The Correlation Between USMLE and COMLEX Exam Scores for ...
Transcript of The Correlation Between USMLE and COMLEX Exam Scores for ...
Lehigh Valley Health NetworkLVHN Scholarly Works
Department of Emergency Medicine
The Correlation Between USMLE and COMLEXExam Scores for Applicants to a Dually- ApprovedEmergency Medicine Residency: An Eight-YearExperienceKathleen E. Kane MDLehigh Valley Health Network, [email protected]
Dawn M. Yenser PGMELehigh Valley Health Network, [email protected]
Laura Semler MSLehigh Valley Health Network, [email protected]
Kevin Weaver DOLehigh Valley Health Network, [email protected]
Gavin Barr Jr. MDLehigh Valley Health Network, [email protected]
See next page for additional authors
Follow this and additional works at: https://scholarlyworks.lvhn.org/emergency-medicine
Part of the Emergency Medicine Commons
This Poster is brought to you for free and open access by LVHN Scholarly Works. It has been accepted for inclusion in LVHN Scholarly Works by anauthorized administrator. For more information, please contact [email protected].
Published In/Presented AtKane, K., Yenser, D., Semler, L., Weaver, K., Barr, Jr, G., Goyke, T., Burckhart, A., Leonetti, A., Yoshioka, I., Kane, B. (2015, April 13).The Correlation Between USMLE and COMLEX Exam Scores for Applicants to a Dually- Approved Emergency Medicine Residency: AnEight-Year Experience. Poster presented at: Council of Emergency Medicine Residency Directors, Annual Academic Assembly,Phoenix, AZ.
AuthorsKathleen E. Kane MD, Dawn M. Yenser PGME, Laura Semler MS, Kevin Weaver DO, Gavin Barr Jr. MD,Terrence E. Goyke DO, Andre Burckhart DO, Adam Leonetti DO, Isamu Yoshioka, and Bryan Kane MD
This poster is available at LVHN Scholarly Works: https://scholarlyworks.lvhn.org/emergency-medicine/340
© 2015 Lehigh Valley Health NetworkFunding Source: Philadelphia College of Osteopathic Medicine, PCOM MEDNet.
Results: The identified 556 applicants are shown in Figure 1. Pair 1 includes applicants with both COMLEX Step-1 and USMLE Step-1 scores (n=486). Pair 2 are those with both COMLEX Step-2 and USMLE Step-2 scores (n=356). For Pair 1, 66% were male with an average age of 28. For Pair 2, 64% were male; the average age was 28. Mean, standard deviation and median for Pair 1 on the COMLEX was 551, 69 and 548, respectively; for the USMLE, it was 216, 16 and 217. Results for Pair 2 on COMLEX were 566, 80 and 562; USMLE results were 228, 18 and 229. As shown in Figure 2, a strong correlation was observed for Pair 1 (r=0.78, p<0.001). A linear regression model controlling for sex and age, a one-point increase in USMLE Step-1 is associated with a 3.55 point increase in the COMLEX Step-1 score (ß=3.55; 95% CI: 3.30-3.80, p<0.001). A similar strong correlation was observed for Pair 2 (r=0.72, p<0.001).
Figure 2a.Scatterplot of USMLE Step-1 and COMLEX Step-1 scores
with least squares regression line
Figure 1.Participant Test Data
COMLEXStep-1N=546
COMLEXStep-2N=491
USMLEStep-2N=386
USMLEStep-1N=489
PAIR ICOMLEX Step-1 USMLE Step-1
N=486
COMLEX Step-1 COMLEX Step-2
N=484
USMLE Step-1 USMLE Step-2
N=324
PAIR IICOMLEX Step-2 USMLE Step-2
N=356
COMLEX Step-1COMLEX Step-2USMLE Step-1USMLE Step-2
N=298
Conclusion: In our cohort, a strong positive correlation between USMLE and COMLEX was found. This relationship may aid EM residency evaluation of applicants who submit test scores with which they are not familiar.
Introduction: With the announced single GME system, EM residencies may see an increase in applicants who submit unfamiliar standardized exam scores. To date, there is limited information positively correlating USMLE and COMLEX scores for EM applicants.
Objective: To determine the correlation between USMLE and COMLEX scores for applicants to an EM residency.
Methods: After IRB approval, we retrospectively gathered all exam scores for applicants to our four-year, 56-member, dually-approved EM residency from 2006-13. Included were applicants who submitted scores for both exams. Demographic analysis was descriptive. Scatterplots were used to visualize pairwise relationships. Multiple linear regression models, stratified by test step, were created with the COMLEX score as the outcome and USMLE score as the predictor value. Participant age and sex were included in each model.
Figure 2b.Scatterplot of USMLE Step-2 and COMLEX Step-2 scores
with least squares regression line
Lehigh Valley Health Network, Allentown, PA
The Correlation Between USMLE and COMLEX Exam Scores for Applicants to a Dually-Approved Emergency Medicine Residency: An Eight-Year Experience
Kathleen E. Kane, MD, Dawn Yenser, Lauren Semler, MS, Kevin R. Weaver, DO, Gavin C. Barr, Jr., MD, Terrence Goyke, DO, Andre J. Burckhart, DO, Adam L. Leonetti, DO, Isamu E. Yoshioka, MD, Bryan G. Kane, MD