Recently the college-age daughter of a friend talked to me about her dream of becoming a doctor. She was doing well as a psychology major and in her pre-medical courses, was working as a research assistant for a pediatrician at a nearby medical school and volunteered on the cancer ward at a children's hospital.
I was impressed.
But her enthusiasm dipped sharply when she told me she was preparing for the MCATs, the Medical College Admission Test, the required standardized test that measures mastery of the pre-medical curriculum. She was putting all her extracurricular work on hold so she could focus on reviewing biology, physics, chemistry and organic chemistry for the exam. "Does my ability to memorize the Krebs cycle and Bernoulli's equation really have anything to do with what kind of doctor I'll be?" she asked.
The answer, it turns out, is yes — and no.
The first MCAT, then referred to as the Scholastic Aptitude Test for Medical Schools, was administered in 1928 and represented an effort to address the significant medical school dropout rates of the time. Up until that point, medical school applicants had been evaluated, and accepted, on the basis of stray bits of biographical information, random letters of endorsement, a few prior grades or the existence of a high school diploma. As many as half of those accepted eventually quit, resulting in huge losses of time, energy, educational resources and money. But thanks in part to the MCAT, by the mid-1940s the medical school attrition rate had plummeted to less than 10 percent, even as the standardized exam was becoming a much-maligned rite of passage for aspiring young doctors.
Over the years, the MCAT has gone through four major revisions and has only strengthened its ability to predict success in medical school, particularly when evaluated in combination with grades. Each year more than 70,000 students take the exam, vying for a little more than 19,500 medical school slots. These days, fewer than 4 percent of those finally accepted drop out.
But the MCAT has had one major failing in its otherwise brilliant performance: It has been unable to consistently predict personal and professional characteristics. As early as 1946, medical educators were trying to design the MCAT in a way that might tease out such information, but they, and those who followed, were unable to succeed.
Now the MCAT is about to undergo its fifth revision, the first in nearly 25 years. Last month, the Association of American Medical Colleges, the national organization that administers the MCAT, released the preliminary recommendations of a 22-member advisory committee that has been studying the issue for the last three years. They recommend, among other things, lengthening the four-and-a-half hour exam by 90 minutes and adding questions on disciplines like sociology and psychology. The new exam would also test analytical and reasoning skills in areas like ethics, philosophy and cross-cultural studies, which could include questions about how someone living in a particular demographic situation, for example, might perceive and interact with others.
Despite what some view as a long overdue re-examination of this linchpin of medical school admissions, many medical educators, including members of the advisory committee, remain cautious about tampering with a test that has proved successful so far.
"It's like trying to improve a Honda," said Dr. Ronald D. Franks, vice chairman of the committee and vice president of health sciences at the University of South Alabama College of Medicine in Mobile. "When you've got something that's working extremely well, you can make improvements, but you've got to be mindful of the services it has rendered."
Those services can be gargantuan. Jefferson Medical College in Philadelphia, for example, receives almost 10,000 applications each year and must whittle those numbers down to 800 for interviews for the 260 available slots in each class. In combination with grades, the MCAT can help admissions officers eliminate a quarter of the applications.
"But we and other medical schools have so many great applications from the standpoint of just numbers that we usually also need to go through other parts of the application as well," said Dr. Clara A. Callahan, dean of student affairs and admissions at Jefferson and the lead author of one of the largest longitudinal studies on the predictive validity of the MCAT. "You want to make sure someone isn't just saying that he or she wants to help people."
It's likely that Dr. Callahan and other medical school admissions officers will have to continue to look beyond the MCAT to learn more about their applicants' personal qualities. The science of personality testing has advanced tremendously over the last 25 years, but the committee felt it was still unclear how accurately a test could predict traits like integrity, altruism and the ability to collaborate. Some members were uncomfortable, too, with the long-term implications. "Will we end up labeling someone forever with a 9.2 for their personality?" Dr. Franks asked.
Only time will tell whether this newest version succeeds where earlier ones have not. But one thing is certain: Taking the MCAT is likely to remain a rite of passage for doctors-to-be for years to come.
"The reality is that we doctors are taking standardized tests – in-service exams, board exams, recertification exams – all our life," Dr. Callahan said. "It's something people have to master in medical school and beyond, so it's nice to be able to accurately predict at the outset how someone will do with them in medical school and beyond."
