Correlations between Accreditation Council for Graduate Medical Education Obstetrics and Gynecology Milestones and American Board of Obstetrics and Gynecology qualifying examination scores: an initial validity study

Background: Since the launch of the Outcome Project in 2001, the graduate medical education community has been working to implement the 6 general competencies. In 2014, all Obstetrics and Gynecology residency programs implemented specialty-specific milestones to advance competency-based assessment. Each clinical competency committee of the Obstetrics and Gynecology program assesses all residents twice a year on the milestones. These data are reported to the Accreditation Council for Graduate Medical Education as part of a continuous quality improvement effort in graduate medical education. Objective: This study aimed to evaluate the correlation between the Accreditation Council for Graduate Medical Education Obstetrics and Gynecology Milestones and residency program graduates’ performance on the American Board of Obstetrics and Gynecology qualifying (written) examination. Study Design: We conducted a validity study of all graduating (postgraduate year 4) Obstetrics and Gynecology residents in 2017 within Accreditation Council for Graduate Medical Education–accredited United States training programs (1260 residents from 242 programs). This cohort of residents began receiving milestone assessments during their postgraduate year 2 in 2014; the first-year milestones were implemented for all Accreditation Council for Graduate Medical Education–accredited Obstetrics and Gynecology programs. This cohort completed their sixth and final milestone assessment at graduation in June 2017 for a total of 6 periods of milestone assessments. Data regarding each resident's milestone ratings in each of the 28 Accreditation Council for Graduate Medical Education subcompetencies for Obstetrics and Gynecology were assessed for their association with candidates’ American Board of Obstetrics and Gynecology qualifying examination scores using a generalized estimating equation regression model. Results: Data were available and analyzed from 1184 residents from 240 programs, representing 94% of the total academic year 2017 graduates of Obstetrics and Gynecology residency training programs. There was a substantial association between most milestone ratings at the 6 assessment points and candidates’ performance on the American Board of Obstetrics and Gynecology qualifying examination. The strongest associations with the American Board of Obstetrics and Gynecology were within all 7 of the subcompetencies of Medical Knowledge (range of slope correlation coefficients at final milestone ratings 3.84–5.17; slope coefficients can be interpreted as the gain in qualifying examination points per unit increase in milestone level). At the final milestone assessment, but more modest associations with the American Board of Obstetrics and Gynecology qualifying examination scores were also seen with 9 of the 11 Patient Care and Procedural Skills subcompetencies, the 2 of 2 Practice-Based Learning and Improvement subcompetencies, the 2 of 2 Systems-Based Practice subcompetencies, and 2 of the 3 Professionalism subcompetencies. Only 1 of the 3 Interpersonal and Communication Skills subcompetencies was associated with American Board of Obstetrics and Gynecology qualifying examination scores. Conclusion: The pattern of associations between the qualifying examination scores and milestone ratings for the 2017 graduating cohort of Obstetrics and Gynecology residents followed a logical pattern, with the strongest associations seen in Medical Knowledge, and lower to no associations in subcompetencies not as effectively assessed on multiple-choice examinations. Although some positive associations were noted for non-Medical Knowledge milestones, these associations could be caused by correlational rating errors with further study needed to better understand these patterns.