty member said, “Students are learning diligently…they cannot be careless anymore”. MS3F01. Additionally, some students said the NMLE established a clear “end goal” and motivates revision of core knowledge, with one student describing the connection between clinical work and study, stating, “When I present during morning sessions, the exam comes into my mind; each time I read about a patient, I also think of the exam” (MS8S08). The unintended consequences The NMLE was viewed as a contributing factor to several unintended consequences, including an emphasis on teaching to the test, significant psychological stress among students, and systemic inefficiencies. The psychological burden: anxiety and burnout Students view the exam as a major source of psychological distress and burnout, especially when combined with internship duties. The pressure from multiple exams is overwhelming, with one student describing it using a metaphor: “a small and repetitive strain can lead to fracture of the bone in a while.” (MS3S04). Another student added that they are both mentally and physically drained. The dominant feeling among students was resentment, with the exam framed as a “hassling” bureaucratic hurdle: “It doesn’t have any advantage except that it’s bothering students” (MS3S03). Faculty and key informants also recognized that the exam is a source of significant stress for students. One faculty said, “…they are prone to significant mental distress” MS7F02. Another key informant stated, “The fear of exit and NMLE may have stressed students.” KI02. The risk of exam-centered teaching and learning A key concern that emerged from the discussion was that the NMLE was perceived to narrow the focus of education. Participants worried that the push for high pass rates might compromise holistic development, leading to an emphasis on the theory over the hands-on skills. One faculty member warned, “If it [NMLE] is not comprehensive and only assesses the knowledge part, it motivates the students and institutions to focus only on that part.” (MS3F05). A waste of time and workforce Many participants believed that the NMLE causes an unnecessary delay in entering the workforce while waiting for the exam. Students expressed their frustration, saying, “We have to stay unemployed for a long time until we take this exam” (MS4S03). Many viewed this as a national loss as well, arguing that “the country is losing services…consequently, this leaves it behind” (MS9S02). Critics of the exam have raised concerns about its purpose, suggesting that it might be better to allocate funding to “capacity-building training and allow graduates to work” (MS3S04) instead of exacerbating the shortage of healthcare workers. Institutional indifference Counterbalancing the themes of adaptation and motivation, many participants believed that the NMLE didn’t bring about much change, despite its potential to do so. While the exam was designed to help improve medical education and practice, many believe that there has been no noticeable change in teaching practices or resource allocation. As one faculty member noted, “I don’t believe that licensing has had a major impact on the medical education process” (MS3F03). Another key informant added, “I don’t believe the exam has an impact on teachers…we don’t usually teach for the licensing exam” (KI08). This view was echoed by other educators, with one stating, “I’m still teaching the same way I did before the exam was introduced” (MS9F03). This indifference within the institution also applied to human resource policies, as a key informant remarked, “There hasn’t been any real change in human resource management due to the national licensing exam” (KI01). This suggests that the exam did not necessitate institutions to implement the intended changes. This institutional inertia led to a strong feeling of abandonment among students. They expressed that their institutions did not provide meaningful educational support. One student said that: “We are told to pay and register for the exam, we got the ID for the exam, and nothing else. We depend on ourselves… Nobody will help you” (MS3S07). This lack of guidance made students feel that they were alone in carrying the burden of the high-stakes exam. Student participants also criticized educators for lacking pedagogical skills. They explained that clinical expertise doesn’t necessarily make a doctor a good teacher, and if they have to be teachers, they must take pedagogical courses. The lack of pedagogical skills contributed to a learning environment described by one participant as militaristic, “I believe we are learning in a military-type environment….” (MS4S03), which suggests a rigid, top-down didactic approach that is ill-suited to fostering the critical thinking and clinical reasoning that the NMLE was intended to assess. Theme 3: Recommendations for reform in the NMLE Participants suggested a constructive way forward to enhance the exam’s legitimacy, fairness, relevance, and impact on improving health workforce education, competence, and healthcare quality. System harmonization A key recommendation was for the Ministry of Health (MOH) and the Ministry of Education (MOE) to collaborate closely, rather than making medical education a “battleground” (MS8S07). Another participant said, “Administering two exams has no value than duplication of efforts and waste of resources.” (MS8S02 & KI07). In addition, all groups recommend merging the NMLE with the qualification examinations to avoid redundancy and reduce student burden. One Participant stated, “The NMLE has to be merged with the qualification exam, or either the qualification exam or the licensing exam should be cancelled.” (MS4S03). Standardized orientation and blueprinting Students have repeatedly expressed the need for clear, standardized information about the examination process. They have advocated for mandatory orientation sessions and, importantly, the implementation of a national examination blueprint. One student explained the need for clarity by stating, “Its purpose should be clearly explained, its content should be well-defined, and, as doctors, we need to understand how it will benefit us” (MS10S03). Another argued that a blueprint would create a fair playing ground: “If I have a blueprint clearly showing the content to be addressed, any failure to meet the required standards will be my responsibility, not the exam” (MS8S08). Comprehensive exam reform The most frequent technical recommendation was to redesign the exam. All groups recommend inclusion of an OSCE or practical component and moving to a step-wise assessment, such as the USMLE. As one participant summarized, “It should not be only a written exam… it should also include practical exams” (MS9F05). This was widely shared to ensure the exam fully assesses a physician’s readiness, with another stating, “It should focus on skill and attitude like those abroad, and this would properly assess the physician’s performance and help patients” (MS8S09). A shift from testing to nurturing Participants advocated allocating resources to enhance the educational process, raising questions about the examination’s cost-effectiveness relative to its perceived value in its current form. A consensus emerged about the necessity to reallocate these resources, as one participant stated, “I believe we should prioritize how we teach students better rather than concentrating solely on the exam” (MS3S05). Another participant emphasized, “The focus should be on the process, not the outcome… So, I don’t think it achieved its goals” (MS9F01). Linkage to meaningful rewards and employment opportunities A significant concern was the connection between the NMLE and actual career opportunities. The lack of employment opportunities was a significant demotivating factor among students. One participant stated, “…We would have fought for that advantage. However, since we feel the exam is pointless and not very fulfilling, we tend to approach it with little attention” (MS3S05). The most critical observation highlighted the irrationality of the current situation: “Those who took the licensing exam, we can see our seniors, they are still on the street looking at the board for employment opportunities” (MS3S01). Discussion This study explored stakeholders’ perceptions of the NMLE in Ethiopia, providing important insights into its perceived relevance, implementation, and impact. Overall, participants recognized the NMLE as an important quality assurance mechanism. This supports the purpose of licensing exams, which are fundamentally tools for safeguarding the public trust [35, 36]. However, they expressed concerns regarding its alignment with training, its added value, and its broader educational consequences. An important finding of this study is that stakeholder perceptions significantly influence the legitimacy and effectiveness of high-stakes assessments. Students’ concerns about the relevance of the NMLE suggest that assessments perceived as redundant can undermine engagement and trust. This underscores that the credibility of assessment systems relies on both their technical quality and how they are perceived by learners and educators [18, 37]. The study also contributes to ongoing discussions on assessment models in medical education. The exam’s reliance on written formats reflects a largely psychometric approach, which may inadequately capture the complexity of clinical competence [38]. As Messick [39] argued, validity is not a property of the test instrument, but of the interpretation of test scores and their consequences. Stakeholders question the NMLE’s consequential validity, arguing that a single written exam score may not adequately predict clinical performance and could narrow the focus of learning [35]. Participants’ call for practical assessments supports a shift toward more comprehensive approaches that integrate knowledge and performance, aligning assessments more closely with real-world clinical practice [40, 41]. However, while incorporating practical assessment can enhance validity arguments [42], it requires balancing rigor with feasibility due to its resource-intensive nature. The recent introduction of the OSCE in the anesthesia licensing examination presents an important opportunity to evaluate its feasibility, acceptability, and sustainability within the Ethiopian context. The implementation of the NMLE presents both positive and negative effects. On the one hand, the exam promotes curricular adaptation and enhances students’ motivation to learn. This aligns with the intended role of high-stakes exams in driving learning and institutional improvement [42, 43]. On the other hand, reports of stress and anxiety among students highlight the unintended consequences of assessment systems on learner well-being [44]. The participants’ metaphor of a “repetitive strain” leading to a bone fracture powerfully captures this cumulative toll. This underscores the need to consider mental health as an integral component of assessment design, particularly in contexts where multiple high-stakes examinations coexist. Moreover, the risk of promoting a narrow, “exam-centered” learning culture threatens to undermine the very goal of producing qualified physicians. This phenomenon can be understood through McClelland’s distinction between competence and intelligence [45]. He argued that traditional assessments predict academic success but not real-world outcomes because they measure a narrow form of academic intelligence rather than the ability to apply knowledge, skills, and judgment in complex contexts. When NMLE focuses predominantly on testable knowledge, it risks superficial learning for the test at the expense of holistic understanding. This concern echoes a study by Asemu et al. on the Ethiopian Anesthetists’ licensing exam [46], a global concern that licensing exams can promote “surface learning” at the expense of a deeper understanding [41, 47]. The findings highlight gaps in faculty engagement and institutional support for adapting teaching practices in response to national licensing examinations. Without targeted faculty development and institutional support, assessments may be seen as external impositions rather than as tools to enhance educational quality. Sociocultural theory emphasizes that learning occurs in a community of practice rather than an individual act [48]. For NMLE to improve educational quality, the whole community, faculty, administration, and clinical supervisors must adapt their practices to support the competencies the exam aims to measure. From the system perspective, participants call for greater harmonization to reduce duplication and improve efficiency. The coexistence of the NMLE administered by the Ministry of Health, Ethiopia, and the national exit examination introduced by the Ministry of Education, Ethiopia, raises important questions about redundancy and coherence in assessment policy. The demand for a national blueprint is also a request for transparency, fairness, and alignment, directly addressing the need for a clear construct definition, a fundamental principle of good assessment practice [49]. Applying a sociocultural perspective, the findings highlight that assessment is not an isolated technical process but is embedded within a broader educational ecosystem. Effective assessment reform needs collaboration among students, educators, institutions, and policymakers, as learning and assessment are shaped by social contexts. This echoes a growing consensus that the ultimate goal of medical education is not to perfect the final examination, but to create robust, supportive, and effective educational systems that reliably produce competent graduates [50]. These findings have important implications for policy and practice in LMICs. Strengthening alignment between national examinations and medical training, expanding the use of performance-based assessments, and investing in faculty development and resources are critical to improving the relevance and effectiveness of the NMLE. In addition, efforts to harmonize national assessment systems could reduce redundancy and enhance coherence. Strengths and limitations The study’s strengths include data triangulation across different participant groups (students, faculty, and deans) and the use of two data collection methods (FGDs and interviews). The sample represents the diversity of medical schools across Ethiopia, including participants from public and private, old and new medical schools, thereby ensuring the transferability of the findings. The limitations of this study include reliance on self-reported perceptions and potential social desirability bias. Such bias may lead participants to withhold honest responses due to the sensitive nature of the topic. Additionally, minimizing bias from both investigators and participants is challenging. To address these concerns, we clearly communicated the study’s purpose and assured confidentiality before data collection. Conclusion The study shows that stakeholders recognized the importance of the NMLE in ensuring minimum competence, standardizing medical training, and building public trust. However, there are concerns about its current design and implementation. Addressing these concerns requires a collaborative dialogue among stakeholders to co-create improvements to the exam. Promoting a more transparent and inclusive reform process is crucial to bridging the gap between policy goals and actual practice. Moving forward, a systems approach that aligns the assessment purpose, implementation feasibility, institutional support, and educational practices should guide the process. Such alignment is crucial for the NMLE to fulfil its promise of protecting the public. Supplementary Information Supplementary Material 1. Acknowledgements We would like to thank Jhpiego Ethiopia, a USAID-funded health workforce improvement program (HWIP), for providing financial support. Clinical trial number Not applicable. Abbreviations Center for International Health Ethiopian Medical Association Focus group discussions Key informant interviews Low and middle-income countries Ludwig-Maximilians-Universität Ministry of Education Ministry of Health National medical licensing examination Objective Structure Clinical Examination Principal Investigator Authors’ contributions LMB designed the study, developed study tools, guided data collection, and led data analysis, manuscript development, and revision. MS, MF, MB, and TYS guided the study design, tool development, data analysis, and reviewed the manuscript. All authors read and approved the final manuscript. Funding The study was funded by Jhpiego Ethiopia, a USAID-funded health workforce improvement program (HWIP), but had no role in the design, data collection, analysis, interpretation, and writing of the manuscript. Data availability The datasets used for the analysis are available upon reasonable request. Declarations Ethics approval and consent to participate Ethical clearance was obtained from the Ethiopian Medical Association Institutional Research Ethics Review Committee (Protocol No.: EMA.0061.01) and the Ludwig-Maximilians-Universität Institutional Review Board (project number 24–0856). Before data collection, informed consent was obtained from study participants. We used codes for analysis to maintain participants’ anonymity. All procedures were conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The datasets used for the analysis are available upon reasonable request.