Affordable Access

deepdyve-link
Publisher Website

United States Medical Licensing Examination and American Board of Pediatrics Certification Examination Results: Does the Residency Program Contribute to Trainee Achievement.

Authors
  • Welch, Thomas R1
  • Olson, Brad G2
  • Nelsen, Elizabeth2
  • Beck Dallaghan, Gary L3
  • Kennedy, Gloria A2
  • Botash, Ann2
  • 1 Department of Pediatrics, State University of New York Upstate Medical University, Syracuse, NY. Electronic address: [email protected]
  • 2 Department of Pediatrics, State University of New York Upstate Medical University, Syracuse, NY.
  • 3 Office of Medical Education, University of Nebraska Medical Center, Omaha, NE.
Type
Published Article
Journal
The Journal of pediatrics
Publication Date
Sep 01, 2017
Volume
188
Identifiers
DOI: 10.1016/j.jpeds.2017.05.057
PMID: 28629684
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

To determine whether training site or prior examinee performance on the US Medical Licensing Examination (USMLE) step 1 and step 2 might predict pass rates on the American Board of Pediatrics (ABP) certifying examination. Data from graduates of pediatric residency programs completing the ABP certifying examination between 2009 and 2013 were obtained. For each, results of the initial ABP certifying examination were obtained, as well as results on National Board of Medical Examiners (NBME) step 1 and step 2 examinations. Hierarchical linear modeling was used to nest first-time ABP results within training programs to isolate program contribution to ABP results while controlling for USMLE step 1 and step 2 scores. Stepwise linear regression was then used to determine which of these examinations was a better predictor of ABP results. A total of 1110 graduates of 15 programs had complete testing results and were subject to analysis. Mean ABP scores for these programs ranged from 186.13 to 214.32. The hierarchical linear model suggested that the interaction of step 1 and 2 scores predicted ABP performance (F[1,1007.70] = 6.44, P = .011). By conducting a multilevel model by training program, both USMLE step examinations predicted first-time ABP results (b = .002, t = 2.54, P = .011). Linear regression analyses indicated that step 2 results were a better predictor of ABP performance than step 1 or a combination of the two USMLE scores. Performance on the USMLE examinations, especially step 2, predicts performance on the ABP certifying examination. The contribution of training site to ABP performance was statistically significant, though contributed modestly to the effect compared with prior USMLE scores. Copyright © 2017 Elsevier Inc. All rights reserved.

Report this publication

Statistics

Seen <100 times