Imbalanced specialty representation of USMLE and NBME test writers.

PURPOSE: The United States Medical Licensing Examination (USMLE) is an examination series required for allopathic physician licensure in the United States (US). USMLE content is created and maintained by the National Board of Medical Examinations (NBME). The specialty composition of the USMLE and NBME taskforce members involved in the creation of examination content is currently unknown. METHODS: Using the 2021 USMLE and 2021 NBME Committees and Task Forces documents, we determined each member's board-certified primary specialty and involvement in test material development committees who we dubbed "test writers". Total active physicians by primary specialty were recorded from the 2020 Physician Specialty Data Report published by the Association of American Medical Colleges (AAMC). Descriptive statistics and chi-square analysis were used to analyze the cohorts. RESULTS: The USMLE and NBME test writer primary specialty composition was found to be significantly different compared to the US active physician population (USMLE χ2[32] = 172, p < .001 and NBME χ2[32] = 200, p < .001). Only nineteen specialties were represented within USMLE test writers, with three specialties being proportionally represented. Two specialties were represented within NBME test writers. Obstetrics and Gynecology physicians were proportionally represented in USMLE but not within NBME test writers. Internal Medicine (IM) accounts for the largest percentage of all USMLE test writers (60/197, 30%) with an excess representation of 31 individuals. CONCLUSIONS: There is an imbalance in the specialty representation of USMLE and NBME test writers compared to the US active physician population. These findings may have implications for the unbiased and accurate portrayal of topics in such national examinations; thus, future investigation is warranted.

Experts' prediction of item difficulty of multiple-choice questions in the Ethiopian Undergraduate Medicine Licensure Examination.

BACKGROUND: The ability of an expert's item difficulty ratings to predict test-taker actual performance is an important aspect of licensure examinations. Expert judgment is used as a primary source of information for users to make prior decisions to determine the pass rate of test takers. The nature of raters involved in predicting item difficulty is central to set credible standards. Therefore, this study aimed to assess and compare raters' prediction and actual Multiple-Choice Questions' difficulty of the undergraduate medicine licensure examination (UGMLE) in Ethiopia. METHOD: 815 examinees' responses to 200 Multiple-Choice Questions (MCQs) were used in this study. The study also included experts' item difficulty ratings of seven physicians who participated in the standard settings of UGMLE. Then, analysis was conducted to understand experts' rating variation in predicting the actual difficulty levels of examinees. Descriptive statistics was used to profile the mean rater's and actual difficulty value for MCQs, and ANOVA was used to compare the mean differences between raters' prediction of item difficulty. Additionally, regression analysis was used to understand the interrater variations in item difficulty predictions compared to the actual difficulty. The proportion of variance of actual difficulty explained from rater prediction was computed using regression analysis. RESULTS: In this study, the mean difference between raters' prediction and examinees' actual performance was inconsistent across the exam domains. The study revealed a statistically significant strong positive correlation between the actual and predicted item difficulty in exam domains eight and eleven. However, a non-statistically significant very weak positive correlation was reported in exam domains seven and twelve. The multiple comparison analysis showed significant differences in mean item difficulty ratings between raters. In the regression analysis, experts' item difficulty ratings of the UGMLE had 33% power in predicting the actual difficulty level. The regression model also showed a moderate positive correlation (R = 0.57) that was statistically significant at F (6, 193) = 15.58, P = 0.001. CONCLUSION: This study demonstrated the complex process for assessing the difficulty level of MCQs in the UGMLE and emphasized the benefits of using experts' ratings in advance. To ensure the exams maintain the necessary reliable and valid scores, raters' accuracy on the UGMLE must be improved. To achieve this, techniques that align with the evolving assessment methodologies must be developed.

How well does the academic performance of Ethiopian anaesthetists predict their National Licensing Examination scores: a cross-sectional study.

OBJECTIVES: This study assessed the relationship between academic performance and National Licensing Examination (NLE) scores of Ethiopian anaesthetists and proposed academic performance thresholds that predict failing the NLE. DESIGN: A retrospective cross-sectional study. SETTING: Academic performance measures were collected from eight universities and matched to total and subdomain NLE scores for anaesthetists. Four universities were first generation (oldest), three were second generation and one was third generation (newest). PARTICIPANTS: 524 anaesthetists. MEASURES: Selected course grades, yearly (Y1-Y4) and cumulative grade point averages (GPAs) and in-school exit exam status were correlated with NLE scores. Linear regression was used to identify predictors of NLE score. Receiver operating characteristic curves were plotted to estimate GPA thresholds that predict failing the NLE. RESULTS: Total and general surgery anaesthesia subdomain NLE scores showed moderate to strong linear relationships with most academic performance indicators. However, correlations between the rest of the subdomain NLE scores and composite course scores were either small or negligible (r<0.5). For every 1-point increase in a 4-point cumulative GPA and composite senior year practicum course score, the first-attempt NLE score increased by 12.61 (95% CI 10.03 to 15.19; p<0.0001) and 3.98 (95% CI 1.77 to 6.19; p=0.001) percentage points, respectively. Those entering university directly from high school and those attending a third-generation university would, on average, score 2.79 (95% CI 0.23 to 5.35; p=0.033) and 4.34 (95% CI 1.42 to 7.26; p=0.004) higher NLE scores compared with nurse entrants and first-generation university graduates, respectively. On a 4-point grading system, the thresholds for failing the NLE were below 2.68 for year-1 GPA, 2.62 for year-2 GPA, 2.72 for year-3 GPA, 3.04 for year-4 GPA and 2.87 for cumulative GPA. CONCLUSIONS: The Ethiopian anaesthetist's NLE score modestly correlated with most academic performance indicators. Based on pass/fail thresholds, the NLE could assist training programmes in improving NLE pass rates.

When can we identify the students at risk of failure in the national medical licensure examination in Japan using the predictive pass rate?

CONTEXT: Failure of students to pass the National Medical Licensure Examination (NMLE) is a major problem for universities and the health system in Japan. To assist students at risk for NMLE failure as early as possible after admission, this study investigated the time points (from the time of admission to graduation) at which predictive pass rate (PPR) can be used to identify students at risk of failing the NMLE. METHODS: Seven consecutive cohorts of medical students between 2012 and 2018 (n = 637) at the Gifu University Graduate School of Medicine were investigated. Using 7 variables before admission to medical school and 10 variables after admission, a prediction model to obtain the PPR for the NMLE was developed using logistic regression analysis at five time points, i.e., at admission and the end of the 1st, 2nd, 4th, and 6th grades. All students were divided into high (PPR < 95%) and low (PPR ≥ 95%) risk groups for failing the NMLE at the five time points, respectively, and the movement between the groups during 6 years in school was simulated. RESULTS: Medical students who passed the NMLE had statistically significant factors at each of the 5 time points, and the number of significant variables increased as their grade in school advanced. In addition, two factors extracted at admission were also selected as significant variables at all other time points. Especially, age at entry had a consistent and significant effect during medical school. CONCLUSIONS: Risk analysis based on multiple variables, such as PPR, can inform more effective intervention compared to a single variable, such as performance in the mock exam. A longer prospective study is required to confirm the validity of PPR.

Family Medicine Clerkship Directors' Perspectives on USMLE Pass/Fail Scoring: A CERA Study.

INTRODUCTION: Reports on the effects of changing the United States Medical Licensing Exam (USMLE) Step 1 examination scoring to pass/fail are evolving in the medical literature. This Council of Academic Family Medicine Educational Research Alliance family medicine clerkship directors' study seeks to describe family medicine clerkship directors' perceptions on the impact of incorporation of Step 1 pass/fail score reporting on students' family medicine clerkship performance. METHODS: Ninety-six clerkship directors responded (56.8% response rate). After exclusion of Canadian schools, we analyzed 88 clerkship directors' responses from US schools. We used descriptive statistics for demographics and responses to survey questions. We used ꭓ2 analysis to determine statistically significant associations between survey items. RESULTS: Clerkship directors did not observe changes in students' overall clinical performance after Step 1 pass/fail scoring (60.8%). Fifty percent of clerkship directors reported changes in Step 1 timing recommendations in the past 3 years. Reasons included curriculum redesign (30.5%), COVID (4.5%), change in Step 1 to pass/fail (11.0%), and other reasons (3.7%). Forty-five percent of these clerkship directors did not observe a change in students' clinical medical knowledge after Step 1 went to pass/fail. Eighty-four percent of these clerkship directors did not compare student performance on clerkship standardized exams before and after Step 1 score changes. We found no significant relationship between Step 1 timing and student performance. CONCLUSIONS: This study represents an early description of family medicine clerkship directors' perceived observations of the impact of Step 1 scoring changes on student performance. Continued investigation of the effects of USMLE Step 1 pass/fail scoring should occur.

Application of the "Plan-Do-Check-Action" plan in improving the pass rate of the "National Medical Licensing Examination".

BACKGROUND: The National Medical Licensing Examination (NMLE) is the only objective, standardized metric to evaluate whether a medical student possessing the professional knowledge and skills necessary to work as a physician. However, the overall pass rate of NMLE in our hospital in 2021 was much lower than that of Peking Union Medical College Hospital, which was required to be further improved. METHODS: To find the reasons for the unsatisfactory performance in 2021, the quality improvement team (QIT) organized regular face-to-face meetings for in-depth discussion and questionnaire, and analyzed the data by "Plato analysis" and "Brainstorming method". After finding out the reasons, the "Plan-Do-Check-Action" (PDCA) cycle was continued to identify and solve problems, which included the formulation and implementation of specific training plans by creating the "Gantt charts", the check of effects, and continuous improvements from 2021 to 2022. Detailed information about the performance of students in 2021 and 2022, and the attendance, assessment, evaluation and suggestions from our hospital were provided by the relevant departments, and the pass rate-associated data was collected online. RESULTS: After the PDCA plan, the pass rate of NMLE in our hospital increased by 10.89% from 80.15% in 2021 to 91.04% in 2022 (P = 0.0109), with the pass rate of skill examination from 95.59% in 2021 to 99.25% in 2022 (P = 0.0581) and theoretical examination from 84.5% in 2021 to 93.13% in 2022 (P = 0.027). Additionally, the mean scores of all examinees increased with the theoretical examination score increasing from 377.0 ± 98.76 in 2021 to 407.6 ± 71.94 in 2022 (P = 0.004). CONCLUSIONS: Our results showed a success application of the PDCA plan in our hospital which improved the pass rate of the NMLE in 2022, and the PDCA plan may provide a practical framework for future medical education and further improve the pass rate of NMLE in the next year.

Performance of ChatGPT Across Different Versions in Medical Licensing Examinations Worldwide: Systematic Review and Meta-Analysis.

BACKGROUND: Over the past 2 years, researchers have used various medical licensing examinations to test whether ChatGPT (OpenAI) possesses accurate medical knowledge. The performance of each version of ChatGPT on the medical licensing examination in multiple environments showed remarkable differences. At this stage, there is still a lack of a comprehensive understanding of the variability in ChatGPT's performance on different medical licensing examinations. OBJECTIVE: In this study, we reviewed all studies on ChatGPT performance in medical licensing examinations up to March 2024. This review aims to contribute to the evolving discourse on artificial intelligence (AI) in medical education by providing a comprehensive analysis of the performance of ChatGPT in various environments. The insights gained from this systematic review will guide educators, policymakers, and technical experts to effectively and judiciously use AI in medical education. METHODS: We searched the literature published between January 1, 2022, and March 29, 2024, by searching query strings in Web of Science, PubMed, and Scopus. Two authors screened the literature according to the inclusion and exclusion criteria, extracted data, and independently assessed the quality of the literature concerning Quality Assessment of Diagnostic Accuracy Studies-2. We conducted both qualitative and quantitative analyses. RESULTS: A total of 45 studies on the performance of different versions of ChatGPT in medical licensing examinations were included in this study. GPT-4 achieved an overall accuracy rate of 81% (95% CI 78-84; P<.01), significantly surpassing the 58% (95% CI 53-63; P<.01) accuracy rate of GPT-3.5. GPT-4 passed the medical examinations in 26 of 29 cases, outperforming the average scores of medical students in 13 of 17 cases. Translating the examination questions into English improved GPT-3.5's performance but did not affect GPT-4. GPT-3.5 showed no difference in performance between examinations from English-speaking and non-English-speaking countries (P=.72), but GPT-4 performed better on examinations from English-speaking countries significantly (P=.02). Any type of prompt could significantly improve GPT-3.5's (P=.03) and GPT-4's (P<.01) performance. GPT-3.5 performed better on short-text questions than on long-text questions. The difficulty of the questions affected the performance of GPT-3.5 and GPT-4. In image-based multiple-choice questions (MCQs), ChatGPT's accuracy rate ranges from 13.1% to 100%. ChatGPT performed significantly worse on open-ended questions than on MCQs. CONCLUSIONS: GPT-4 demonstrates considerable potential for future use in medical education. However, due to its insufficient accuracy, inconsistent performance, and the challenges posed by differing medical policies and knowledge across countries, GPT-4 is not yet suitable for use in medical education. TRIAL REGISTRATION: PROSPERO CRD42024506687; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=506687.

Association of USMLE Step 1 Pass/Fail Reporting with Interview and Match Outcomes.

OBJECTIVE: To investigate interview and match outcomes of medical students who received pass/fail USMLE reporting vs medical students with numeric scoring during the same period. DESIGN: Retrospective analysis of a cross-sectional survey-based study. SETTING: United States 2023 residency match. PARTICIPANTS: Medical student applicants in the 2023 residency match cycle who responded to the Texas Seeking Transparency in Application to Residency (STAR) survey. RESULTS: Among 6756 applicants for the 2023 match, 496 (7.3%) took USMLE Step 1 with pass/fail reporting. Pass/fail reporting was associated with lower USMLE Step 2-CK scores (245.9 vs 250.7), fewer honored clerkships (2.4 vs 3.1), and lower Alpha Omega Alpha membership (12.5% vs 25.2%) (all p < 0.001). Applicants with numeric USMLE Step 1 scores received more interview offers after adjusting for academic performance (beta coefficient 1.04 (95% CI 0.28-1.79); p = 0.007). Numeric USMLE Step 1 scoring was associated with more interview offers in nonsurgical specialties (beta coefficient 1.64 [95% CI 0.74-2.53]; p < 0.001), but not in general surgery (beta coefficient 3.01 [95% CI -0.82 to 6.84]; p = 0.123) or surgical subspecialties (beta coefficient 1.92 [95% CI -0.78 to 4.62]; p = 0.163). Numeric USMLE Step 1 scoring was not associated with match outcome. CONCLUSIONS: Applicants with numeric USMLE Step 1 scoring had stronger academic profiles than those with pass/fail scoring; however, adjusted analyses found only weak associations with interview or match outcomes. Further research is warranted to assess longitudinal outcomes.

[The National Competency-Based Catalogue of Learning Objectives for Medicine 2.0 for the subject Physical and Rehabilitative Medicine, Naturopathic Medicine. Consensus of Scientific Medical Societies]. / Der Nationale Kompetenzbasierte Lernzielkatalog 2.0 für das Fach "Physikalische und Rehabilitative Medizin, Naturheilverfahren". Konsentierte Ergebnisse beteiligter Fachgesellschaften der AWMF.

BACKGROUND: The learning objectives in the current cross-sectional subject "Rehabilitation, Physical Medicine, Naturopathic Medicine" have been revised as part of the further development of the National Competency-Based Catalogue of Learning Objectives for Medicine (NKLM) to its new version 2.0. Since the NKLM is designed as an interdisciplinary catalogue, a subject assignment seemed necessary from the point of view of various stakeholders. Thus, the German Association of Scientific Medical Societies (AWMF) and the German medical faculties initiated a subject assignment process. The assignment process for the subject "Physical and Rehabilitative Medicine, Naturopathic Medicine" (PRM-NHV; according to the subject list of the first draft of the planned novel medical license regulations from 2020) is presented in this paper. MATERIAL AND METHODS: The AWMF invited its member societies to participate in the assignment of learning objectives of chapters VI, VII, and VIII of the NKLM 2.0 to the individual subjects to which they consider to contribute in teaching. For "PRM-NHV", representatives of the societies for rehabilitation sciences (DGRW), physical and rehabilitation medicine (DGPRM), orthopaedics and traumatology (DGOU), as well as for naturopathy (DGNHK) participated. In a structured consensus process according to the DELPHI methodology, the learning objectives were selected and consented. Subsequently, subject recommendations were made by the AWMF for each learning objective. RESULTS: From the NKLM 2.0, a total of 100 competency-based learning objectives of chapters VII and VIII for the subject "PRM-NHV" were consented by the representatives of the involved societies for presentation on the NKLM 2.0 online platform. CONCLUSIONS: In the context of the revision process of medical studies in Germany and under the umbrella of the AWMF and the German medical faculties, a broad consensus of competency-based learning objectives in the subject "PRM-NHV" could be achieved. This provides an important orientation for all medical faculties both for the further development of teaching in the cross-sectional subject "Rehabilitation, Physical Medicine, Naturopathic Medicine" according to the 9th revision of the medical license regulations, which has been valid for twenty years, and for the preparation of the corresponding subjects in the draft bill of the novel license regulations.

Out-of-Hospital Birth in Medical Education: A Thematic Analysis of Question Banks for Medical Licensing Examinations in the United States.

OBJECTIVES: This study assessed the content of US Medical Licensing Examination question banks with regard to out-of-hospital births and whether the questions aligned with current evidence. METHODS: Three question banks were searched for key words regarding out-of-hospital births. A thematic analysis was then utilized to analyze the results. RESULTS: Forty-seven questions were identified, and of these, 55% indicated a lack of inadequate, limited, or irregular prenatal care in the question stem. CONCLUSIONS: Systematic studies comparing prenatal care in out-of-hospital births versus hospital births are nonexistent, leading to the potential for bias and adverse outcomes. Adjustments to question stems that accurately portray current evidence are recommended.

Letters of Reference in the Current Era.

ABSTRACT: Letters of reference (LORs) are a common component of the application process for residency training programs. With the United States Medical Licensing Examination Step 1 transitioning to pass/fail grading and with the increasing use of holistic review, the potential role of LORs is rising in importance. Among some key benefits are the ability to provide a broader and more holistic view of applicants, which can include highlighting elements of experiences or skills that could be missed in their application, as well as providing a third-party assessment of the applicant external to their rotation experiences. However, LORs also face issues, including variation in quality, challenges with comparability, and risk of bias. In this article, the authors discuss the unique benefits, limitations, and best practice recommendations for LORs in academic medicine. The authors also discuss future directions, including the role of artificial intelligence, unblinded, and co-created LORs.

From admissions to licensure: education data associations from a multi-centre undergraduate medical education collaboration.

This paper reports the findings of a Canada based multi-institutional study designed to investigate the relationships between admissions criteria, in-program assessments, and performance on licensing exams. The study's objective is to provide valuable insights for improving educational practices across different institutions. Data were gathered from six medical schools: McMaster University, the Northern Ontario School of Medicine University, Queen's University, University of Ottawa, University of Toronto, and Western University. The dataset includes graduates who undertook the Medical Council of Canada Qualifying Examination Part 1 (MCCQE1) between 2015 and 2017. The data were categorized into five distinct sections: demographic information as well as four matrices: admissions, course performance, objective structured clinical examination (OSCE), and clerkship performance. Common and unique variables were identified through an extensive consensus-building process. Hierarchical linear regression and a manual stepwise variable selection approach were used for analysis. Analyses were performed on data set encompassing graduates of all six medical schools as well as on individual data sets from each school. For the combined data set the final model estimated 32% of the variance in performance on licensing exams, highlighting variables such as Age at Admission, Sex, Biomedical Knowledge, the first post-clerkship OSCE, and a clerkship theta score. Individual school analysis explained 41-60% of the variance in MCCQE1 outcomes, with comparable variables to the analysis from of the combined data set identified as significant independent variables. Therefore, strongly emphasising the need for variety of high-quality assessment on the educational continuum. This study underscores the importance of sharing data to enable educational insights. This study also had its challenges when it came to the access and aggregation of data. As such we advocate for the establishment of a common framework for multi-institutional educational research, facilitating studies and evaluations across diverse institutions. This study demonstrates the scientific potential of collaborative data analysis in enhancing educational outcomes. It offers a deeper understanding of the factors influencing performance on licensure exams and emphasizes the need for addressing data gaps to advance multi-institutional research for educational improvements.

Gender differences in learning and study strategies impact medical students' preclinical and USMLE step 1 examination performance.

BACKGROUND: Evaluation of students' learning strategies can enhance academic support. Few studies have investigated differences in learning strategies between male and female students as well as their impact on United States Medical Licensing Examination® (USMLE) Step 1 and preclinical performance. METHODS: The Learning and Study Strategies Inventory (LASSI) was administered to the classes of 2019-2024 (female (n = 350) and male (n = 262)). Students' performance on preclinical first-year (M1) courses, preclinical second-year (M2) courses, and USMLE Step 1 was recorded. An independent t-test evaluated differences between females and males on each LASSI scale. A Pearson product moment correlation determined which LASSI scales correlated with preclinical performance and USMLE Step 1 examinations. RESULTS: Of the 10 LASSI scales, Anxiety, Attention, Information Processing, Selecting Main Idea, Test Strategies and Using Academic Resources showed significant differences between genders. Females reported higher levels of Anxiety (p < 0.001), which significantly influenced their performance. While males and females scored similarly in Concentration, Motivation, and Time Management, these scales were significant predictors of performance variation in females. Test Strategies was the largest contributor to performance variation for all students, regardless of gender. CONCLUSION: Gender differences in learning influence performance on STEP1. Consideration of this study's results will allow for targeted interventions for academic success.

Does small talk with a medical provider affect ChatGPT's medical counsel? Performance of ChatGPT on USMLE with and without distractions.

Efforts are being made to improve the time effectiveness of healthcare providers. Artificial intelligence tools can help transcript and summarize physician-patient encounters and produce medical notes and medical recommendations. However, in addition to medical information, discussion between healthcare and patients includes small talk and other information irrelevant to medical concerns. As Large Language Models (LLMs) are predictive models building their response based on the words in the prompts, there is a risk that small talk and irrelevant information may alter the response and the suggestion given. Therefore, this study aims to investigate the impact of medical data mixed with small talk on the accuracy of medical advice provided by ChatGPT. USMLE step 3 questions were used as a model for relevant medical data. We use both multiple-choice and open-ended questions. First, we gathered small talk sentences from human participants using the Mechanical Turk platform. Second, both sets of USLME questions were arranged in a pattern where each sentence from the original questions was followed by a small talk sentence. ChatGPT 3.5 and 4 were asked to answer both sets of questions with and without the small talk sentences. Finally, a board-certified physician analyzed the answers by ChatGPT and compared them to the formal correct answer. The analysis results demonstrate that the ability of ChatGPT-3.5 to answer correctly was impaired when small talk was added to medical data (66.8% vs. 56.6%; p = 0.025). Specifically, for multiple-choice questions (72.1% vs. 68.9%; p = 0.67) and for the open questions (61.5% vs. 44.3%; p = 0.01), respectively. In contrast, small talk phrases did not impair ChatGPT-4 ability in both types of questions (83.6% and 66.2%, respectively). According to these results, ChatGPT-4 seems more accurate than the earlier 3.5 version, and it appears that small talk does not impair its capability to provide medical recommendations. Our results are an important first step in understanding the potential and limitations of utilizing ChatGPT and other LLMs for physician-patient interactions, which include casual conversations.

Evaluating Urology Program Directors' Perception on Resident Application Parameters Following the Transition of USMLE Step 1 to Pass/Fail.

OBJECTIVE: To investigate how the shift of the United States Medical Licensing Examination (USMLE) Step 1 to a Pass/Fail (P/F) scoring system impacts the perceptions of Urology Program Directors (PDs) on evaluating urology residency applicants. METHODS AND MATERIALS: A cross-sectional survey was sent to 117 PDs, including questions about program characteristics, perceptions of shelf scores and medical school rank post-transition, beliefs about the predictive value of Step 1 and Step 2 Clinical Knowledge (CK) scores for board success and residency performance, and changes in applicant parameter ranking. RESULTS: Forty-five PDs (38% response rate) participated. Notably, 49% favored releasing quantitative clerkship grades, and 71% valued medical school rank more. Opinions on Step 1 scores' correlation with board success were split (49% agreed), and 44% endorsed Step 2 CK scores' connection to board performance. As predictors of good residents, only 9% and 22% considered Step 1 and Step 2 CK scores, respectively, indicative. Clerkship grades and Urology rotation recommendation letters maintained significance, while research experience gained importance. Step 2 CK scores' importance rose but did not match Step 1 scores' previous significance. CONCLUSION: The transition to P/F for USMLE Step 1 adds intricacies to urology residency selection, exposing PDs' uncertainties regarding clerkship grades and the relevance of medical school rank. This research underscores the dynamic nature of urology residency admissions, emphasizing the increasing importance of research in evaluating applicants and a diminishing emphasis on volunteering and leadership.

A validity study of COMLEX-USA Level 3 with the new test design.

CONTEXT: The National Board of Osteopathic Medical Examiners (NBOME) administers the Comprehensive Osteopathic Medical Licensing Examination of the United States (COMLEX-USA), a three-level examination designed for licensure for the practice of osteopathic medicine. The examination design for COMLEX-USA Level 3 (L3) was changed in September 2018 to a two-day computer-based examination with two components: a multiple-choice question (MCQ) component with single best answer and a clinical decision-making (CDM) case component with extended multiple-choice (EMC) and short answer (SA) questions. Continued validation of the L3 examination, especially with the new design, is essential for the appropriate interpretation and use of the test scores. OBJECTIVES: The purpose of this study is to gather evidence to support the validity of the L3 examination scores under the new design utilizing sources of evidence based on Kane's validity framework. METHODS: Kane's validity framework contains four components of evidence to support the validity argument: Scoring, Generalization, Extrapolation, and Implication/Decision. In this study, we gathered data from various sources and conducted analyses to provide evidence that the L3 examination is validly measuring what it is supposed to measure. These include reviewing content coverage of the L3 examination, documenting scoring and reporting processes, estimating the reliability and decision accuracy/consistency of the scores, quantifying associations between the scores from the MCQ and CDM components and between scores from different competency domains of the L3 examination, exploring the relationships between L3 scores and scores from a performance-based assessment that measures related constructs, performing subgroup comparisons, and describing and justifying the criterion-referenced standard setting process. The analysis data contains first-attempt test scores for 8,366 candidates who took the L3 examination between September 2018 and December 2019. The performance-based assessment utilized as a criterion measure in this study is COMLEX-USA Level 2 Performance Evaluation (L2-PE). RESULTS: All assessment forms were built through the automated test assembly (ATA) procedure to maximize parallelism in terms of content coverage and statistical properties across the forms. Scoring and reporting follows industry-standard quality-control procedures. The inter-rater reliability of SA rating, decision accuracy, and decision consistency for pass/fail classifications are all very high. There is a statistically significant positive association between the MCQ and the CDM components of the L3 examination. The patterns of associations, both within the L3 subscores and with L2-PE domain scores, fit with what is being measured. The subgroup comparisons by gender, race, and first language showed expected small differences in mean scores between the subgroups within each category and yielded findings that are consistent with those described in the literature. The L3 pass/fail standard was established through implementation of a defensible criterion-referenced procedure. CONCLUSIONS: This study provides some additional validity evidence for the L3 examination based on Kane's validity framework. The validity of any measurement must be established through ongoing evaluation of the related evidence. The NBOME will continue to collect evidence to support validity arguments for the COMLEX-USA examination series.