The arrival of COVID-19 into the United States and subsequent surge of cases in early 2020 led to significant efforts first at containment and then mitigation of the virus. Looking back from our vantage point a year later, it’s hard to find much that hasn’t been impacted.

One recent notable casualty is the Step 2 Clinical Skills (CS)—the performance-based component within the United States Medical Licensing Examination (USMLE). The exam had been suspended since March 2020 when on January 26 of this year, the USMLE announced its discontinuation. [Note: On February 11, the National Board of Osteopathic Medical Examiners (NBOME) announced that its Level 2-PE was “postponed indefinitely.”] Citing no one factor as the sole determinant, USMLE shut down its CS exam explaining that it had become clear the program would be unable to relaunch with an exam “appreciably better” than its pre-COVID version.

From a medical regulatory perspective, this marks a significant—and one could argue disappointing—milestone in the history of the medical licensing exam. I say this because if we step back to consider the long history of the licensing examination, one can discern a persistent aspiration to assess more fully and deeply into the qualifications of the prospective physician.

Some of the earliest exams for medical licensure were oral exams perhaps not too far akin from traditional academic and ecclesiastical forms of disputation. Before the end of the 19^th century most boards committed to a written examination featuring long-answer or essay style questions designed to allow the candidates to demonstrate their depth and breadth of knowledge.

Over time, however, many lamented that these test items relied too often upon simple recall and memorization of specific “facts”—in essence, too little in the way of clinical reasoning that demonstrates the application of medical knowledge. These critics didn’t require Miller’s pyramid of [clinical] competence to recognize the licensing exam should aspire to more. State medical boards desperately sought to assess beyond purely cognitive assessment.

The same era of medical education reforms that delivered the Flexner Report saw state medical boards (starting with Ohio) institute performance-based and/or procedural exams incorporating laboratory work and patient encounters into their licensing exam. By 1919, roughly twenty boards incorporated some type of procedural or skill-based component to their exams. These long-ago predecessors to the NBME Part III bedside and later Step 2CS did not persist as high costs and logistical challenges to such assessments led to their demise over the course of the 1920s.

The late 1960s-70s saw the complete transition to exams wholly developed and administered nationally, e.g., Federation Licensing Exam (FLEX); NBME Parts…and later USMLE and COMLEX. Such exams delivered better-quality assessment with higher reliability and over time a growing technical literature supporting the validity of the licensing decisions made, at least in part, upon passing these tests. The shift to computer-delivery in the early 2000s facilitated the use of new formats better able to assess specific competencies such as patient management with the computer-case simulations on USMLE Step 3.

With the inclusion of Step 2 CS in 2004, licensing boards enjoyed another tool designed to help ensure at least minimal proficiency in communication and clinical skills of their prospective licensees. Throughout the exam’s existence, state medical boards tended to be among its strongest and most reliable supporters.

Future prospects for assessing (within the licensing exam) competencies associated with clinical skills are a work in progress. In the wake of the USMLE decision, several avenues seem open. Some see this as the opportunity for a fresh start—to reimagine assessment of the competencies we associate with clinical skills by drawing upon the formative assessments now embedded in the medical education curriculum since the introduction of CS nearly twenty years ago. Schools are now free from the practical constraints of aligning their OSCEs, at least in part, as assessments that must prepare their students for Step 2CS. With the practical constraints of a “preparatory” role removed, educators should feel empowered to consider innovative approaches to teaching and assessing these critical competencies.

Such enthusiasm leads some to feel that teaching clinical skills can now safely be left to the medical schools—after all, it’s an accreditation standard, right? In fact, why worry about assessment of these competencies within licensing? Why not simply let the schools “attest” to the proficiency of their graduates in the wake of Step 2CS discontinuation. Wouldn’t such an attestation alleviate any lingering concerns among the medical regulatory community?  

One would like to think so yet there are those among us quietly concerned that over time, the financial pressures arising from the cost of maintaining an OSCE-style assessment will chip away at medical schools’ collective commitment to assess these competencies—or at the very least within the context of an OSCE-style format.

We would all be well-advised to monitor closely how schools respond to the discontinuation of Step 2 CS and the current suspension of Level 2-PE. Will this present moment give way to a flowering of innovative assessment…or will the grinding pressures of financial entrenchment prove this moment to be a false spring?

The views expressed are those of the author and do not reflect those of the Federation of State Medical Boards or the USMLE program