Featured, Opinions
Leave a comment

It’s Time to Find a Better Way to Test Soon-To-Be Doctors


Six nervous medical students lined up in front of six heavy doors. I uncovered the piece of paper tacked to the door in front of me: “Mr. Smith: 59 year-old. Chief complaint: chest pain.” I glanced down the narrow hallway to see five other students swimming in identical white coats, a row of ducks shifting weight in our dress shoes, awaiting orders.

“You may now enter the exam room,” a voice projected overhead. With a rush of nerves, I stepped forward. Knock, knock echoed down the hall, as we each push open our doors to greet our patient-actors.

These paid actors, formally known as “standardized patients” or SPs, have been used in medical education since the early 1960s. They improvise along a guiding script that details their symptoms and pieces of their history while a student, such as myself, plays doctor – asks questions, examines, and shares information. The SP visits allow students to practice their history-taking and physical exam skills in a controlled environment before awkwardly fumbling through a sexual history or with the blood pressure cuff in front of a real patient.

Just as importantly, standardized encounters enable faculty – often observing through a live video stream – to evaluate a student’s ability to gather necessary information, order appropriate labs and tests, and arrive at the correct diagnosis and treatment plan. These evaluations happen at the individual medical school level throughout training, at the state level in select states, and then once on a national level – that is, until the national testing centers closed due to COVID-19.

The United States Medical Licensing Exam (USMLE) program, which organizes all medical licensing tests including the national standardized patient portion known as Step 2 Clinical Skills (CS), recently announced that it will suspend the exam for the next 12-18 months. This decision followed a short-lived effort to explore remote testing options, which the USMLE concluded couldn’t be put together in time for current fourth-year students who would need to take the exam in the coming months in order to graduate.

The suspension was received with a wave of support on medical Twitter, the tweets often concluded with the hashtag #EndStep2CS. Along with many other students and physicians, I believe that the exam stakeholders and the wider medical education community should use this disruption in testing to reexamine the costs and benefits of Step 2 CS and to consider reinventing the exam — or possibly ending it permanently.

In a typical year, medical students have to pass this one final patient actor bonanza before they can become doctors. Like all other USMLE exams, Step 2 CS is eight hours long. However, this is the only Step exam that isn’t administered on a computer; rather, it’s offered at just five centers in the country, located in Atlanta, Chicago, Philadelphia, Houston, and Los Angeles.

Students are expected to pay their way and stay in one of these five metropolitan areas on top of the $1,300 exam registration fee.

In a way, medical students have become inured to the expectation that they pony up and hop on a plane. In the days before the pandemic, during the final year of school, students would spend a few months flying — often at a moment’s notice — to interview at different residency programs to secure a spot for their next phase of training. Students would have little control over the schedule and, if interviewing along and between both coasts, could bounce back and forth across the country. On average, students spent between $1,000 and $5,000 on lodging and transportation for interviews. These numbers soared even higher for those applying to more competitive specialties.

On the testing side, students spend $3,485 on registration for Step exams – almost $2,000 of which is spent during the same year students travel for interviews. This figure doesn’t include the costs of study aids, practice tests, or fees incurred to move their exam date.

For the past 16 years, Step 2 CS has been treated like another drop in the bucket, but in reality, it is more like a lake in the ocean. The exam poses a huge financial burden that disproportionately affects lower-income students.

Furthermore, the test’s utility has been called into question, making it even more difficult to justify the expense. Since 2010, 95-98% of North American medical students passed Step 2 CS on their first attempt, and of those who failed, 91% passed on their second try. Not unexpectedly, several single-institution studies have shown that students who failed the first time around were more likely to have had poor pre-clinical grades, failed Step 1, struggled on their clerkships, or performed poorly on the shorter subject exams taken throughout the clinical years.

These associations beg the question: what new information does Step 2 CS really provide about 3-5% of students? And is it worth $1,300 out of the pockets of some of the most deep-in-debt students?

Another concerning finding is that international medical students fail Step 2 CS at 5 times the rate as North American medical students. This statistic has prompted medical educators to ask whether factors such as English proficiency and variations in accent have substantial effects on performance, although insufficient data is made available to evaluate the discrepancy. If English proficiency or accent were major predictors of a student’s ability to pass, that would raise serious doubts about the clinical utility of the exam. Instead, Step 2 CS would be a test of language skills.

In response to the cost and concerns, the American Medical Association Medical Students Section has called to end Step 2 CS. As a rising fourth-year medical student, I echo their unified voice but deviate slightly in my stance.

My hope is that third parties invested in medical education – and not the USMLE alone, which brings in millions of dollars each year from testing fees – will come together to evaluate the costs and benefits of Step 2 CS. As others have suggested, in a healthcare system that strives to eliminate waste and reduce costs, we should apply the same scrutiny to our medical licensing exams.

If a third-party review concludes that an in-person standardized clinical skills exam is warranted before graduating new doctors, then we’ll continue forward. But in that case, this break in routine provides the perfect opportunity to modify the exam’s format and finances. This is the time to minimize the exam’s redundancy, to do away with the need to travel to one of only five cities, and to lower the price of the test itself.

It’s possible that an external review may find that the exam’s benefits pale in comparison to its costs. A thorough review may conclude that the test is redundant, that students’ clinical skills can adequately be assessed using the many other test results and data points collected throughout medical school. Or it may even find that we’re not really testing clinical skills as much as English proficiency. In these cases, I’ll gladly join the chorus of voices and unequivocally state: it is time to end Step 2 CS.

Image credit: final exam (CC BY 2.0) by dcJohn

Orly Farber (1 Posts)

Contributing Writer

Standford University School of Medicine


Orly Farber is a fourth-year medical student class of 2021 at Stanford University School of Medicine in Palo Alto, California. She graduated from the University of Chicago in 2015 with a BA in comparative human development, and then spent two years working in an allergy lab at the National Institutes of Health. In her free time, Orly enjoys running in the nearby foothills and cooking. She plans to pursue a career in surgery this coming year.