Institutional predictors of minimum proficiency in Brazil’s new National Medical Education Assessment (ENAMED): a Bayesian hierarchical nationwide cross-sectional analysis

Institutional predictors of minimum proficiency in Brazil’s new National Medical Education Assessment (ENAMED): a Bayesian hierarchical nationwide cross-sectional analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Institutional predictors of minimum proficiency in Brazil’s new National Medical Education Assessment (ENAMED): a Bayesian hierarchical nationwide cross-sectional analysis Bruno B. Andrade, Kluass Villalva-Serra, Rodrigo C. Menezes, Quécia H. Brito, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8904720/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Brazil’s medical education system has expanded at unprecedented pacing, largely propelled by policies meant to increase the physician workforce and reduce regional inequities. This expansion succeeded in scale, but it reignited an old question: are we expanding training capacity or expanding diplomas? In 2025, Brazil introduced the National Medical Education Assessment Examination (ENAMED), a regulatory instrument grounded in a simple principle: programs should be judged, and effectively ranked, by the proportion of graduating students who meet a defined minimum standard of professional proficiency (knowledge-based), rather than by mean scores alone. Here, we evaluated institutional and structural determinants of ENAMED performance, focusing on the system’s ability to guarantee minimum proficiency across medical schools amid recent expansion patterns. Methods We conducted a nationwide retrospective observational analysis linking ENAMED-2025 results to institutional and regulatory characteristics of medical programs. Our outcome was systemic assurance of minimum proficiency, defined as meeting a regulatory benchmark in which ≥ 60% of graduates are classified as proficient. We fit hierarchical Bayesian models estimating each program’s posterior probability of meeting this benchmark, accounting for course size and geographic clustering (state and municipality). Results ENAMED revealed pronounced heterogeneity in the probability of meeting the minimum proficiency benchmark across institutional segments. Federal and state public medical schools had the highest assurance, followed by community-based institutions. Private programs, especially for-profit schools, showed substantially lower probabilities of meeting the benchmark. Programs created during the most recent waves of expansion and those with larger volumes of authorized enrollments consistently exhibited weaker assurance. Independently, greater enrollment capacity was associated with lower probability of achieving the ≥ 60% proficiency threshold, with the steepest gradient observed among for-profit institutions. Conclusions ENAMED’s signal is strongly shaped by institutional design and conditions under which expansion occurred, far more than by chance variation. By privileging the share of graduates who reach a minimum proficiency standard, ENAMED shifts regulation from “average performance” to “minimum guarantees,” offering a more defensible lens for accountability in medical education. Its promise, however, depends on policy responses that strengthen academic governance, clinical training capacity, and faculty support, rather than short-term, reactive strategies aimed at gaming exam performance. medical education academic performance education business undergraduate training Brazil Figures Figure 1 Figure 2 Introduction Over the past two decades, medical education in Brazil has been profoundly shaped by an accelerated expansion, with a marked increase in the number of medical schools and training positions 1 – 4 . This growth has been concentrated in the private sector, particularly among for-profit institutions, and was encouraged by federal policies intended to reduce regional disparities in physician supply and expand access to medical training 2 – 5 . Key initiatives included the Programa Mais Médicos (More Doctors Program) 6 , the interiorization of medical schools, the expansion of scholarships for students in private institutions, and regulatory measures aimed at aligning training with the needs of the Brazilian Unified Health System (SUS) 6 – 8 . Although these policies were justified by workforce needs, accumulating evidence indicates that quantitative expansion has not been consistently accompanied by proportional investments in infrastructure, faculty development, and adequate clinical training environments 1 , 2 , 5 , 8 – 11 . Consequently, expansion has had uneven effects on educational quality, reinforcing differences between institutional segments and regions of the country 4 , 5 , 12 . In prior national evaluations, public institutions have generally outperformed private schools, especially for-profit programs, and larger class sizes and accelerated growth in training slots have been associated with worse academic performance, supporting calls for stronger regulatory oversight 4 , 5 . In this context, the Ministry of Education created the National Medical Education Assessment Examination (ENAMED) in April 2025 13 . ENAMED is an annual national examination applied to graduating medical students, designed as a standardized test whose scores are used by graduates to compete for residency positions through a unified national system, integrating formative assessment, institutional accountability, and professional progression. Importantly, ENAMED shifts the evaluative focus from comparing courses by mean scores alone to assessing whether each course ensures a minimum competency floor, operationalized as the proportion of graduating examinees who meet a predefined proficiency standard within the accountability framework. Previous studies have documented structural inequalities between institutional segments and the consequences of accelerated expansion in Brazilian medical education 2 , 4 , 5 , 12 . Despite its relevance, evidence remains limited regarding which institutional characteristics are associated with course-level performance on ENAMED. In particular, it is unclear whether institutional metrics such as administrative category, enrollment capacity, curricular workload, geographic setting, and existing Ministry of Education indicators, relate to the likelihood that a medical course meets the minimum proficiency standard defined by the ENAMED accountability framework. Accordingly, this study aims to identify institutional predictors of satisfactory ENAMED performance and to inform policies for regulating and supervising medical education in Brazil. Methods Study design We conducted a retrospective, observational nationwide analysis of course-level performance in Brazil’s ENAMED using data from the 2025 edition (ENAMED-2025). Data sources and unit of analysis We used publicly available secondary data from Brazil’s National Institute for Educational Studies and Research "Anísio Teixeira (INEP) 14 , a branch of the Brazil’s Ministry of Education (MEC). ENAMED course-level performance metrics were linked to institutional and geographic metadata using official course identifiers. The unit of analysis was each individual medical course (program). Inclusion and exclusion criteria We included all medical courses that participated in ENAMED-2025 and had sufficient information to compute course-level proficiency (number of proficient graduates and total examinees), as well as valid identifiers enabling linkage to institutional and geographic characteristics. Outcomes and operationalization of variables The analysis adopted a system-assurance perspective, focusing on whether each medical course met a predefined minimum performance standard rather than comparing mean scores. Consistent with the INEP accountability framework, the primary outcome was attainment of the MEC “Satisfactory” standard at the course level, operationalized as having ≥ 60% of graduating examinees classified as proficient on ENAMED. Course-level datasets were constructed by cleaning and integrating ENAMED performance results with institutional characteristics, including administrative category, authorized annual enrollment slots, total curricular workload, capital-city location, and official MEC quality indicators, particularly the Preliminary Course Concept (CPC). Variables were derived through harmonization of identifiers and standardization of institutional classifications, and their definitions are provided in the Supplementary Methods . Administrative categories were aggregated a priori to support stable estimation and policy-relevant comparisons. Federal and state public institutions were combined as the reference category given comparable governance structures and higher observed probabilities of meeting the ENAMED minimum proficiency threshold in preliminary analyses. Municipal public institutions and those classified as “special” were grouped due to small counts and similar proficiency distributions. Private institutions were retained as for-profit and non-profit categories, given their substantial representation in the system and their distinct regulatory and governance models. ( Supplementary Table S1 ). Statistical Analysis Primary Outcome and System-Assurance Estimation Framework Analyses were conducted at the medical-course level (the unit of analysis), using ENAMED 2025 course-level counts. Inference focused on system assurance, defined as the posterior probability that a course meets the regulatory benchmark of at least 60% of graduating examinees classified as proficient, rather than comparisons of mean examination scores. Course-level performance was modeled using the number of proficient graduates relative to the total number of examinees, allowing uncertainty to appropriately reflect differences in course size. We report posterior probabilities of meeting the benchmark and summarize system assurance as the expected proportion of courses exceeding the 60% threshold within relevant institutional strata, such as administrative category. Primary modeling framework: Bayesian hierarchical beta-binomial regression We used a Bayesian hierarchical beta-binomial regression to accommodate key features of the data: (i) outcomes are proportions derived from counts (proficient students out of all participants); (ii) course denominators vary substantially (different courses have different numbers of examinees), and (iii) courses are geographically clustered within municipalities and states. Hierarchical modeling supported partial pooling, with the inclusion of random intercepts for state and city-within-state, assisting in the stabilization estimates for smaller courses and small municipalities by borrowing strength from the broader state and city level associations, while also allowing for fully propagating uncertainty into a posterior distribution of probability estimates of meeting the ≥ 60% threshold. Details on Bayesian prior specification and likelihood function used for the analysis are present in the Supplementary Methods Estimation and Posterior-Based Reporting Model estimation and inference were conducted in R using Integrated Nested Laplace Approximation (INLA). Results are presented as posterior means with 95% credible intervals (CrI). System-level summaries and predicted quantities were derived from posterior predictive distributions as described in the Supplementary Methods. System-Level Summaries and Assurance Curves To evaluate how assurance varies across enrollment capacity, we generated assurance curves 𝑝(𝑣), defined as the posterior probability that a course meets the ≥ 60% standard when authorized enrollment slots are set to a given value 𝑣, holding other covariates fixed. Predictions were restricted to empirically supported ranges (5th–95th percentiles of observed enrollment, overall and within strata) to avoid extrapolation. Posterior predictions were generated using 4,000 draws from the fitted model. At each value of v, posterior means and 95% CrI were computed. For interpretability, results were also summarized as (i) the expected change in assurance per + 50 enrollment slots and (ii) the proportion of posterior draws indicating a negative slope. Secondary analyses: mean scores, subgroup model, and concordance with CPC To provide complementary context, we modeled ENAMED mean scores (overall and by clinical sub-area) using beta-family regression after rescaling scores to the unit interval. Given prior evidence of greater heterogeneity among for-profit private schools, we also estimated a stratified hierarchical beta-binomial model within this subgroup, including an indicator for affiliation with private educational conglomerates. Concordance between ENAMED-based classifications and MEC/CPC categories was evaluated using two complementary approaches: (i) a Dirichlet posterior model applied to the full 5×5 joint classification table to account for uncertainty in cell probabilities, and (ii) quadratic-weighted kappa (QWK) statistics to quantify agreement while penalizing larger discrepancies more heavily. Sensitivity Analyses and Model Diagnostics Model adequacy was assessed using leave-one-out predictive diagnostics available in INLA. Because the relationship between enrollment size and ENAMED performance is not assumed to be linear a priori, multiple model specifications were evaluated, including linear effects, interaction terms with institutional category, and nonlinear smooth effects using second-order random-walk (RW2). Model selection was based on lowest WAIC and verified numerical stability. More details on diagnostics and model selection are presented in the Supplementary Methods , Supplementary Figure S1 and Supplementary Table S2 . Results The analytic cohort included 346 medical courses with valid ENAMED proficiency data: 116 (33.5%) Public (Federal/State) institutions; 36 (10.4%) Community/Confessional; 63 (18.2%) Non-Profit Private; 112 (32.4%) For-Profit Private; and 19 (5.5%) Public (Municipal/Special). Model comparison results supported a parsimonious linear specification for enrollment slots ( Supplementary Methods and Supplementary Table S2 ). Institutional Characteristics Associated with Proportion of Proficient Medical Graduates In the adjusted hierarchical model (including authorized enrollment slots, curricular workload, CPC rating, capital-city location, and policy period), administrative category was the strongest predictor of meeting the ≥ 60% proficiency standard (Table 1). Relative to Public (Federal + State) courses, the odds of meeting the standard were lower for Community/Confessional courses (OR 0.67; 95% CrI 0.52–0.88), Non-Profit Private courses (OR 0.44; 0.36–0.55), Public (Municipal/Special) courses (OR 0.34; 0.25–0.47), and For-Profit Private courses (OR 0.31; 0.26–0.38). These relative differences were consistent with the absolute assurance gradients observed in Fig. 1 and Table 2. Among covariates, courses established after the More Doctors Program policy milestone (≥ year 2014) had lower odds of meeting the standard (OR 0.62; 0.53–0.73). Table 1 Institutional Characteristics Associated with Proportion of Proficient Medical Graduates Institution Characteristics: Posterior Coefficient Mean [95CrI] Odds ratio (OR) Mean [95 CrI] Institutional type (vs. Public [Federal + State]): Community / Confessional -0.395 [-0.656, -0.133] 0.674 [0.52, 0.88] For-profit private -1.163 [-1.351, -0.972] 0.313 [0.26, 0.38] Non-profit private -0.817 [-1.036, -0.597] 0.442 [0.36, 0.55] Public (Municipal / Special) -1.081 [-1.397, -0.765] 0.339 [0.25, 0.47] MEC score (vs. CPC 1): CPC = 2 -0.667 [-1.243, -0.091] 0.513 [0.29, 0.91] CPC = 3 -0.08 [-0.615, 0.457] 0.924 [0.54, 1.58] CPC = 4 0.462 [-0.08, 1.005] 1.588 [0.92, 2.73] CPC = 5 0.528 [-0.067, 1.123] 1.696 [0.94, 3.08] Capital-city location (yes vs. no) 0.104 [-0.09, 0.296] 1.11 [0.91, 1.34] Post–More Doctors Program period (≥ year 2014) -0.472 [-0.627, -0.316] 0.624 [0.53, 0.73] Curricular workload (per 1,000 hours) -0.065 [-0.168, 0.038] 0.937 [0.85, 1.04] Enrollment Slots -0.111 [-0.234, 0.012] 0.894 [0.79, 1.01] Table 1 Note: Table 1 reports adjusted fixed effects from the primary Bayesian hierarchical beta-binomial model. Effects are shown on the log-odds scale ( ) and as odds ratios (ORs) with 95% credible intervals. The reference category for institutional type is Public (Federal + State), while for MEC Score is the lowest CPC grading (1). Table 2 Probability of achieving a satisfactory ENAMED score by institutional category Institutional Category Number of Courses Probability of Satisfactory Score (ENAMED > = 3) Overall 346 73.4[69.4% − 77.5%] Public (Federal + State) 116 99.4[97.4% − 100.0%] Community/Confessional 36 95.2[86.1% − 100.0%] Non-Profit Private 63 76.6[66.7% − 85.7%] For-Profit Private 112 43.5[33.0% − 53.6%] Public (Municipal/Special) 19 39.5[21.1% − 57.9%] Table 2 Note: Table 2 summarizes the posterior mean probability (with 95% credible interval) that a medical course meets the INEP “Satisfactory” standard (≥60% proficient graduates) by institutional category and reports the number of courses in each category. Of note, MEC/CPC score displayed limited monotonic alignment with ENAMED-based proficiency assurance, with only CPC = 2 (vs CPC = 1) being consistently associated with lower odds of meeting proficiency (OR 0.51; 0.29 to 0.91), whereas CPC = 3 to CPC = 5 estimates were imprecise and compatible with both modest decreases and increases in odds. Other institutional characteristics showed weaker evidence of association with ≥ 60% proficiency. For each additional 1,000 hours of curricular workload in a given course, the adjusted association was close to null (OR 0.94; 95% CrI 0.85–1.04). Capital-city location was also not strongly associated (OR 1.11; 0.91–1.34). Finally, each + 100 increase in authorized enrollment slots was associated with a modest reduction in the odds of meeting the ≥ 60% proficiency benchmark (OR 0.89; 0.79–1.01), although the uncertainty still overlaps the null. Probability of achieving a satisfactory ENAMED score by institutional category System assurance differed sharply across institutional categories (Table 2). Overall, the posterior mean probability that a course meets the “Satisfactory” threshold was 73.4% (69.4% to 77.5%). Public (Federal/State) courses exhibited near-universal assurance at 99.4% (97.4% to 100.0%), with Community/Confessional courses also high at 95.2% (86.1% to 100.0%). In contrast, Non-Profit Private courses demonstrated intermediate assurance at 76.6% (66.7% to 85.7%), while For-Profit Private and Public (Municipal/Special) courses showed substantially lower assurance at 43.5% (33.0% to 53.6%) and 39.5% (21.1% to 57.9%), respectively, indicating that in these segments the average probability of meeting the national benchmark was below 50%. Bayesian Assurance Curves according to Enrollment capacity Across the medical courses, higher authorized enrollment was generally associated with lower system assurance of reaching the proficiency threshold (Table 3). Overall, the posterior mean change was − 2.6% per + 50 slots (95% CrI − 5.8 to + 0.2), with strong posterior probability for a negative trend (Pr[decrease] = 95.2%). Table 3 Probability of achieving a satisfactory ENAMED score by institutional category Institutional category Slot range evaluated Estimated change per + 50 slots Posterior probability of decrease Overall 40–210 -2.6 pp (-5.6, + 0.2) 95.2% For-Profit Private 50–250 -5.0 pp (-10.9, + 0.5) 95.0% Non-Profit Private 60–300 -2.6 pp (-5.9, + 0.1) 93.2% Public (Municipal/Special) 50–200 -3.5 pp (-10.3, + 0.0) 79.1% Community/Confessional 40–200 -1.9 pp (-6.2, + 0.0) 76.3% Public (Federal + State) 30–170 -0.3 pp (-1.4, + 0.0) 52.2% Table 3 Note: Table 3 summarizes the posterior mean change in assurance per +50 slots (percentage points) and the posterior probability that the trend is negative (Pr[decrease] = Pr[slope < 0 | data],). Assurance is defined as the course-weighted probability that a randomly selected medical course meets the ENAMED satisfactory performance threshold (≥60% proficient). For each group, the posterior assurance curve π( v ) was computed over the same slot grid used in Figure 2, by varying enrollment slots v , while holding all other course-level covariates constant at their observed values. A linear slope was estimated within each posterior draw by regressing π(v) on slots v across all grid points. The decline was most likely among For-Profit Private courses: -5.0% per + 50 slots (95% CrI − 10.8 to + 0.4), with Pr[decrease] = 95.0%. Non-Profit Private courses also showed high likelihood of declining assurance with increasing enrollment (-2.9% per + 50 slots; -7.2 to + 0.1; Pr[decrease] = 93.2%). Notably, these groups also experienced the highest range (5th to 95th quartiles) among yearly enrollment capacity, ranging from 50 to 250 among for-profit, and 60 to 300 in non-profit private schools (Table 3). In contrast, Public (Federal + State) courses showed minimal change across the observed range (− 0.3 per + 50 slots; −1.2 to + 0.0; Pr[decrease] = 52.3%) (Fig. 2). Although credible intervals for the per-50-slot change often included 0 (indicating uncertainty about exact magnitude), posterior probabilities of decrease were high for the overall system and the private sector strata, implying that scaling up enrollment has an important likelihood of being associated with reduced probability of meeting the proficiency benchmark. Concordance Between MEC/CPC Scores and ENAMED Evaluation Agreement between ENAMED and CPC classifications was limited (Table 4). Overall, the probability of exact agreement was estimated to be 32.9% (95% CrI 28.1% to 37.7%), and the quadratic-weighted kappa was 0.31, consistent with only fair concordance beyond chance. Directional disagreement was asymmetric: CPC rated courses higher than ENAMED by at least one level in 38.2% of cases (33.2% to 43.1%), whereas ENAMED was rated higher than CPC in 28.9% (24.3% to 33.8%). Severe disagreement (≥ 2 levels) was also more common in the CPC-higher direction (13.0% [9.8% to 16.7%]) than in the ENAMED-higher direction (4.6% [2.7% to 7.1%]), and the mean category difference was negative (Δ = -0.19; -0.32 to -0.07), indicating net CPC inflation relative to ENAMED on average. Table 4 Concordance and directional disagreement between MEC (CPC) and ENAMED categories Institutional category Agreement Direction of disagreement (≥ 1 level) Severe disagreement (≥ 2 levels) Summary Exact agreement CPC higher (≥ 1 level) ENAMED Higher (≥ 1 level) CPC Higher (≥ 2 levels) ENAMED Higher (≥ 2 levels) Mean Δ (ENAMED − CPC) Quadratic-weighted κappa Overall 32.9% [28.1%–37.7%] 38.2% [33.2%–43.1%] 28.9% [24.3%–33.8%] 13.0% [9.8%–16.7%] 4.6% [2.7%–7.1%] -0.19 (-0.32, -0.07) 0.31 For-Profit Private 25.6% [18.3%–33.4%] 66.2% [57.8%–74.4%] 8.1% [3.9%–13.5%] 29.6% [21.9%–38.0%] 2.4% [0.5%–5.8%] -0.92 (-1.12, -0.71) 0.22 Public (Municipal/Special) 16.4% [5.4%–32.0%] 51.0% [32.6%–69.4%] 32.6% [16.7%–50.7%] 18.1% [6.5%–34.7%] 10.8% [2.4%–24.6%] -0.26 (-0.82, 0.33) 0.51 Non-Profit Private 32.5% [22.2%–43.4%] 49.1% [37.8%–60.3%] 18.4% [10.8%–27.7%] 12.0% [5.7%–20.1%] 4.0% [0.8%–9.4%] -0.40 (-0.66, -0.14) 0.48 Community/Confessional 44.2% [30.9%–58.0%] 20.7% [10.6%–32.8%] 35.1% [22.6%–48.9%] 6.3% [1.3%–14.5%] 10.3% [3.6%–20.4%] 0.20 (-0.16, 0.55) 0.14 Public (Federal + State) 34.2% [26.2%–43.0%] 8.2% [4.0%–13.8%] 57.5% [48.3%–66.1%] 3.3% [0.9%–7.1%] 11.5% [6.4%–17.7%] 0.59 (0.41, 0.77) 0.33 Table 4 Legend: Table 4 summarizes agreement between ENAMED and CPC ordinal categories (1-5), including exact agreement, directional disagreement (CPC higher vs ENAMED higher by >=1 level), severe disagreement (>=2 levels), mean difference Delta (ENAMED - CPC), and quadratic-weighted kappa (QWK). Proportions and mean Delta are reported as posterior means with 95% credible intervals from a Dirichlet posterior on the 5x5 joint table. This misalignment was most pronounced among For-Profit Private courses. In this segment, there was a 66.2% [57.8% to 74.4%] probability that CPC exceeded ENAMED by at least one level and 29.6% [21.9% to 38.0%] by at least two levels, with a large negative mean Delta of -0.92 (-1.12 to -0.71) and kappa of 0.22. By contrast, among Public (Federal/State) courses, ENAMED more frequently exceeded CPC (57.5% [48.3% to 66.1%]) with a positive mean Delta of + 0.59 (0.41 to 0.77). Together, these patterns suggest that CPC is not a reliable proxy for ENAMED-based competency assurance and may systematically overrate performance in the for-profit sector. Subgroup Analysis – Bayesian Model Specific to For-Profit Schools Within the For-Profit Private subgroup, two predictors showed clear evidence of association with lower proficiency assurance ( Supplementary Table S3 ). Curricular workload (per 1,000 hours) was negatively associated with proficiency odds (OR 0.79; 95% CrI 0.66 to 0.95), and the post More-Doctors Program period (> 2013) was strongly associated with reduced proficiency levels (OR 0.53; 0.42 to 0.68). Increases in enrollment slots showcases a modest, yet non-significant, negative association with proficiency level. Other predictors were not strongly associated within the subgroup: capital-city location (OR 0.93; 0.71 to 1.23), and private conglomerate affiliation (OR 0.91; 0.73 to 1.13) were compatible with both modest decreases and increases in odds. Discussion This nationwide analysis of ENAMED-2025 revealed marked heterogeneity in course-level attainment of the MEC “Satisfactory” standard across Brazil’s medical education system. After adjustment for enrollment capacity, curricular workload, CPC rating, geographic location, and policy period, the institutional administrative category remained the most consistent predictor of meeting the minimum proficiency threshold. Federal and state public courses showed near-universal assurance of meeting the benchmark, Community/Confessional institutions also performed strongly, whereas non-profit private, municipal/special public, and especially for-profit private courses exhibited markedly lower assurance. These patterns are coherent with Brazil’s broader institutional landscape in higher education, where governance models, funding arrangements, academic density, and the ability to secure stable clinical training environments vary considerably across segments 15 . Because medical education depends on supervised practice, service integration, and infrastructure-intensive learning settings, differences in institutional capacity to sustain faculty careers, integrate with health services, and maintain supervised practice settings plausibly translate into differences in the likelihood of meeting minimum proficiency standards. The superior ENAMED proficiency outcomes observed among public federal and state universities likely reflect structural advantages historically concentrated in this segment. Public institutions are typically tuition-free and constitutionally oriented toward integrating teaching, research, and extension, and they more often concentrate stricto sensu postgraduate programs, full-time doctoral faculty, and university hospitals that support clinical training at scale 16 , 17 . Municipal public institutions, by contrast, often operate on a smaller scale and depend on local fiscal capacity, which may limit long-term investments in faculty consolidation and training environments 18 . Within the private sector, governance and incentives also differ in ways that plausibly shape training conditions. Non-profit institutions, particularly Community and Confessional models, tend to reinvest surpluses in academic activities and may sustain stronger regional embeddedness 19 , 20 . By contrast, for-profit institutions, now the dominant segment, operate under market-oriented expansion strategies and charge some of the highest tuition fees in medical education 16 , 21 , 22 . Although central to the rapid expansion of training positions, this model has raised concerns regarding alignment between accelerated growth and the infrastructure-intensive requirements of medical education 1 , 23 – 25 . In this context, ENAMED’s threshold-based accountability provides a relevant lens for assessing whether these heterogeneous institutional arrangements can consistently ensure a minimum proficiency standard at graduation. Importantly, the observed gradient favoring public institutions emerged despite a prolonged period of fiscal austerity and structural underfunding affecting Brazilian public universities 26 – 29 . This reinforces prior evidence linking institutional governance models to medical training quality. The findings also align with earlier evidence that predominantly private expansion, especially among for-profit programs, has been associated with weaker performance in prior national assessments and with massification processes not adequately supported by proportional growth in infrastructure and training fields 4 , 12 , 24 . Because ENAMED is aligned with the National Curricular Guidelines and with competencies expected for practice in the Brazilian Unified Health System, and because it influences access to residency training, it may render these structural asymmetries more visible than earlier evaluation models 30 , 31 . Our results should also be interpreted against the backdrop of Brazil’s large-scale expansion in medical education, driven by policy initiatives, including the More Doctors Program, intended to increase physician supply and reduce regional disparities 1 , 5 , 32 – 41 . Although expansion goals were framed around workforce equity, the long-term educational consequences have remained contested, particularly given the rapid proliferation of private programs and concerns about regulatory fragility 2 , 36 , 37 . Consistent with these debates, we found that courses established in the post–More Doctors Program expansion period showed a lower likelihood of meeting the ENAMED minimum proficiency standard, a pattern that was especially pronounced in the for-profit private sector. This suggests that in parts of the system, the pace of expansion may have outstripped regulatory capacity and the practical training infrastructure required for competency-based graduation. The association between higher authorized enrollment and lower assurance, most evident in private-sector strata, points in the same direction. While the magnitude varied, the posterior evidence supports the premise that scaling enrollment is not neutral: when expansion is not accompanied by proportional growth in clinical placement capacity, supervision, and training infrastructure, the probability of meeting a minimum proficiency benchmark declines. This is consistent with prior work linking accelerated expansion and larger class sizes to poorer performance in national assessments (4), and with evidence describing constraints in practice-field allocation, school–service coordination, and insufficient ratios of hospital beds per student, all of which can undermine the supervised learning environment essential for internship quality 1 , 4 , 20 , 33 , 42 – 44 . A further policy-relevant finding is the limited alignment between ENAMED-based threshold attainment and CPC categories. CPC is a multidimensional composite that incorporates ENADE performance, faculty characteristics, and student-reported measures of infrastructure and pedagogical resources 45 , 46 . However, the modest concordance observed here suggests that CPC may not reliably detect whether a program ensures that a sufficient proportion of graduates reach a minimum proficiency floor. If regulatory oversight relies primarily on CPC, it may fail in two directions: it may overlook programs that appear structurally adequate yet do not achieve minimum proficiency assurance, and it may penalize programs that meet the proficiency floor while scoring modestly on other components. The presence of high-CPC programs with low ENAMED proficiency attainment raises questions about whether the previous framework adequately captures professional competence as developed across training, especially in a system where incentives may not map cleanly onto competency assurance. These results also intersect with an evolving regulatory response. Following the first ENAMED cycle, the Ministry of Education adopted graduated supervisory measures targeting low-performing programs, including suspension of new admissions and reductions in authorized enrollment capacity 47 . Subsequently, the federal government revoked the active regulatory instrument governing the authorization of new medical programs, after conducting a broader technical reassessment of expansion policies, considering the effects of judicially driven growth, infrastructure capacity constraints, and the emerging ENAMED result 48 . In parallel, the debate has extended to legislative proposals concerning a National Medical Proficiency Examination as a condition for licensure and to discussions within the Federal Council of Medicine regarding the potential use of ENAMED performance in professional registration decisions 49 . The initial regulatory call had defined procedural and evaluative criteria for authorizing new private-sector medical programs 50 . From an assessment perspective, ENAMED holds potential to become a central mechanism for monitoring medical training in Brazil, particularly if employed as an instrument for ensuring minimum competency standards rather than merely ranking institutions. Although it is primarily a cognitive-theoretical examination, prior studies support the validity of well-designed assessments of this nature in measuring relevant medical competencies. Furthermore, they link low proficiency performance to deficits that extend beyond cognition into clinically meaningful domains 51 , and demonstrate associations between theoretical examination performance, critical thinking measures 52 , and practical assessments such as the Objective Structured Clinical Examination (OSCE) 53 . The implications are practical, as our results support the strengthening of regulatory oversight of expansion, including closer scrutiny of rapid scale increases in segments where assurance declines with enrollment. They also indicate that ENAMED can function as a continuous monitoring instrument for minimum competency assurance, while highlighting the need to reassess how existing indicators, such as CPC, are used in supervision and authorization decisions. Collectively, the findings suggest that expansion without assurance of minimum standards creates systemic risks for health services and for the credibility of training pathways. The results reported here also invite reflection on pathways toward excellence within the current landscape. With regulatory sanctions linked to ENAMED performance, there is a risk that institutions may redirect pedagogical strategies narrowly toward examination outcomes. However, meaningful improvement requires structured academic reconstruction strategies centered on substantive training quality. This includes strengthening internship governance, aligning internal assessment systems with competencies, supporting faculty development, and cultivating institutional cultures grounded in formative mission. In the medium and long term, building academic density, institutional identity, and internal self-regulatory mechanisms remains essential for sustainable improvement, ultimately benefiting institutions, students, patients, and the Brazilian Unified Health System. This study has limitations inherent to its cross-sectional and ecological design, as analyses were conducted at the institutional rather than individual level of the medical graduates. Thus, identified associations should not be interpreted as direct causal relationships and cannot be extrapolated to the student level. Unobserved variables, including students’ socioeconomic background, faculty-to-student ratios, quality of clinical training environments, and degree of integration between teaching and healthcare services, may contribute to the heterogeneity observed. Additionally, ENAMED remains in its early implementation phase, and future methodological adjustments may affect longitudinal comparability. Despite these limitations, the study presents important strengths. It employs national microdata, applies hierarchical Bayesian modeling appropriate for proportion-based outcomes with varying denominators, explicitly incorporates statistical uncertainty, and translates complex findings into metrics directly relevant for regulatory decision-making. By focusing on the probability of achieving a minimum competency threshold, this work shifts the debate from simple comparisons of mean performance toward a framework centered on formative safety and the social accountability of medical education. Conclusion This nationwide analysis of the Brazilian ENAMED-2025 indicates that course-level attainment of the MEC “Satisfactory” standard is strongly patterned by institutional characteristics, particularly administrative category, timing of program implementation, and authorized enrollment scale, revealing a consistent gradient in which federal and state public schools show the highest competency assurance, while private programs, especially for-profit schools established during the recent expansion period, display greater risk of formative insufficiency. By reframing evaluation from mean-score comparisons to the assurance of a minimum competency floor aligned with the National Curricular Guidelines and SUS needs, ENAMED can become a central instrument to guide supervision and expansion policy. However, its value will depend on prioritizing substantive structural improvements in training, such as internship governance, internal competency-based assessment, faculty development, and institutional educational culture, rather than narrow test-oriented responses, reinforcing that minimum-quality assurance in medical education is an ethical and system-level imperative to protect learners, patients, and the SUS. Abbreviations Brazilian Unified Health System (SUS); National Medical Education Assessment Examination (ENAMED); National Institute for Educational Studies and Research "Anísio Teixeira (INEP), Ministry of Education (MED); Preliminary Course Concept (CPC), ntegrated Nested Laplace Approximation (INLA); Credible intervals (CrI); Quadratic-weighted kappa (QWK); Second-order random-walk (RW2); Objective Structured Clinical Examination (OSCE) Declarations Author Contributions BBA contributed to conceptualization, supervision, critical intellectual revision of the manuscript, and project administration. KVS contributed to data curation, formal analysis, and writing – review and editing. RCM contributed to formal analysis and writing – review and editing. QHB contributed to data acquisition and writing – review and editing. LFQ contributed to conceptualization, writing – original draft preparation, and critical intellectual revision. KMA contributed to conceptualization, writing – original draft preparation, critical intellectual revision, and supervision. All authors contributed to the article and approved the submitted version. Ethics statement All data used in this study were obtained from publicly available government databases and did not involve identifiable personal information. Therefore, it did not require submission to the Brazilian Research Ethics Committee, as dictated by Resolution No. 510/2016 of the Brazilian Health Council on norms applicable to Human and Social Sciences research. Consent for publication Not Applicable Funding The study was supported by the Intramural research programs of the Oswaldo Cruz Foundation, Brazil and of the Clariens Educação, Brazil. Acknowledgments Not Applicable Competing Interests The authors declare that they have no competing interests Data/Materials Availability The datasets supporting the conclusions of this article are publicly available in (https://www.gov.br/inep/pt-br/acesso-a-informacao/dados-abertos/microdados/enamed). References Avena KM, Quintanilha LF, Luzardo Filho RL, Andrade BB. Lessons learned from the expansion of medical schools in Brazil: a review of challenges and opportunities. Front Educ (Lausanne). 2024;9:1–10. Andrade BB. The Dark Side of Private Medical Education in Brazil. Front Med (Lausanne). 2025;12:1504794. Scheffer M. Medical Demography in Brazil 2025 [Internet]., Brasilia DF. Ministry of Health of Brazil; 2025 [cited 2025 May 26]. 448 p. Available from: https://bvsms.saude.gov.br/bvs/publicacoes/demografia_medica_brasil_2025.pdf Andrade BB, Villalva-Serra K, Menezes RC, Quintanilha LF, Avena KM. For-Profit Growth and Academic Decline: A Retrospective Nationwide Assessment of Brazilian Medical Schools. Front Med (Lausanne). 2025;12:1617885. Scheffer M, Mosquera P, Cassenote A, McPake B, Russo G. Brazil’s experiment to expand its medical workforce through private and public schools: Impacts and consequences of the balance of regulatory and market forces in resource-scarce settings. Global Health. 2025;21(1). Brazil. Law No. 12,871 of October 22, 2013: Institutes the Mais Médicos Program, amends laws no. 8,745, of december 9, 1993, and no. 6,932, of july 7, 1981, and makes other provisions. [Internet]. Brasília, DF: Presidency of the Republic of Brazil; 2013 [cited 2025 Sep 16]. Available from: https://www.planalto.gov.br/ccivil_03/_ato2011-2014/2013/lei/l12871.htm Machado CV, Silva GA. Political struggles for a universal health system in Brazil: Successes and limits in the reduction of inequalities. Global Health. 2019;15(Suppl 1):1–12. Mattos E, Mazetto D. Assessing the impact of more doctors’ program on healthcare indicators in Brazil. World Dev. 2019;123:104617. Campos VMS, Maggitti-Bezerril M, Menezes RC, Sobral L, Avena KM, Andrade BB. Doctors in the making, or degrees for sale? A student’s view of Brazil’s medical education crisis. Front Med (Lausanne). 2025;12. Federal Council of Medicine. 78% of municipalities that host medical schools do not have adequate infrastructure [Internet]. 2024 [cited 2025 Jan 11]. Available from: https://portal.cfm.org.br/noticias/78-dos-municipios-que-sediam-escolas-medicas-nao-possuem-a-infraestrutura-adequada-para-a-formacao-dos-profissionais Federal Council of Medicine. 73% of municipalities applying to receive medical schools do not have adequate infrastructure [Internet]. 2024 [cited 2025 Jan 11]. Available from: https://portal.cfm.org.br/noticias/73-dos-municipios-candidatos-a-receber-escolas-medicas-nao-tem-infraestrutura-adequada-para-formar-novos-profissionais Fatel QH, Silva FG, Avena KM, Menezes RC, Andrade BB, Quintanilha LF. Shaping Tomorrow’s Doctors: The Impact of Socioeconomic and Institutional Factors on Medical Education Quality in Brazil: An Ecological Study. J Med Educ Curric Dev. 2025;12:1. Brazil, ENAMED - Brazilian Medical Training Assessment Exam [Internet]. Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira (Inep). 2025 [cited 2025 May 26]. Available from: https://www.gov.br/inep/pt-br/areas-de-atuacao/avaliacao-e-exames-educacionais/enamed National Institute of Educational Studies and Research Anísio Teixeira (INEP). INEP: National Institute of Educational Studies and Research Anísio Teixeira [Internet], Brasília DF. 2024 [cited 2024 Apr 3]. Available from: http://www.inep.gov.br/ Gomes MF. Evaluation and administrative nature of the institutions of higher education. Ensaio: Avaliação e Políticas Públicas em Educação. 2010;18(68):589–610. Brazil, National Institute of Educational Studies and Research Anísio Teixeira (INEP). Higher Education Census 2021: statistical notes. Brasília, DF: INEP; 2022. Zorzetto R, Razzouk D, Dubugras MTB, Gerolin J, Schor N, Guimarães JA, et al. The scientific production in health and biological sciences of the top 20 Brazilian universities. Braz J Med Biol Res. 2006;39(12):1513–20. Neves RS. Municipal institutions of higher education. Interesse Público. 2018;20(109):191–210. Fioreze C, McCowan T. Community universities in the South of Brazil: prospects and challenges of a model of non-state public higher education. Comp Educ. 2018;54(3):370–89. Tagliari C. For-profit expansion and nonprofit retraction in Brazilian higher education. Plural. 2022;29(1):36–59. McCowan T. The growth of private higher education in Brazil: implications for issues of equity, quality and public benefit. Educ Policy Anal Arch [Internet]. 2004;12:30. Available from: http://epaa.asu.edu/epaa/v13n27 Carvalho CHA, Sousa Filho EM, Goellner IA, Freire NR, Amitrano CR. Private higher education financialization in Brazil: expansion of medical programs in publicly traded companies (2013–2022). Educ Soc. 2024;45:e286692. Weber CAT, da Silva AG. New medical schools: solution or problem? Debates em Psiquiatria. 2024;14:1–6. Scheffer MC, Poz MRD. The privatization of medical education in Brazil: trends and challenges. Hum Resour Health. 2015;13(1):96. Nassar LM, Couto MHC, Pereira Junior GA. Public funding (FIES and PROUNI) for medical teaching in Brazil: a literature review and the distortions created. Educação em Revista. 2021;37:e25246. Silva S, Leher R. Austerity is jeopardizing the future of public universities and their social function. Temporalis. 2024;24(47):15–34. Amaral NC, Junior WT da, Salles S. Federal universities: autonomy undermined by funding? Caderno CRH. 2025;38:e025020. Silva ARR, Oliveira LRPF, Pinho MJ. The higher public education in political context: reflections concerning the university dismantling process from de 2016. Revista Cocar. 2024;21:39. Nota-Publica-ABC. -SBPC-maio-2025. 2025. Andrade BB, Sasaki KCM, Menezes RC, Luzardo Filho RL, Avena KM. From framework to fitness for the 21st century: How Brazil’s 2025 National Curricular Guidelines recast priorities for training physicians. Front Med (Lausanne). 2026;12. Rodrigues DLG. Challenges and opportunities in the Brazilian Unified Health System: Pathways to sustainability and equity. Public Health. 2025;247. Lopes AC. The numerical explosion in Brazilian medical schools. Educación Médica. 2018;19:19–24. Oliveira BLCA, Lima SF, Pereira MUL, Pereira Júnior GA. Evolution, distribution and expansion of medicine courses in Brazil (1808–2018). Trabalho, Educação e Saúde. 2019;17(1):e0018317. Pereira DVR, Fernandes DLR, Mari JF, Lage ALF, Fernandes APPC. Mapping of medical schools: the distribution of undergraduate courses and annual vacancies in Brazilian cities in 2020. Rev Bras Educ Med. 2021;45(1):1–10. Santos LMP, Oliveira A, Trindade JS, Barreto ICHC, Palmeira PA, Comes Y, et al. Implementation research: towards universal health coverage with more doctors in Brazil. Bull World Health Organ. 2017;95(2):103. Dolci JEL. The proliferation of medical schools in Brazil: a threat to the quality of medical education? Braz J Otorhinolaryngol. 2023;89(6):101354. Oliveira JPA, Sanchez MN, Santos LMP. The Mais Médicos (More Doctors) Program: the placement of physicians in priority municipalities in Brazil from 2013 to 2014. Cien Saude Colet. 2016;21(9):2719–27. Figueiredo AM, McKinley DW, Massuda A, Azevedo GD. Evaluating medical education regulation changes in Brazil: workforce impact. Hum Resour Health. 2021;19(1):1–12. Figueiredo AM, McKinley DW, Lima KC, Azevedo GD. Medical school expansion policies: educational access and physician distribution. Med Educ. 2019;53(11):1121–31. Santos LMP, Costa AM, Girardi SN. Mais Medicos Program: an effective action to reduce health inequities in Brazil. Cien Saude Colet [Internet]. 2015 Jan 1 [cited 2026 Feb 8];20(11):3547–52. Available from: https://www.scielo.br/j/csc/a/mFYpCXL3q4XknsR58Pk5gmS/?lang=en Girardi SN, Van Stralen AC, de Cella S, Der Maas JN, Carvalho LW, de Faria CL. Impact of the Mais Médicos (More Doctors) Program in reducing physician shortage in Brazilian Primary Healthcare. Cien Saude Colet. 2016;21(9):2675–84. Oliveira BLCA, Lima SF, Pereira MUL, Pereira Júnior GA. Evolution, distribution and expansion of medicine courses in Brazil (1808–2018). Trabalho, Educação e Saúde. 2019;17(1):e0018317. Silva FGM, Santos PAJ, Quintanilha LF. Mais Médicos - More Physicians law: evaluating compliance with municipal criteria for the operation of medical schools. Rev Bras Educ Med. 2025;49(2):e067. Morais HMM, Sa RGR, Albuquerque M, do SV, Oliveira RS. Expansion and privatization of medical courses and the integration of teaching and service: the case of the state of Pernambuco. Saúde em Debate. 2023;47(137 abr–jun):182–95. Indicators of higher education quality. 91.6% of courses have a Course Performance Concept (CPC) in 2019 that falls between bands 3 and 5 [Internet]. [cited 2026 Feb 8]. Available from: https://abmes.org.br/noticias/detalhe/4136/indicadores-de-qualidade-da-educacao-superior-cpc-2019-de-91-6-dos-cursos-esta-entre-as-faixas-3-e-5 Preliminary Course Concept (CPC) [Internet]. [cited 2026 Feb 8]. Available from: https://www.gov.br/inep/pt-br/areas-de-atuacao/pesquisas-estatisticas-e-indicadores/indicadores-de-qualidade-da-educacao-superior/conceito-preliminar-de-curso-cpc Brazil E. Medical course evaluations and supervision measures released [Internet]. Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira (INEP). 2026 [cited 2026 Feb 8]. Available from: https://www.gov.br/mec/pt-br/assuntos/noticias/2026/janeiro/enamed-divulgadas-avaliacao-dos-cursos-de-medicina-e-medidas-de-supervisao Brazil. Revocation of Public Notice No. 01, dated October 4, 2023, and its amendments. Ministério da Educação. Conselho Nacional de Educação. Brasilia, DF; 2026. Agência, Senado. A bill creating a medical proficiency exam is in its final stages in the Senate [Internet]. 2026 [cited 2026 Feb 8]. Available from: https://www12.senado.leg.br/noticias/materias/2026/01/26/projeto-que-cria-exame-de-proficiencia-em-medicina-esta-em-fase-final-no-senado Brazil MEC. Notice No. 1/2023. Public Call for Proposals for Authorization to Operate Medical Courses within the More Doctors Program. Brasilia, DF: Ministério da Educação, Conselho Nacional de Educação; 2023. Andreou V, Eggermont J, Gielis G, Schoenmakers B. Proficiency testing for identifying underperforming students before postgraduate education: a longitudinal study. BMC Med Educ. 2020;20(1). Mafinejad MK, Arabshahi SKS, Monajemi A, Jalili M, Soltani A, Rasouli J. Use of Multi-Response Format Test in the Assessment of Medical Students’ Critical Thinking Ability. J Clin Diagn Res. 2017;11(9):LC10–3. Castellani L, Quintanilha LF, Arriaga MB, Lima ML, Andrade BB. Objective Structured Clinical Examination (OSCE) As a Reliable Evaluation Strategy: Evidence From a Brazilian Medical School. Probl Educ 21st Century. 2020;78(5):674–87. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterialENAMED2026.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8904720","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":594641053,"identity":"51a34784-0bee-49dc-aee8-f476bf74df0f","order_by":0,"name":"Bruno B. Andrade","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYBACAzjrAAODBEOFHA+CS5yWM8akamFsM0bi4gDm0mfMHnxgsMvju3344Y2P8wxk+NnPPpP8wXAnH5cWy74cc8MZDMnFkufSjC1nbjPgkexJN5OQYHhm2YDLYWd4zKR5GJgTN5xhMJPm3faHx+AGG5uEAcNhAxw6YFrqgVrYv0n/nWMA0ZJAWMthoBYgg7EBquUAHi2WPWxlkjMMjifOPMNTbNlzDOSXNGbLBoNnOLWY8zBvk/hQUZ3Yd4Z9440fNQb2/OzHGG/+qLiDUwvUeUSIjIJRMApGwSggAQAAICFN5ytEt38AAAAASUVORK5CYII=","orcid":"","institution":"Oswaldo Cruz Foundation","correspondingAuthor":true,"prefix":"","firstName":"Bruno","middleName":"B.","lastName":"Andrade","suffix":""},{"id":594641054,"identity":"c21bf8d5-3943-4e71-965c-dbdf6e65c0bc","order_by":1,"name":"Kluass Villalva-Serra","email":"","orcid":"","institution":"Vanderbilt University","correspondingAuthor":false,"prefix":"","firstName":"Kluass","middleName":"","lastName":"Villalva-Serra","suffix":""},{"id":594641055,"identity":"42c4964c-dc07-401f-8953-224d6110288a","order_by":2,"name":"Rodrigo C. Menezes","email":"","orcid":"","institution":"Oswaldo Cruz Foundation","correspondingAuthor":false,"prefix":"","firstName":"Rodrigo","middleName":"C.","lastName":"Menezes","suffix":""},{"id":594641056,"identity":"ab963271-e799-4649-98fb-08b424e5b490","order_by":3,"name":"Quécia H. Brito","email":"","orcid":"","institution":"Universidade Salvador","correspondingAuthor":false,"prefix":"","firstName":"Quécia","middleName":"H.","lastName":"Brito","suffix":""},{"id":594641057,"identity":"12828204-98be-4625-9002-1e4b547a4fea","order_by":4,"name":"Luiz F. Quintanilha","email":"","orcid":"","institution":"Faculdade Zarns","correspondingAuthor":false,"prefix":"","firstName":"Luiz","middleName":"F.","lastName":"Quintanilha","suffix":""},{"id":594641058,"identity":"caa94265-5edb-4b87-8fa6-2465df7b0c98","order_by":5,"name":"Kátia M. Avena","email":"","orcid":"","institution":"Oswaldo Cruz Foundation","correspondingAuthor":false,"prefix":"","firstName":"Kátia","middleName":"M.","lastName":"Avena","suffix":""}],"badges":[],"createdAt":"2026-02-17 23:38:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8904720/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8904720/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104307826,"identity":"8a6366bb-a6c9-48aa-b9d1-4ce8b591d163","added_by":"auto","created_at":"2026-03-10 10:16:57","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":408932,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProbability of achieving a satisfactory ENAMED score by institutional category\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003edisplays posterior probabilities of meeting the \u0026gt;=60% proficiency threshold across institutional categories. Intervals represent posterior uncertainty (95% credible intervals). This figure emphasizes system assurance: the likelihood that a course meets the INEP’s national competency standard.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8904720/v1/54a9c28798aeec0baf379e2d.png"},{"id":104307825,"identity":"d366c493-b051-4fbf-bf69-1062cbbe52a2","added_by":"auto","created_at":"2026-03-10 10:16:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":343925,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBayesian Assurance Curves for Probability of Meeting INEP’s Proficiency Target by Enrollment Size\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAssurance curves show predicted posterior probabilities of meeting the \u0026gt;=60% proficiency threshold across the evaluated enrollment slot range, stratified by institutional category. Probabilities were obtained from posterior simulations of the fitted Bayesian hierarchical beta-binomial model and represent uncertainty-aware estimates conditional on observed data. All curves are adjusted for institutional administrative category, authorized enrollment slots, curricular workload (per 1,000 hours, mean-centered), prior course quality as measured by the MEC Preliminary Course Concept (CPC), capital-city location, and course establishment in the post–More Doctors policy period (≥2014), with additional random intercepts for state and city-within-state to account for geographic clustering. Shaded bands indicate 95% Bayesian credible intervals.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8904720/v1/1479988c00faa04643105276.png"},{"id":104779686,"identity":"d69903e6-9276-4af0-b2bb-7b1afd9ee7d0","added_by":"auto","created_at":"2026-03-17 07:44:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2218516,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8904720/v1/e8382496-dae9-45a2-b6de-7156cad1c0b6.pdf"},{"id":104405402,"identity":"8c60a385-7049-481e-982a-3053be17fd08","added_by":"auto","created_at":"2026-03-11 12:22:47","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":161893,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialENAMED2026.docx","url":"https://assets-eu.researchsquare.com/files/rs-8904720/v1/1c35b38909ffd3b1782eec90.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eInstitutional predictors of minimum proficiency in Brazil’s new National Medical Education Assessment (ENAMED): a Bayesian hierarchical nationwide cross-sectional analysis\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOver the past two decades, medical education in Brazil has been profoundly shaped by an accelerated expansion, with a marked increase in the number of medical schools and training positions \u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. This growth has been concentrated in the private sector, particularly among for-profit institutions, and was encouraged by federal policies intended to reduce regional disparities in physician supply and expand access to medical training \u003csup\u003e\u003cspan additionalcitationids=\"CR3 CR4\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Key initiatives included the \u003cem\u003ePrograma Mais M\u0026eacute;dicos\u003c/em\u003e (More Doctors Program) \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e, the interiorization of medical schools, the expansion of scholarships for students in private institutions, and regulatory measures aimed at aligning training with the needs of the Brazilian Unified Health System (SUS) \u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAlthough these policies were justified by workforce needs, accumulating evidence indicates that quantitative expansion has not been consistently accompanied by proportional investments in infrastructure, faculty development, and adequate clinical training environments \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Consequently, expansion has had uneven effects on educational quality, reinforcing differences between institutional segments and regions of the country \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. In prior national evaluations, public institutions have generally outperformed private schools, especially for-profit programs, and larger class sizes and accelerated growth in training slots have been associated with worse academic performance, supporting calls for stronger regulatory oversight \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn this context, the Ministry of Education created the National Medical Education Assessment Examination (ENAMED) in April 2025 \u003csup\u003e13\u003c/sup\u003e. ENAMED is an annual national examination applied to graduating medical students, designed as a standardized test whose scores are used by graduates to compete for residency positions through a unified national system, integrating formative assessment, institutional accountability, and professional progression. Importantly, ENAMED shifts the evaluative focus from comparing courses by mean scores alone to assessing whether each course ensures a minimum competency floor, operationalized as the proportion of graduating examinees who meet a predefined proficiency standard within the accountability framework.\u003c/p\u003e \u003cp\u003ePrevious studies have documented structural inequalities between institutional segments and the consequences of accelerated expansion in Brazilian medical education \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Despite its relevance, evidence remains limited regarding which institutional characteristics are associated with course-level performance on ENAMED. In particular, it is unclear whether institutional metrics such as administrative category, enrollment capacity, curricular workload, geographic setting, and existing Ministry of Education indicators, relate to the likelihood that a medical course meets the minimum proficiency standard defined by the ENAMED accountability framework. Accordingly, this study aims to identify institutional predictors of satisfactory ENAMED performance and to inform policies for regulating and supervising medical education in Brazil.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design\u003c/h2\u003e \u003cp\u003eWe conducted a retrospective, observational nationwide analysis of course-level performance in Brazil\u0026rsquo;s ENAMED using data from the 2025 edition (ENAMED-2025).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eData sources and unit of analysis\u003c/h3\u003e\n\u003cp\u003eWe used publicly available secondary data from Brazil\u0026rsquo;s National Institute for Educational Studies and Research \"An\u0026iacute;sio Teixeira (INEP) \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e, a branch of the Brazil\u0026rsquo;s Ministry of Education (MEC). ENAMED course-level performance metrics were linked to institutional and geographic metadata using official course identifiers. The unit of analysis was each individual medical course (program).\u003c/p\u003e\n\u003ch3\u003eInclusion and exclusion criteria\u003c/h3\u003e\n\u003cp\u003eWe included all medical courses that participated in ENAMED-2025 and had sufficient information to compute course-level proficiency (number of proficient graduates and total examinees), as well as valid identifiers enabling linkage to institutional and geographic characteristics.\u003c/p\u003e\n\u003ch3\u003eOutcomes and operationalization of variables\u003c/h3\u003e\n\u003cp\u003eThe analysis adopted a system-assurance perspective, focusing on whether each medical course met a predefined minimum performance standard rather than comparing mean scores. Consistent with the INEP accountability framework, the primary outcome was attainment of the MEC \u0026ldquo;Satisfactory\u0026rdquo; standard at the course level, operationalized as having\u0026thinsp;\u0026ge;\u0026thinsp;60% of graduating examinees classified as proficient on ENAMED.\u003c/p\u003e \u003cp\u003eCourse-level datasets were constructed by cleaning and integrating ENAMED performance results with institutional characteristics, including administrative category, authorized annual enrollment slots, total curricular workload, capital-city location, and official MEC quality indicators, particularly the Preliminary Course Concept (CPC). Variables were derived through harmonization of identifiers and standardization of institutional classifications, and their definitions are provided in the \u003cb\u003eSupplementary Methods\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eAdministrative categories were aggregated a priori to support stable estimation and policy-relevant comparisons. Federal and state public institutions were combined as the reference category given comparable governance structures and higher observed probabilities of meeting the ENAMED minimum proficiency threshold in preliminary analyses. Municipal public institutions and those classified as \u0026ldquo;special\u0026rdquo; were grouped due to small counts and similar proficiency distributions. Private institutions were retained as for-profit and non-profit categories, given their substantial representation in the system and their distinct regulatory and governance models. (\u003cb\u003eSupplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003ePrimary Outcome and System-Assurance Estimation Framework\u003c/h2\u003e \u003cp\u003eAnalyses were conducted at the medical-course level (the unit of analysis), using ENAMED 2025 course-level counts. Inference focused on system assurance, defined as the posterior probability that a course meets the regulatory benchmark of at least 60% of graduating examinees classified as proficient, rather than comparisons of mean examination scores.\u003c/p\u003e \u003cp\u003eCourse-level performance was modeled using the number of proficient graduates relative to the total number of examinees, allowing uncertainty to appropriately reflect differences in course size. We report posterior probabilities of meeting the benchmark and summarize system assurance as the expected proportion of courses exceeding the 60% threshold within relevant institutional strata, such as administrative category.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003ePrimary modeling framework: Bayesian hierarchical beta-binomial regression\u003c/h3\u003e\n\u003cp\u003eWe used a Bayesian hierarchical beta-binomial regression to accommodate key features of the data: (i) outcomes are proportions derived from counts (proficient students out of all participants); (ii) course denominators vary substantially (different courses have different numbers of examinees), and (iii) courses are geographically clustered within municipalities and states. Hierarchical modeling supported partial pooling, with the inclusion of random intercepts for state and city-within-state, assisting in the stabilization estimates for smaller courses and small municipalities by borrowing strength from the broader state and city level associations, while also allowing for fully propagating uncertainty into a posterior distribution of probability estimates of meeting the \u0026ge;\u0026thinsp;60% threshold. Details on Bayesian prior specification and likelihood function used for the analysis are present in the \u003cb\u003eSupplementary Methods\u003c/b\u003e\u003c/p\u003e\n\u003ch3\u003eEstimation and Posterior-Based Reporting\u003c/h3\u003e\n\u003cp\u003eModel estimation and inference were conducted in R using Integrated Nested Laplace Approximation (INLA). Results are presented as posterior means with 95% credible intervals (CrI). System-level summaries and predicted quantities were derived from posterior predictive distributions as described in the \u003cb\u003eSupplementary Methods.\u003c/b\u003e\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSystem-Level Summaries and Assurance Curves\u003c/h2\u003e \u003cp\u003eTo evaluate how assurance varies across enrollment capacity, we generated assurance curves \u0026#119901;(\u0026#119907;), defined as the posterior probability that a course meets the \u0026ge;\u0026thinsp;60% standard when authorized enrollment slots are set to a given value \u0026#119907;, holding other covariates fixed. Predictions were restricted to empirically supported ranges (5th\u0026ndash;95th percentiles of observed enrollment, overall and within strata) to avoid extrapolation.\u003c/p\u003e \u003cp\u003ePosterior predictions were generated using 4,000 draws from the fitted model. At each value of v, posterior means and 95% CrI were computed. For interpretability, results were also summarized as (i) the expected change in assurance per +\u0026thinsp;50 enrollment slots and (ii) the proportion of posterior draws indicating a negative slope.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eSecondary analyses: mean scores, subgroup model, and concordance with CPC\u003c/h2\u003e \u003cp\u003eTo provide complementary context, we modeled ENAMED mean scores (overall and by clinical sub-area) using beta-family regression after rescaling scores to the unit interval. Given prior evidence of greater heterogeneity among for-profit private schools, we also estimated a stratified hierarchical beta-binomial model within this subgroup, including an indicator for affiliation with private educational conglomerates.\u003c/p\u003e \u003cp\u003eConcordance between ENAMED-based classifications and MEC/CPC categories was evaluated using two complementary approaches: (i) a Dirichlet posterior model applied to the full 5\u0026times;5 joint classification table to account for uncertainty in cell probabilities, and (ii) quadratic-weighted kappa (QWK) statistics to quantify agreement while penalizing larger discrepancies more heavily.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eSensitivity Analyses and Model Diagnostics\u003c/h2\u003e \u003cp\u003eModel adequacy was assessed using leave-one-out predictive diagnostics available in INLA. Because the relationship between enrollment size and ENAMED performance is not assumed to be linear a priori, multiple model specifications were evaluated, including linear effects, interaction terms with institutional category, and nonlinear smooth effects using second-order random-walk (RW2). Model selection was based on lowest WAIC and verified numerical stability. More details on diagnostics and model selection are presented in the \u003cb\u003eSupplementary Methods\u003c/b\u003e, \u003cb\u003eSupplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e and \u003cb\u003eSupplementary Table S2\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eThe analytic cohort included 346 medical courses with valid ENAMED proficiency data: 116 (33.5%) Public (Federal/State) institutions; 36 (10.4%) Community/Confessional; 63 (18.2%) Non-Profit Private; 112 (32.4%) For-Profit Private; and 19 (5.5%) Public (Municipal/Special). Model comparison results supported a parsimonious linear specification for enrollment slots (\u003cstrong\u003eSupplementary Methods\u003c/strong\u003e and \u003cstrong\u003eSupplementary Table S2\u003c/strong\u003e).\u003c/p\u003e\n\u003cdiv id=\"Sec15\"\u003e\n \u003ch2\u003eInstitutional Characteristics Associated with Proportion of Proficient Medical Graduates\u003c/h2\u003e\n \u003cp\u003eIn the adjusted hierarchical model (including authorized enrollment slots, curricular workload, CPC rating, capital-city location, and policy period), administrative category was the strongest predictor of meeting the ≥ 60% proficiency standard (Table\u0026nbsp;1). Relative to Public (Federal + State) courses, the odds of meeting the standard were lower for Community/Confessional courses (OR 0.67; 95% CrI 0.52–0.88), Non-Profit Private courses (OR 0.44; 0.36–0.55), Public (Municipal/Special) courses (OR 0.34; 0.25–0.47), and For-Profit Private courses (OR 0.31; 0.26–0.38). These relative differences were consistent with the absolute assurance gradients observed in Fig.\u0026nbsp;1 and Table\u0026nbsp;2. Among covariates, courses established after the More Doctors Program policy milestone (≥ year 2014) had lower odds of meeting the standard (OR 0.62; 0.53–0.73).\u003c/p\u003e\n \u003cdiv\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 1\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eInstitutional Characteristics Associated with Proportion of Proficient Medical Graduates\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eInstitution Characteristics:\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePosterior Coefficient\u003c/p\u003e\n \u003cp\u003eMean [95CrI]\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eOdds ratio (OR)\u003c/p\u003e\n \u003cp\u003eMean [95 CrI]\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInstitutional type (vs. Public [Federal + State]):\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCommunity / Confessional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.395 [-0.656, -0.133]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.674 [0.52, 0.88]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFor-profit private\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-1.163 [-1.351, -0.972]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.313 [0.26, 0.38]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-profit private\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.817 [-1.036, -0.597]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.442 [0.36, 0.55]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePublic (Municipal / Special)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-1.081 [-1.397, -0.765]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.339 [0.25, 0.47]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMEC score (vs. CPC 1):\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCPC = 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.667 [-1.243, -0.091]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.513 [0.29, 0.91]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCPC = 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.08 [-0.615, 0.457]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.924 [0.54, 1.58]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCPC = 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.462 [-0.08, 1.005]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.588 [0.92, 2.73]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCPC = 5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.528 [-0.067, 1.123]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.696 [0.94, 3.08]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCapital-city location (yes vs. no)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.104 [-0.09, 0.296]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.11 [0.91, 1.34]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePost–More Doctors Program period (≥ year 2014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.472 [-0.627, -0.316]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.624 [0.53, 0.73]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCurricular workload (per 1,000 hours)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.065 [-0.168, 0.038]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.937 [0.85, 1.04]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEnrollment Slots\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.111 [-0.234, 0.012]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.894 [0.79, 1.01]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 1 Note:\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eTable 1 reports adjusted fixed effects from the primary Bayesian hierarchical beta-binomial model. Effects are shown on the log-odds scale ( ) and as odds ratios (ORs) with 95% credible intervals. The reference category for institutional type is Public (Federal + State), while for MEC Score is the lowest CPC grading (1).\u003c/p\u003e\n \u003cdiv\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 2\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eProbability of achieving a satisfactory ENAMED score by institutional category\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eInstitutional Category\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eNumber of Courses\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eProbability of Satisfactory Score\u003c/p\u003e\n \u003cp\u003e(ENAMED \u0026gt; = 3)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eOverall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e346\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e73.4[69.4% − 77.5%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePublic (Federal + State)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e116\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e99.4[97.4% − 100.0%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCommunity/Confessional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e95.2[86.1% − 100.0%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Profit Private\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e76.6[66.7% − 85.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFor-Profit Private\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e112\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e43.5[33.0% − 53.6%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePublic (Municipal/Special)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e39.5[21.1% − 57.9%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 2 Note:\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eTable 2 summarizes the posterior mean probability (with 95% credible interval) that a medical course meets the INEP “Satisfactory” standard (≥60% proficient graduates) by institutional category and reports the number of courses in each category.\u003c/p\u003e\n \u003cp\u003eOf note, MEC/CPC score displayed limited monotonic alignment with ENAMED-based proficiency assurance, with only CPC = 2 (vs CPC = 1) being consistently associated with lower odds of meeting proficiency (OR 0.51; 0.29 to 0.91), whereas CPC = 3 to CPC = 5 estimates were imprecise and compatible with both modest decreases and increases in odds.\u003c/p\u003e\n \u003cp\u003eOther institutional characteristics showed weaker evidence of association with ≥ 60% proficiency. For each additional 1,000 hours of curricular workload in a given course, the adjusted association was close to null (OR 0.94; 95% CrI 0.85–1.04). Capital-city location was also not strongly associated (OR 1.11; 0.91–1.34). Finally, each + 100 increase in authorized enrollment slots was associated with a modest reduction in the odds of meeting the ≥ 60% proficiency benchmark (OR 0.89; 0.79–1.01), although the uncertainty still overlaps the null.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\"\u003e\n \u003ch2\u003eProbability of achieving a satisfactory ENAMED score by institutional category\u003c/h2\u003e\n \u003cp\u003eSystem assurance differed sharply across institutional categories (Table\u0026nbsp;2). Overall, the posterior mean probability that a course meets the “Satisfactory” threshold was 73.4% (69.4% to 77.5%). Public (Federal/State) courses exhibited near-universal assurance at 99.4% (97.4% to 100.0%), with Community/Confessional courses also high at 95.2% (86.1% to 100.0%).\u003c/p\u003e\n \u003cp\u003eIn contrast, Non-Profit Private courses demonstrated intermediate assurance at 76.6% (66.7% to 85.7%), while For-Profit Private and Public (Municipal/Special) courses showed substantially lower assurance at 43.5% (33.0% to 53.6%) and 39.5% (21.1% to 57.9%), respectively, indicating that in these segments the average probability of meeting the national benchmark was below 50%.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\"\u003e\n \u003ch2\u003eBayesian Assurance Curves according to Enrollment capacity\u003c/h2\u003e\n \u003cp\u003eAcross the medical courses, higher authorized enrollment was generally associated with lower system assurance of reaching the proficiency threshold (Table\u0026nbsp;3). Overall, the posterior mean change was − 2.6% per + 50 slots (95% CrI − 5.8 to + 0.2), with strong posterior probability for a negative trend (Pr[decrease] = 95.2%).\u003c/p\u003e\n \u003cdiv\u003e\n \u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 3\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eProbability of achieving a satisfactory ENAMED score by institutional category\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"4\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eInstitutional category\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSlot range evaluated\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eEstimated change per + 50 slots\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePosterior probability of decrease\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eOverall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40–210\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.6 pp (-5.6, + 0.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e95.2%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFor-Profit Private\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e50–250\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-5.0 pp (-10.9, + 0.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e95.0%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Profit Private\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e60–300\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.6 pp (-5.9, + 0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e93.2%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePublic (Municipal/Special)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e50–200\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-3.5 pp (-10.3, + 0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e79.1%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCommunity/Confessional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40–200\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.9 pp (-6.2, + 0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e76.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePublic (Federal + State)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e30–170\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.3 pp (-1.4, + 0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52.2%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 3 Note:\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eTable 3 summarizes the posterior mean change in assurance per +50 slots (percentage points) and the posterior probability that the trend is negative (Pr[decrease]\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e= Pr[slope \u0026lt; 0 | data],). Assurance is defined as the course-weighted probability that a randomly selected medical course meets the ENAMED satisfactory performance threshold (≥60% proficient). For each group, the posterior assurance curve π(\u003cem\u003ev\u003c/em\u003e) was computed over the same slot grid used in Figure 2, by varying enrollment slots\u0026nbsp;\u003cem\u003ev\u003c/em\u003e, while holding all other course-level covariates constant at their observed values. A linear slope was estimated within each posterior draw by regressing π(v) on slots v across all grid points.\u003c/p\u003e\n \u003cp\u003eThe decline was most likely among For-Profit Private courses: -5.0% per + 50 slots (95% CrI − 10.8 to + 0.4), with Pr[decrease] = 95.0%. Non-Profit Private courses also showed high likelihood of declining assurance with increasing enrollment (-2.9% per + 50 slots; -7.2 to + 0.1; Pr[decrease] = 93.2%). Notably, these groups also experienced the highest range (5th to 95th quartiles) among yearly enrollment capacity, ranging from 50 to 250 among for-profit, and 60 to 300 in non-profit private schools (Table\u0026nbsp;3). In contrast, Public (Federal + State) courses showed minimal change across the observed range (− 0.3 per + 50 slots; −1.2 to + 0.0; Pr[decrease] = 52.3%) (Fig.\u0026nbsp;2).\u003c/p\u003e\n \u003cp\u003eAlthough credible intervals for the per-50-slot change often included 0 (indicating uncertainty about exact magnitude), posterior probabilities of decrease were high for the overall system and the private sector strata, implying that scaling up enrollment has an important likelihood of being associated with reduced probability of meeting the proficiency benchmark.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\"\u003e\n \u003ch2\u003eConcordance Between MEC/CPC Scores and ENAMED Evaluation\u003c/h2\u003e\n \u003cp\u003eAgreement between ENAMED and CPC classifications was limited (Table\u0026nbsp;4). Overall, the probability of exact agreement was estimated to be 32.9% (95% CrI 28.1% to 37.7%), and the quadratic-weighted kappa was 0.31, consistent with only fair concordance beyond chance. Directional disagreement was asymmetric: CPC rated courses higher than ENAMED by at least one level in 38.2% of cases (33.2% to 43.1%), whereas ENAMED was rated higher than CPC in 28.9% (24.3% to 33.8%). Severe disagreement (≥ 2 levels) was also more common in the CPC-higher direction (13.0% [9.8% to 16.7%]) than in the ENAMED-higher direction (4.6% [2.7% to 7.1%]), and the mean category difference was negative (Δ = -0.19; -0.32 to -0.07), indicating net CPC inflation relative to ENAMED on average.\u003c/p\u003e\n \u003cdiv\u003e\n \u003ctable id=\"Tab4\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 4\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eConcordance and directional disagreement between MEC (CPC) and ENAMED categories\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"8\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003eInstitutional category\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAgreement\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eDirection of disagreement (≥ 1 level)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eSevere disagreement (≥ 2 levels)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eSummary\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eExact agreement\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCPC higher (≥ 1 level)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eENAMED Higher\u003c/p\u003e\n \u003cp\u003e(≥ 1 level)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCPC Higher\u003c/p\u003e\n \u003cp\u003e(≥ 2 levels)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eENAMED Higher\u003c/p\u003e\n \u003cp\u003e(≥ 2 levels)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMean Δ\u003c/p\u003e\n \u003cp\u003e(ENAMED − CPC)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eQuadratic-weighted κappa\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eOverall\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e32.9% [28.1%–37.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e38.2% [33.2%–43.1%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e28.9% [24.3%–33.8%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13.0% [9.8%–16.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.6% [2.7%–7.1%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.19 (-0.32, -0.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eFor-Profit Private\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e25.6% [18.3%–33.4%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e66.2% [57.8%–74.4%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8.1% [3.9%–13.5%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e29.6% [21.9%–38.0%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.4% [0.5%–5.8%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.92 (-1.12, -0.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePublic (Municipal/Special)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e16.4% [5.4%–32.0%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51.0% [32.6%–69.4%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e32.6% [16.7%–50.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18.1% [6.5%–34.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10.8% [2.4%–24.6%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.26 (-0.82, 0.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNon-Profit Private\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e32.5% [22.2%–43.4%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e49.1% [37.8%–60.3%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18.4% [10.8%–27.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12.0% [5.7%–20.1%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.0% [0.8%–9.4%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.40 (-0.66, -0.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eCommunity/Confessional\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e44.2% [30.9%–58.0%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e20.7% [10.6%–32.8%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e35.1% [22.6%–48.9%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6.3% [1.3%–14.5%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10.3% [3.6%–20.4%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.20 (-0.16, 0.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePublic (Federal + State)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e34.2% [26.2%–43.0%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8.2% [4.0%–13.8%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.5% [48.3%–66.1%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.3% [0.9%–7.1%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11.5% [6.4%–17.7%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.59 (0.41, 0.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 4 Legend:\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eTable 4 summarizes agreement between ENAMED and CPC ordinal categories (1-5), including exact agreement, directional disagreement (CPC higher vs ENAMED higher by \u0026gt;=1 level), severe disagreement (\u0026gt;=2 levels), mean difference Delta (ENAMED - CPC), and quadratic-weighted kappa (QWK). Proportions and mean Delta are reported as posterior means with 95% credible intervals from a Dirichlet posterior on the 5x5 joint table.\u003c/p\u003e\n \u003cp\u003eThis misalignment was most pronounced among For-Profit Private courses. In this segment, there was a 66.2% [57.8% to 74.4%] probability that CPC exceeded ENAMED by at least one level and 29.6% [21.9% to 38.0%] by at least two levels, with a large negative mean Delta of -0.92 (-1.12 to -0.71) and kappa of 0.22. By contrast, among Public (Federal/State) courses, ENAMED more frequently exceeded CPC (57.5% [48.3% to 66.1%]) with a positive mean Delta of + 0.59 (0.41 to 0.77). Together, these patterns suggest that CPC is not a reliable proxy for ENAMED-based competency assurance and may systematically overrate performance in the for-profit sector.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\"\u003e\n \u003ch2\u003eSubgroup Analysis – Bayesian Model Specific to For-Profit Schools\u003c/h2\u003e\n \u003cp\u003eWithin the For-Profit Private subgroup, two predictors showed clear evidence of association with lower proficiency assurance (\u003cstrong\u003eSupplementary Table S3\u003c/strong\u003e). Curricular workload (per 1,000 hours) was negatively associated with proficiency odds (OR 0.79; 95% CrI 0.66 to 0.95), and the post More-Doctors Program period (\u0026gt; 2013) was strongly associated with reduced proficiency levels (OR 0.53; 0.42 to 0.68). Increases in enrollment slots showcases a modest, yet non-significant, negative association with proficiency level. Other predictors were not strongly associated within the subgroup: capital-city location (OR 0.93; 0.71 to 1.23), and private conglomerate affiliation (OR 0.91; 0.73 to 1.13) were compatible with both modest decreases and increases in odds.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis nationwide analysis of ENAMED-2025 revealed marked heterogeneity in course-level attainment of the MEC \u0026ldquo;Satisfactory\u0026rdquo; standard across Brazil\u0026rsquo;s medical education system. After adjustment for enrollment capacity, curricular workload, CPC rating, geographic location, and policy period, the institutional administrative category remained the most consistent predictor of meeting the minimum proficiency threshold. Federal and state public courses showed near-universal assurance of meeting the benchmark, Community/Confessional institutions also performed strongly, whereas non-profit private, municipal/special public, and especially for-profit private courses exhibited markedly lower assurance.\u003c/p\u003e \u003cp\u003eThese patterns are coherent with Brazil\u0026rsquo;s broader institutional landscape in higher education, where governance models, funding arrangements, academic density, and the ability to secure stable clinical training environments vary considerably across segments \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Because medical education depends on supervised practice, service integration, and infrastructure-intensive learning settings, differences in institutional capacity to sustain faculty careers, integrate with health services, and maintain supervised practice settings plausibly translate into differences in the likelihood of meeting minimum proficiency standards.\u003c/p\u003e \u003cp\u003eThe superior ENAMED proficiency outcomes observed among public federal and state universities likely reflect structural advantages historically concentrated in this segment. Public institutions are typically tuition-free and constitutionally oriented toward integrating teaching, research, and extension, and they more often concentrate stricto sensu postgraduate programs, full-time doctoral faculty, and university hospitals that support clinical training at scale \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Municipal public institutions, by contrast, often operate on a smaller scale and depend on local fiscal capacity, which may limit long-term investments in faculty consolidation and training environments \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWithin the private sector, governance and incentives also differ in ways that plausibly shape training conditions. Non-profit institutions, particularly Community and Confessional models, tend to reinvest surpluses in academic activities and may sustain stronger regional embeddedness \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. By contrast, for-profit institutions, now the dominant segment, operate under market-oriented expansion strategies and charge some of the highest tuition fees in medical education \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Although central to the rapid expansion of training positions, this model has raised concerns regarding alignment between accelerated growth and the infrastructure-intensive requirements of medical education \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan additionalcitationids=\"CR24\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. In this context, ENAMED\u0026rsquo;s threshold-based accountability provides a relevant lens for assessing whether these heterogeneous institutional arrangements can consistently ensure a minimum proficiency standard at graduation.\u003c/p\u003e \u003cp\u003eImportantly, the observed gradient favoring public institutions emerged despite a prolonged period of fiscal austerity and structural underfunding affecting Brazilian public universities \u003csup\u003e\u003cspan additionalcitationids=\"CR27 CR28\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. This reinforces prior evidence linking institutional governance models to medical training quality. The findings also align with earlier evidence that predominantly private expansion, especially among for-profit programs, has been associated with weaker performance in prior national assessments and with massification processes not adequately supported by proportional growth in infrastructure and training fields \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Because ENAMED is aligned with the National Curricular Guidelines and with competencies expected for practice in the Brazilian Unified Health System, and because it influences access to residency training, it may render these structural asymmetries more visible than earlier evaluation models \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur results should also be interpreted against the backdrop of Brazil\u0026rsquo;s large-scale expansion in medical education, driven by policy initiatives, including the More Doctors Program, intended to increase physician supply and reduce regional disparities \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan additionalcitationids=\"CR33 CR34 CR35 CR36 CR37 CR38 CR39 CR40\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Although expansion goals were framed around workforce equity, the long-term educational consequences have remained contested, particularly given the rapid proliferation of private programs and concerns about regulatory fragility \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Consistent with these debates, we found that courses established in the post\u0026ndash;More Doctors Program expansion period showed a lower likelihood of meeting the ENAMED minimum proficiency standard, a pattern that was especially pronounced in the for-profit private sector. This suggests that in parts of the system, the pace of expansion may have outstripped regulatory capacity and the practical training infrastructure required for competency-based graduation.\u003c/p\u003e \u003cp\u003eThe association between higher authorized enrollment and lower assurance, most evident in private-sector strata, points in the same direction. While the magnitude varied, the posterior evidence supports the premise that scaling enrollment is not neutral: when expansion is not accompanied by proportional growth in clinical placement capacity, supervision, and training infrastructure, the probability of meeting a minimum proficiency benchmark declines. This is consistent with prior work linking accelerated expansion and larger class sizes to poorer performance in national assessments (4), and with evidence describing constraints in practice-field allocation, school\u0026ndash;service coordination, and insufficient ratios of hospital beds per student, all of which can undermine the supervised learning environment essential for internship quality \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan additionalcitationids=\"CR43\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eA further policy-relevant finding is the limited alignment between ENAMED-based threshold attainment and CPC categories. CPC is a multidimensional composite that incorporates ENADE performance, faculty characteristics, and student-reported measures of infrastructure and pedagogical resources \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. However, the modest concordance observed here suggests that CPC may not reliably detect whether a program ensures that a sufficient proportion of graduates reach a minimum proficiency floor. If regulatory oversight relies primarily on CPC, it may fail in two directions: it may overlook programs that appear structurally adequate yet do not achieve minimum proficiency assurance, and it may penalize programs that meet the proficiency floor while scoring modestly on other components. The presence of high-CPC programs with low ENAMED proficiency attainment raises questions about whether the previous framework adequately captures professional competence as developed across training, especially in a system where incentives may not map cleanly onto competency assurance.\u003c/p\u003e \u003cp\u003eThese results also intersect with an evolving regulatory response. Following the first ENAMED cycle, the Ministry of Education adopted graduated supervisory measures targeting low-performing programs, including suspension of new admissions and reductions in authorized enrollment capacity \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Subsequently, the federal government revoked the active regulatory instrument governing the authorization of new medical programs, after conducting a broader technical reassessment of expansion policies, considering the effects of judicially driven growth, infrastructure capacity constraints, and the emerging ENAMED result \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. In parallel, the debate has extended to legislative proposals concerning a National Medical Proficiency Examination as a condition for licensure and to discussions within the Federal Council of Medicine regarding the potential use of ENAMED performance in professional registration decisions \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. The initial regulatory call had defined procedural and evaluative criteria for authorizing new private-sector medical programs \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFrom an assessment perspective, ENAMED holds potential to become a central mechanism for monitoring medical training in Brazil, particularly if employed as an instrument for ensuring minimum competency standards rather than merely ranking institutions. Although it is primarily a cognitive-theoretical examination, prior studies support the validity of well-designed assessments of this nature in measuring relevant medical competencies. Furthermore, they link low proficiency performance to deficits that extend beyond cognition into clinically meaningful domains \u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e, and demonstrate associations between theoretical examination performance, critical thinking measures \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e, and practical assessments such as the Objective Structured Clinical Examination (OSCE) \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe implications are practical, as our results support the strengthening of regulatory oversight of expansion, including closer scrutiny of rapid scale increases in segments where assurance declines with enrollment. They also indicate that ENAMED can function as a continuous monitoring instrument for minimum competency assurance, while highlighting the need to reassess how existing indicators, such as CPC, are used in supervision and authorization decisions. Collectively, the findings suggest that expansion without assurance of minimum standards creates systemic risks for health services and for the credibility of training pathways.\u003c/p\u003e \u003cp\u003eThe results reported here also invite reflection on pathways toward excellence within the current landscape. With regulatory sanctions linked to ENAMED performance, there is a risk that institutions may redirect pedagogical strategies narrowly toward examination outcomes. However, meaningful improvement requires structured academic reconstruction strategies centered on substantive training quality. This includes strengthening internship governance, aligning internal assessment systems with competencies, supporting faculty development, and cultivating institutional cultures grounded in formative mission. In the medium and long term, building academic density, institutional identity, and internal self-regulatory mechanisms remains essential for sustainable improvement, ultimately benefiting institutions, students, patients, and the Brazilian Unified Health System.\u003c/p\u003e \u003cp\u003eThis study has limitations inherent to its cross-sectional and ecological design, as analyses were conducted at the institutional rather than individual level of the medical graduates. Thus, identified associations should not be interpreted as direct causal relationships and cannot be extrapolated to the student level. Unobserved variables, including students\u0026rsquo; socioeconomic background, faculty-to-student ratios, quality of clinical training environments, and degree of integration between teaching and healthcare services, may contribute to the heterogeneity observed. Additionally, ENAMED remains in its early implementation phase, and future methodological adjustments may affect longitudinal comparability. Despite these limitations, the study presents important strengths. It employs national microdata, applies hierarchical Bayesian modeling appropriate for proportion-based outcomes with varying denominators, explicitly incorporates statistical uncertainty, and translates complex findings into metrics directly relevant for regulatory decision-making. By focusing on the probability of achieving a minimum competency threshold, this work shifts the debate from simple comparisons of mean performance toward a framework centered on formative safety and the social accountability of medical education.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis nationwide analysis of the Brazilian ENAMED-2025 indicates that course-level attainment of the MEC \u0026ldquo;Satisfactory\u0026rdquo; standard is strongly patterned by institutional characteristics, particularly administrative category, timing of program implementation, and authorized enrollment scale, revealing a consistent gradient in which federal and state public schools show the highest competency assurance, while private programs, especially for-profit schools established during the recent expansion period, display greater risk of formative insufficiency. By reframing evaluation from mean-score comparisons to the assurance of a minimum competency floor aligned with the National Curricular Guidelines and SUS needs, ENAMED can become a central instrument to guide supervision and expansion policy. However, its value will depend on prioritizing substantive structural improvements in training, such as internship governance, internal competency-based assessment, faculty development, and institutional educational culture, rather than narrow test-oriented responses, reinforcing that minimum-quality assurance in medical education is an ethical and system-level imperative to protect learners, patients, and the SUS.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eBrazilian Unified Health System (SUS); National Medical Education Assessment Examination (ENAMED); National Institute for Educational Studies and Research \u0026quot;An\u0026iacute;sio Teixeira (INEP), Ministry of Education (MED); Preliminary Course Concept (CPC), ntegrated Nested Laplace Approximation (INLA); Credible intervals (CrI); Quadratic-weighted kappa (QWK); Second-order random-walk (RW2); Objective Structured Clinical Examination (OSCE)\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBBA contributed to conceptualization, supervision, critical intellectual revision of the manuscript, and project administration. KVS contributed to data curation, formal analysis, and writing \u0026ndash; review and editing. RCM contributed to formal analysis and writing \u0026ndash; review and editing. QHB contributed to data acquisition and writing \u0026ndash; review and editing. LFQ contributed to conceptualization, writing \u0026ndash; original draft preparation, and critical intellectual revision. KMA contributed to conceptualization, writing \u0026ndash; original draft preparation, critical intellectual revision, and supervision. All authors contributed to the article and approved the submitted version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data used in this study were obtained from publicly available government databases and did not involve identifiable personal information. Therefore, it did not require submission to the Brazilian Research Ethics Committee, as dictated by Resolution No. 510/2016 of the Brazilian Health Council on norms applicable to Human and Social Sciences research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was supported by the Intramural research programs of the Oswaldo Cruz Foundation, Brazil and of the Clariens Educa\u0026ccedil;\u0026atilde;o, Brazil.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData/Materials Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets supporting the conclusions of this article are publicly available in (https://www.gov.br/inep/pt-br/acesso-a-informacao/dados-abertos/microdados/enamed).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAvena KM, Quintanilha LF, Luzardo Filho RL, Andrade BB. Lessons learned from the expansion of medical schools in Brazil: a review of challenges and opportunities. Front Educ (Lausanne). 2024;9:1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrade BB. The Dark Side of Private Medical Education in Brazil. Front Med (Lausanne). 2025;12:1504794.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheffer M. Medical Demography in Brazil 2025 [Internet]., Brasilia DF. Ministry of Health of Brazil; 2025 [cited 2025 May 26]. 448 p. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://bvsms.saude.gov.br/bvs/publicacoes/demografia_medica_brasil_2025.pdf\u003c/span\u003e\u003cspan address=\"https://bvsms.saude.gov.br/bvs/publicacoes/demografia_medica_brasil_2025.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrade BB, Villalva-Serra K, Menezes RC, Quintanilha LF, Avena KM. For-Profit Growth and Academic Decline: A Retrospective Nationwide Assessment of Brazilian Medical Schools. Front Med (Lausanne). 2025;12:1617885.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheffer M, Mosquera P, Cassenote A, McPake B, Russo G. Brazil\u0026rsquo;s experiment to expand its medical workforce through private and public schools: Impacts and consequences of the balance of regulatory and market forces in resource-scarce settings. Global Health. 2025;21(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrazil. Law No. 12,871 of October 22, 2013: Institutes the Mais M\u0026eacute;dicos Program, amends laws no. 8,745, of december 9, 1993, and no. 6,932, of july 7, 1981, and makes other provisions. [Internet]. Bras\u0026iacute;lia, DF: Presidency of the Republic of Brazil; 2013 [cited 2025 Sep 16]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.planalto.gov.br/ccivil_03/_ato2011-2014/2013/lei/l12871.htm\u003c/span\u003e\u003cspan address=\"https://www.planalto.gov.br/ccivil_03/_ato2011-2014/2013/lei/l12871.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMachado CV, Silva GA. Political struggles for a universal health system in Brazil: Successes and limits in the reduction of inequalities. Global Health. 2019;15(Suppl 1):1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMattos E, Mazetto D. Assessing the impact of more doctors\u0026rsquo; program on healthcare indicators in Brazil. World Dev. 2019;123:104617.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampos VMS, Maggitti-Bezerril M, Menezes RC, Sobral L, Avena KM, Andrade BB. Doctors in the making, or degrees for sale? A student\u0026rsquo;s view of Brazil\u0026rsquo;s medical education crisis. Front Med (Lausanne). 2025;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFederal Council of Medicine. 78% of municipalities that host medical schools do not have adequate infrastructure [Internet]. 2024 [cited 2025 Jan 11]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.cfm.org.br/noticias/78-dos-municipios-que-sediam-escolas-medicas-nao-possuem-a-infraestrutura-adequada-para-a-formacao-dos-profissionais\u003c/span\u003e\u003cspan address=\"https://portal.cfm.org.br/noticias/78-dos-municipios-que-sediam-escolas-medicas-nao-possuem-a-infraestrutura-adequada-para-a-formacao-dos-profissionais\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFederal Council of Medicine. 73% of municipalities applying to receive medical schools do not have adequate infrastructure [Internet]. 2024 [cited 2025 Jan 11]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.cfm.org.br/noticias/73-dos-municipios-candidatos-a-receber-escolas-medicas-nao-tem-infraestrutura-adequada-para-formar-novos-profissionais\u003c/span\u003e\u003cspan address=\"https://portal.cfm.org.br/noticias/73-dos-municipios-candidatos-a-receber-escolas-medicas-nao-tem-infraestrutura-adequada-para-formar-novos-profissionais\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFatel QH, Silva FG, Avena KM, Menezes RC, Andrade BB, Quintanilha LF. Shaping Tomorrow\u0026rsquo;s Doctors: The Impact of Socioeconomic and Institutional Factors on Medical Education Quality in Brazil: An Ecological Study. J Med Educ Curric Dev. 2025;12:1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrazil, ENAMED - Brazilian Medical Training Assessment Exam [Internet]. Instituto Nacional de Estudos e Pesquisas Educacionais An\u0026iacute;sio Teixeira (Inep). 2025 [cited 2025 May 26]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gov.br/inep/pt-br/areas-de-atuacao/avaliacao-e-exames-educacionais/enamed\u003c/span\u003e\u003cspan address=\"https://www.gov.br/inep/pt-br/areas-de-atuacao/avaliacao-e-exames-educacionais/enamed\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNational Institute of Educational Studies and Research An\u0026iacute;sio Teixeira (INEP). INEP: National Institute of Educational Studies and Research An\u0026iacute;sio Teixeira [Internet], Bras\u0026iacute;lia DF. 2024 [cited 2024 Apr 3]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.inep.gov.br/\u003c/span\u003e\u003cspan address=\"http://www.inep.gov.br/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGomes MF. Evaluation and administrative nature of the institutions of higher education. Ensaio: Avalia\u0026ccedil;\u0026atilde;o e Pol\u0026iacute;ticas P\u0026uacute;blicas em Educa\u0026ccedil;\u0026atilde;o. 2010;18(68):589\u0026ndash;610.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrazil, National Institute of Educational Studies and Research An\u0026iacute;sio Teixeira (INEP). Higher Education Census 2021: statistical notes. Bras\u0026iacute;lia, DF: INEP; 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZorzetto R, Razzouk D, Dubugras MTB, Gerolin J, Schor N, Guimar\u0026atilde;es JA, et al. The scientific production in health and biological sciences of the top 20 Brazilian universities. Braz J Med Biol Res. 2006;39(12):1513\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeves RS. Municipal institutions of higher education. Interesse P\u0026uacute;blico. 2018;20(109):191\u0026ndash;210.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFioreze C, McCowan T. Community universities in the South of Brazil: prospects and challenges of a model of non-state public higher education. Comp Educ. 2018;54(3):370\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTagliari C. For-profit expansion and nonprofit retraction in Brazilian higher education. Plural. 2022;29(1):36\u0026ndash;59.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCowan T. The growth of private higher education in Brazil: implications for issues of equity, quality and public benefit. Educ Policy Anal Arch [Internet]. 2004;12:30. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://epaa.asu.edu/epaa/v13n27\u003c/span\u003e\u003cspan address=\"http://epaa.asu.edu/epaa/v13n27\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarvalho CHA, Sousa Filho EM, Goellner IA, Freire NR, Amitrano CR. Private higher education financialization in Brazil: expansion of medical programs in publicly traded companies (2013\u0026ndash;2022). Educ Soc. 2024;45:e286692.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeber CAT, da Silva AG. New medical schools: solution or problem? Debates em Psiquiatria. 2024;14:1\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheffer MC, Poz MRD. The privatization of medical education in Brazil: trends and challenges. Hum Resour Health. 2015;13(1):96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNassar LM, Couto MHC, Pereira Junior GA. Public funding (FIES and PROUNI) for medical teaching in Brazil: a literature review and the distortions created. Educa\u0026ccedil;\u0026atilde;o em Revista. 2021;37:e25246.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilva S, Leher R. Austerity is jeopardizing the future of public universities and their social function. Temporalis. 2024;24(47):15\u0026ndash;34.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmaral NC, Junior WT da, Salles S. Federal universities: autonomy undermined by funding? Caderno CRH. 2025;38:e025020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilva ARR, Oliveira LRPF, Pinho MJ. The higher public education in political context: reflections concerning the university dismantling process from de 2016. Revista Cocar. 2024;21:39.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNota-Publica-ABC. -SBPC-maio-2025. 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrade BB, Sasaki KCM, Menezes RC, Luzardo Filho RL, Avena KM. From framework to fitness for the 21st century: How Brazil\u0026rsquo;s 2025 National Curricular Guidelines recast priorities for training physicians. Front Med (Lausanne). 2026;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodrigues DLG. Challenges and opportunities in the Brazilian Unified Health System: Pathways to sustainability and equity. Public Health. 2025;247.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLopes AC. The numerical explosion in Brazilian medical schools. Educaci\u0026oacute;n M\u0026eacute;dica. 2018;19:19\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliveira BLCA, Lima SF, Pereira MUL, Pereira J\u0026uacute;nior GA. Evolution, distribution and expansion of medicine courses in Brazil (1808\u0026ndash;2018). Trabalho, Educa\u0026ccedil;\u0026atilde;o e Sa\u0026uacute;de. 2019;17(1):e0018317.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePereira DVR, Fernandes DLR, Mari JF, Lage ALF, Fernandes APPC. Mapping of medical schools: the distribution of undergraduate courses and annual vacancies in Brazilian cities in 2020. Rev Bras Educ Med. 2021;45(1):1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantos LMP, Oliveira A, Trindade JS, Barreto ICHC, Palmeira PA, Comes Y, et al. Implementation research: towards universal health coverage with more doctors in Brazil. Bull World Health Organ. 2017;95(2):103.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDolci JEL. The proliferation of medical schools in Brazil: a threat to the quality of medical education? Braz J Otorhinolaryngol. 2023;89(6):101354.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliveira JPA, Sanchez MN, Santos LMP. The Mais M\u0026eacute;dicos (More Doctors) Program: the placement of physicians in priority municipalities in Brazil from 2013 to 2014. Cien Saude Colet. 2016;21(9):2719\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFigueiredo AM, McKinley DW, Massuda A, Azevedo GD. Evaluating medical education regulation changes in Brazil: workforce impact. Hum Resour Health. 2021;19(1):1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFigueiredo AM, McKinley DW, Lima KC, Azevedo GD. Medical school expansion policies: educational access and physician distribution. Med Educ. 2019;53(11):1121\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantos LMP, Costa AM, Girardi SN. Mais Medicos Program: an effective action to reduce health inequities in Brazil. Cien Saude Colet [Internet]. 2015 Jan 1 [cited 2026 Feb 8];20(11):3547\u0026ndash;52. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.scielo.br/j/csc/a/mFYpCXL3q4XknsR58Pk5gmS/?lang=en\u003c/span\u003e\u003cspan address=\"https://www.scielo.br/j/csc/a/mFYpCXL3q4XknsR58Pk5gmS/?lang=en\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGirardi SN, Van Stralen AC, de Cella S, Der Maas JN, Carvalho LW, de Faria CL. Impact of the Mais M\u0026eacute;dicos (More Doctors) Program in reducing physician shortage in Brazilian Primary Healthcare. Cien Saude Colet. 2016;21(9):2675\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliveira BLCA, Lima SF, Pereira MUL, Pereira J\u0026uacute;nior GA. Evolution, distribution and expansion of medicine courses in Brazil (1808\u0026ndash;2018). Trabalho, Educa\u0026ccedil;\u0026atilde;o e Sa\u0026uacute;de. 2019;17(1):e0018317.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilva FGM, Santos PAJ, Quintanilha LF. Mais M\u0026eacute;dicos - More Physicians law: evaluating compliance with municipal criteria for the operation of medical schools. Rev Bras Educ Med. 2025;49(2):e067.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorais HMM, Sa RGR, Albuquerque M, do SV, Oliveira RS. Expansion and privatization of medical courses and the integration of teaching and service: the case of the state of Pernambuco. Sa\u0026uacute;de em Debate. 2023;47(137 abr\u0026ndash;jun):182\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIndicators of higher education quality. 91.6% of courses have a Course Performance Concept (CPC) in 2019 that falls between bands 3 and 5 [Internet]. [cited 2026 Feb 8]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://abmes.org.br/noticias/detalhe/4136/indicadores-de-qualidade-da-educacao-superior-cpc-2019-de-91-6-dos-cursos-esta-entre-as-faixas-3-e-5\u003c/span\u003e\u003cspan address=\"https://abmes.org.br/noticias/detalhe/4136/indicadores-de-qualidade-da-educacao-superior-cpc-2019-de-91-6-dos-cursos-esta-entre-as-faixas-3-e-5\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePreliminary Course Concept (CPC) [Internet]. [cited 2026 Feb 8]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gov.br/inep/pt-br/areas-de-atuacao/pesquisas-estatisticas-e-indicadores/indicadores-de-qualidade-da-educacao-superior/conceito-preliminar-de-curso-cpc\u003c/span\u003e\u003cspan address=\"https://www.gov.br/inep/pt-br/areas-de-atuacao/pesquisas-estatisticas-e-indicadores/indicadores-de-qualidade-da-educacao-superior/conceito-preliminar-de-curso-cpc\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrazil E. Medical course evaluations and supervision measures released [Internet]. Instituto Nacional de Estudos e Pesquisas Educacionais An\u0026iacute;sio Teixeira (INEP). 2026 [cited 2026 Feb 8]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gov.br/mec/pt-br/assuntos/noticias/2026/janeiro/enamed-divulgadas-avaliacao-dos-cursos-de-medicina-e-medidas-de-supervisao\u003c/span\u003e\u003cspan address=\"https://www.gov.br/mec/pt-br/assuntos/noticias/2026/janeiro/enamed-divulgadas-avaliacao-dos-cursos-de-medicina-e-medidas-de-supervisao\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrazil. Revocation of Public Notice No. 01, dated October 4, 2023, and its amendments. Minist\u0026eacute;rio da Educa\u0026ccedil;\u0026atilde;o. Conselho Nacional de Educa\u0026ccedil;\u0026atilde;o. Brasilia, DF; 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAg\u0026ecirc;ncia, Senado. A bill creating a medical proficiency exam is in its final stages in the Senate [Internet]. 2026 [cited 2026 Feb 8]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www12.senado.leg.br/noticias/materias/2026/01/26/projeto-que-cria-exame-de-proficiencia-em-medicina-esta-em-fase-final-no-senado\u003c/span\u003e\u003cspan address=\"https://www12.senado.leg.br/noticias/materias/2026/01/26/projeto-que-cria-exame-de-proficiencia-em-medicina-esta-em-fase-final-no-senado\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrazil MEC. Notice No. 1/2023. Public Call for Proposals for Authorization to Operate Medical Courses within the More Doctors Program. Brasilia, DF: Minist\u0026eacute;rio da Educa\u0026ccedil;\u0026atilde;o, Conselho Nacional de Educa\u0026ccedil;\u0026atilde;o; 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndreou V, Eggermont J, Gielis G, Schoenmakers B. Proficiency testing for identifying underperforming students before postgraduate education: a longitudinal study. BMC Med Educ. 2020;20(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMafinejad MK, Arabshahi SKS, Monajemi A, Jalili M, Soltani A, Rasouli J. Use of Multi-Response Format Test in the Assessment of Medical Students\u0026rsquo; Critical Thinking Ability. J Clin Diagn Res. 2017;11(9):LC10\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastellani L, Quintanilha LF, Arriaga MB, Lima ML, Andrade BB. Objective Structured Clinical Examination (OSCE) As a Reliable Evaluation Strategy: Evidence From a Brazilian Medical School. Probl Educ 21st Century. 2020;78(5):674\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"medical education, academic performance, education business, undergraduate training, Brazil","lastPublishedDoi":"10.21203/rs.3.rs-8904720/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8904720/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eBrazil\u0026rsquo;s medical education system has expanded at unprecedented pacing, largely propelled by policies meant to increase the physician workforce and reduce regional inequities. This expansion succeeded in scale, but it reignited an old question: are we expanding training capacity or expanding diplomas? In 2025, Brazil introduced the National Medical Education Assessment Examination (ENAMED), a regulatory instrument grounded in a simple principle: programs should be judged, and effectively ranked, by the proportion of graduating students who meet a defined minimum standard of professional proficiency (knowledge-based), rather than by mean scores alone. Here, we evaluated institutional and structural determinants of ENAMED performance, focusing on the system\u0026rsquo;s ability to guarantee minimum proficiency across medical schools amid recent expansion patterns.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe conducted a nationwide retrospective observational analysis linking ENAMED-2025 results to institutional and regulatory characteristics of medical programs. Our outcome was systemic assurance of minimum proficiency, defined as meeting a regulatory benchmark in which\u0026thinsp;\u0026ge;\u0026thinsp;60% of graduates are classified as proficient. We fit hierarchical Bayesian models estimating each program\u0026rsquo;s posterior probability of meeting this benchmark, accounting for course size and geographic clustering (state and municipality).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eENAMED revealed pronounced heterogeneity in the probability of meeting the minimum proficiency benchmark across institutional segments. Federal and state public medical schools had the highest assurance, followed by community-based institutions. Private programs, especially for-profit schools, showed substantially lower probabilities of meeting the benchmark. Programs created during the most recent waves of expansion and those with larger volumes of authorized enrollments consistently exhibited weaker assurance. Independently, greater enrollment capacity was associated with lower probability of achieving the \u0026ge;\u0026thinsp;60% proficiency threshold, with the steepest gradient observed among for-profit institutions.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eENAMED\u0026rsquo;s signal is strongly shaped by institutional design and conditions under which expansion occurred, far more than by chance variation. By privileging the share of graduates who reach a minimum proficiency standard, ENAMED shifts regulation from \u0026ldquo;average performance\u0026rdquo; to \u0026ldquo;minimum guarantees,\u0026rdquo; offering a more defensible lens for accountability in medical education. Its promise, however, depends on policy responses that strengthen academic governance, clinical training capacity, and faculty support, rather than short-term, reactive strategies aimed at gaming exam performance.\u003c/p\u003e","manuscriptTitle":"Institutional predictors of minimum proficiency in Brazil’s new National Medical Education Assessment (ENAMED): a Bayesian hierarchical nationwide cross-sectional analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-10 10:16:52","doi":"10.21203/rs.3.rs-8904720/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1ee074d4-6e53-4381-9381-ae0adddd46e9","owner":[],"postedDate":"March 10th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-10T10:16:52+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-10 10:16:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8904720","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8904720","identity":"rs-8904720","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}