In its early years, Major League Baseball discovered talent in what can only be described as a haphazard fashion, auditioning promising players in cornfields and hoping for the best as they took their place on big-city team rosters.
Then, about 100 years ago, everything changed. The St. Louis Cardinals, under Sam Breadon and Branch Rickey, created the farm system, which revolutionized baseball. It helped teams discover a wide swath of talent across America that wasn’t being previously captured. It gave teams the opportunity to evaluate players over time and assess them holistically. And it helped them create a deep bench as well as a reliable, data-driven way to develop promising young players.
In the process, the farm system helped take baseball from a genteel sport to the national pastime.
A hundred years after that process started in St. Louis and spread worldwide, we have the ability to use the same principles — data, a wider net, and careful talent development — to revolutionize a much more vital sector: medicine.
Changing the way we select who can enter and study in medical school will not only diversify medicine at a key time but also address a looming shortage of doctors to serve low-income rural and urban communities. The Association of American Medical Colleges this summer said the U.S. could see a shortage of as many as 139,000 physicians by 2033.
The change starts with giving prospective students new ways to show they have the skills to become quality physicians — figuring out how to give them a chance that they don’t always have now. A lot of these students are smart and highly educated. Some just tripped up in college, likely during freshman or sophomore year. Or they came from high schools where they didn’t have access to rigorous coursework, and started college behind. We must — and we can — figure out how to get them back on track and into medical school.
For prospective medical students, the all-important indicator of whether young people qualify for medical school right now is the 7.5-hour Medical College Admission Test. According to a 2017 Kaplan survey, more than half of medical school admissions officers say a low MCAT score is “the biggest application deal breaker.”
What’s worse, many students who do poorly on the test come from low-income communities, or communities of color, both of which have the greatest need for good medical professionals.
By contrast, let’s look at how professional baseball develops talent: It evaluates prospective Major League players over the long haul, not just by one 7.5-hour test, one at-bat or even one game. It nurtures them, not by loosening standards but by doing just the opposite: by subjecting them to data-rich coaching and direction to help improve specific parts of their game.
Part of the reason most medical schools don’t change their system is attachment to the status quo. Schools get thousands of applicants for perhaps 150 slots. They sort the names on an Excel spreadsheet to determine those with the best grades and test scores and select only the top scorers. They’re managing the situation with the data points they have. But grade-point averages and test scores are also imperfect metrics. Students with 4.0s may make great researchers but not such great physicians.
There’s a way to do things differently, using predictive modeling, among other techniques. It also allows schools to open medical schools worldwide, matching promising students with needy communities anywhere. For instance, talented Somali students can be matched with hospitals in Minneapolis, where decades of immigration by Somalis fleeing civil war have both revitalized parts of the city and increased the need for responsive medicine.
Such modeling and matching takes a lot more work, but if we want to solve this problem, we don’t get to take the easy solution.
I’m hopeful a new generation of medical schools can show the way to a new system that finds and nurtures young people with potential to be physicians, starting with analytics that identify potentially successful students (see the baseball scouting system). It’s a system that also acknowledges that some students may need more time to mature. A holistic evaluation system can better identify non-traditional but promising candidates.