USMLE Step 2 CK Facts

When should I take Step 2?

Many U.S. medical schools now require a passing score on the USMLE Step 2 examination before graduation. It is also recommended that students take the exam after finishing all basic clinical clerkships. When the test was hand-written (pre-2000), it was only offered twice annually, but now that the exam is computerized students can essentially take it anytime during their fourth year. With regards to the application process, taking the examination early in fourth year typically makes the results available to program directors whereas taking the examination later means that the score will not be factored into the residency selection process. Students disappointed in their Step 1 score may opt to take the test earlier in hopes of achieving a more impressive score for residency selection committees to consider.

A 2004 study looked at the relationship between the timing of when students took the exam and their eventual score. Students who took the examination in June through August of fourth year tended to score better than they had on Step 1, whereas students who took the exam in spring of fourth year tended to do worse than they had on Step I. This may reflect better recall of clerkship knowledge when taking the exam earlier, as well as stronger motivation to perform well when taking the test at a time that would make the results available to residency selection committees. While it seems that for most students, the primary consideration in timing Step 2 is strategy with regards to residency application, students who feel they may be at risk of failing the exam may benefit from taking the test earlier in the fourth year so that key clinical information is fresher in their minds.

Reference

Pohl CA, Robeson MR, Veloski J. USMLE Step 2 Performance and Test Administration Date in the Fourth Year of Medical School. Academic Medicine, 79(10), October Supplement 2004, S49-S51.

Why is Step 2 important?

With the extreme focus placed on USMLE Step 1 both by program directors and medical students, many may overlook Step 2 and perhaps invest considerably less effort preparing for this exam. However, as the material on Step 2 is more clinically relevant and patient centered than on Step 1, the material studied and tested is likely to be more relevant to students in their future careers. Though there is evidence that much of the material tested by Step 1 may not be retained as students progress in their careers, one would hope that the more clinically relevant material of Step 2 might be retained better in the long run (see "How much basic science material from USMLE Step 1 will I end up remembering in the long term?").

In addition to its educational value, Step 2 has also been shown to be an important factor in the residency selection process. A 2006 survey issued to program directors nationwide determined that Step 2 CK and CS are the fifth and sixth most important factors, respectively, in resident selection (notably ranking higher than AOA status, research experience, class rank, and medical school reputation). Step 1 scores were considered the second most important factor. (There was some variation with respect to Step 2 importance between specialties, with program directors from fields considered 'less competitive' giving more weight to Step 2 CK and CS scores than program directors from more competitive specialties.) Given their importance, fourth year students should strongly consider taking these tests early enough in the application process to allow the results to be available to residency program directors.

Reference

Green M, Jones P, Thomas, JX. Selection Criteria for Residency: Results of a National Program Directors Survey. Academic Medicine 84(3), March 2009, pp 362-367.

Is the Step 1 score predictive of the Step 2 score?

While the USMLE Step 1 and 2 exams cover different material, basic sciences and clinical knowledge respectively, there are similarities between the tests as they are both written by the same governing body. A study of USMLE performance in the mid 1990s (when the exam was converted from the NBME to the USMLE exams) showed a strong correlation between Step 1 and 2 performance. A more recent study in 2006 similarly showed correlated results.

Interestingly, the correlation was stronger for students from medical schools with a higher average USMLE Step 1 score; this may be reflective of an environment that encourages strong study habits and drive to achieve. The correlation was also stronger at larger medical schools, possibly because some smaller schools have a more narrow curricular focus or less diverse patient population.

Multiple studies have also demonstrated a difference in performance on the tests based on gender. While the overall pass rates between genders are similar, studies have demonstrated that male students generally outperform females on Step 1 while females perform equal to or better than males on Step 2 CK. While controversial, some believe this may be in part because female students may have less of a basic science background upon entering medical school leading to lower Step 1 scores while Step 2 CK includes areas which have traditionally been of more interest to female students, such as pediatrics, psychiatry, and obstetrics/gynecology, in turn leading to higher scores. Overall, there was a stronger correlation of Step 1 with Step 2 performance in males than females.

References

Cuddy MM, Swanson DB, Dillon GF, Holtman MC, Clauser BE. A Multilevel Analysis of the Relationships Between Selected Examinee Characteristics and United States Medical Licensing Examination Step 2 Clinical Knowledge Performance: Revisiting Old Findings and Asking New Questions. Academic Medicine 81(10), October 2006 Supplement, S103-S107.

Case SM, Swanson DB, Ripkey DR, Bowles LT, Melnick DE. Performance of the Class of 1994 in the new era of the USMLE. Academic Medicine 71 (10), October supplement 1996, S91-S93.

Is there a correlation between performance on USMLE exams and specialty board licensing exams later in training?

Residency programs weigh USMLE performance heavily in the resident selection process. One often-cited justification for this is that students who perform well on the USMLE likely have the skills to perform well on board certifying exams too. This is important, as poor resident performance on board certifying exams reflects poorly on the residency program.

A 2009 study examined the relationship between USMLE performance and resident performance on the internal medicine in-training examination. While the in-training examination is not directly involved in board certification for internal medicine, performance on the test has been independently correlated with performance on the certifying exam. The study demonstrated that there was a positive correlation between USMLE scores and performance on the in-training exam. Of Step 1, Step 2, and Step 3, the correlation was strongest by far for the USMLE Step 2. It seems this correlation holds even for specialty fields; a 2006 study examined the relationship between performance on USMLE Steps 1 and 2 with performance on the orthopedic surgery in-service training exam and found similar results (specifically, a positive correlation between scores on USMLE Step 2 and on the in-service training exam, but no significant correlation with Step 1 scores). Overall, these results suggest that performance on USMLE tests may indicate future success on board certification exams whether the student chooses a broad or narrow specialty, with the Step 2 exam likely having the closest correlation.

References

Perez JA, Greer S. Correlation of United States Medical Licensing Examination and Internal Medicine In-Training Examination performance. Adv in Health Sci Educ (2009) 14:753–758.

Black KP, Abzug JM, Chinchilli VM. Orthopedic In-Training Examination Scores: A Correlation with USMLE Results. Journal of Bone and Joint Surgery. 88A(3), March 2006, 671-676.

Which aspects of my clinical training will most impact my score on the USMLE Step 2 exam?

Clinical clerkships vary widely at different medical schools. Even students from within the same medical school may have quite different clerkship experiences. A 2009 study attempted to quantify the effect that different variables pertaining to clinical clerkships have on USMLE Step 2 performance. The most significant factor appeared to be the number of patients cared for in a day. Students who had a significant improvement in their Step 2 score over their Step 1 score saw significantly more patients per day. Students who had a significant drop in score cared for significantly fewer patients per day. However, the study did not examine the effect of the precise number of patients cared for. One might expect an eventual threshold effect, beyond which increased patient load brought diminishing returns secondary to students becoming overwhelmed.

Among students with high Step 1 scores, those with better Step 2 scores tended to have more exposure to hospitalists, higher NBME shelf exam scores, and longer attending rounds. In contrast, for students with lower Step 1 scores, more structured learning formats such as small groups, computer-based instruction, and separate rounds with a teaching attending seemed to have more benefit for the Step 2 score. Both groups of students tended to do better on Step 2 with 4 week rather than 2 week clinical clerkships. These results suggest that some curriculum reforms may not benefit all students equally.

Reference

Griffith CH, Wilson JF, Haist SA, Albritton TA, Bognar BA, Cohen SJ, Hoesley CJ, Fagan MJ, Ferenchick GS, Pryor OW, Friedman E, Harrell HE, Hemmer PA, Houghton BL, Kovach R, Lambert D, Loftus TH, Painter TD, Udden MM, Watkins RS, Wong RY. Internal Medicine Clerkship Characteristics Associated with Enhanced Student Examination Performance. Academic Medicine 84(7), July 2009, 895-901.