Fleecing Medical Students: A Cost-Benefit Analysis of Step 2 CS

By: Abhas Gupta

Update (04/02/08): This analysis refers exclusively to US-based M.D. Students, not D.O. or Foreign Medical Graduates.

As 4th-year rolls around, the inevitable costs of applying to residencies begin to weigh one down. The list seems endless: application fees, flights, lodging, new suit, new tie, PDA, and of course, licensing exams—Step 2 CS, alone, now costs $1050. Further investigation into the costs and benefits of Step 2 CS suggest that not all licensing requirements are constructive and perhaps, this time, medical students are being taken for a ride.

Why is Step 2 CS administered in the first place? The USMLE website states:

Step 2 CS uses standardized patients to test medical students and graduates on their ability to gather information from patients, perform physical examinations, and communicate their findings to patients and colleagues.

The NBME® reports that in the 2005 academic year, 16,936 medical students took the Step 2 CS and 98% passed². Of the test takers, 325 had failed previously but overwhelmingly passed this time around (97%) . Put another way: In the 2005 academic year, 4th-year medical students were made to spend $17.8 million (16,936 students x $1050/student) on Step 2 CS alone so that 363 students with unsatisfactory clinical skills could be identified and retested. After retesting, only about 11 students would still fall short of satisfactory clinical proficiency.

The cost of Step 2 CS to students is substantial; the $17.8 million calculated above does not even include the ancillary costs of traveling and housing—an additional $2 - $3 million—required to attend one of the only five nationwide testing centers. But what about the test's benefits, both to the student and to society? Presumably, identifying poor-performers is better for students and better for patients; students that need remediation are made aware of their weaknesses and patients are safer at the hands of well-vetted interns.

The relationship between Step 2 CS and these benefits is not as direct as suggested above. There are numerous mechanisms already in place to identify poor-performers both at the medical school and residency levels. A 1999 study published in Academic Medicine² found that 48% of medical schools had incorporated standardized patients to evaluate the clinical skills of their students. One can reasonably expect that this fraction has increased since the study's publication. Also, all residency programs have numerous internal teaching tools and safety nets as well as external licensing requirements to cultivate and monitor their interns. The value of Step 2 CS can thus be considered minimal in the short term and all-but-useless by residency.

The justifications for an expensive Step 2 CS are therefore few and trivial. The following are three recommendations that could potentially mitigate the sizeable cost to students and still preserve the test's professional purpose:

1. Implement other metrics to identify students at high-risk for failing the Step 2 CS - factors such as scores on Step 1 or Step 2 CK, serious clinical infractions, or consistently poor-performance in clerkships could help identify students with clinical skills weaknesses. The Step 2 CS could then be made mandatory for these students and these students only


2. Minimize redundancy in clinical skills testing – many institutions already test clinical skills, albeit with different formats and styles. The NBME could establish a set of national guidelines dictating how the evaluation should be conducted. Students at institutions that comply with these guidelines would then be exempt from taking the Step 2 CS


3. Eliminate Step 2 CS from residency programs' entrance criteria – leadership from these programs is necessary for students and medical school administrators to consider any changes to the current system. These programs should be made aware of the negligible benefit of Step 2 CS and come to the aid of their future interns

These three actions would make the Step 2 CS much more palatable to medical students without detracting from the licensing exam's purpose. I am eager to hear more from other medical students—post your comments anonymously, if you prefer. If there's enough interest, perhaps we can consider broaching this issue with the NBME.

With the excitement of the democratic primaries, I have been negligent in posting to this blog. If you have specific topics you would like to see discussed, please forward them here. I am currently considering writing on the health care payment system and pay-for-performance incentives.

References:
1. 2006 USMLE Performance Data.
2. Kassebaum, DG, Eaglen, RH. Shortcomings in the Evaluation of Students’ Clinical Skills and Behaviors in Medical School