Medical Statistics

Register now

One of my neighbors recently solicited my opinion of her high school senior daughter's college application choices. The girl is a top student, looking to take a pre-med curriculum in college. Knowing I was an alum of Johns Hopkins, she asked my thoughts on four prestigious schools for pre-med concentration: Notre Dame, Northwestern, Washington University in St. Louis and Hopkins.

I think my neighbor erroneously thought I'd been involved with the sciences at JHU. I decided not to tell her I was never closer than 100 yards to an organic chemistry lab in my undergrad days.

My neighbor's question is one that many parents ask: What college will give my son/daughter the best chance to gain acceptance to medical school? Colleges are certainly not reluctant to tout their successes with medical school placement. Indeed, most top private schools claim placement rates far in excess of the national average of 45 percent. And there's no shortage of dialog – i.e opinion – on the topic from the college admissions website College Confidential, Yahoo!Answers and Facebook, among other sources. So I decided to do a little research to see if I could offer some “performance measurement” insight to my neighbor.

The noise-to-signal ratio on the pre-med Internet chat threads is high, not surprisingly. It seems there're plenty of strong opinions on which schools are better pre-med, but little evidence to support them. Some naively question whether acceptance figures reflect those who start in pre-med, rather than those who finish. Others obsess on acceptance to only the most elite med schools. And still others argue it's less about the school and more about the student.

Some of the confusion comes from the schools themselves as they market their pre-med prowess. One small liberal arts college touts a 95 percent acceptance rate based on a single year's data from 20 applicants. But there's a lot more year-to-year variation than many schools acknowledge. I'd need to see at least five years of data for the small college to be convinced. Other programs combine allopathic and osteopathic school admits, inflating the acceptance rate. Elite liberal arts college Amherst presents a pre-med outcome table I can't understand, though the text following explains several flavors of acceptance rates, each very impressive.  

An interesting page on the Hopkins pre-professional advising site elucidates the complexity of  acceptance measurement. For 2008, Hopkins reports an overall 63 percent admission rate to medical school based on 223 acceptances for 352 applicants – seemingly solid but initially not especially noteworthy. Of the 352 applicants, 295 earned the recommendation of a faculty committee based on academic credentials. The 295 students are subdivided into 238 first-time applicants and 57 re-applicants – those who've previously applied unsuccessfully. The first-time applicants are further divided into 111 post-junior year and 127 alumni applicants.

Historically, the acceptance rate for committee-recommended first-time junior applicants > rate for committee-recommended first-time alumni >> rate for re-applicants.  Hopkins reports that 88 percent of first-time applicants in 2003-08 with a gpa >= 3.3 were accepted to at least one allopathic medical school. So JHU, like many pre-med powerhouses, generally uses committee-recommended, first-time applicants as the basis for its acceptance bragging. And, of course, committee recommendation is a proxy for outstanding academic performance. The more select the applicant pool, the smaller the denominator – and, for a given numerator, the higher the acceptance percentage.

For me, the most telling numbers of all are the detailed acceptance figures by college grade point average (gpa) and medical school admissions  test  scores (MCAT) tallied by the Association of American Medical Colleges. The relationship between grades, MCAT scores and med school acceptance is very strong, as can be discerned reading the table from lower left to top right. Applicants in 2007-09 whose gpa was between 3.4 and 3.59 with MCAT scores between 27 and 29, had a 38.6 percent chance of gaining admission to medical school. Those in the 3.6-3.79 gpa and 30-32 MCAT cell, by contrast, gained admission at the rate of 73.2 percent – a stark difference. The upper-right quadrant is clearly the holy grail for med school aspirants. Straight-shooting Emory University confirms the strength of this correlation, reporting an overall 56 percent acceptance rate in 2010 that climbs to 67 percent for applicants with a >= 3.3 gpa and 89 percent for students with a gpa >= 3.60 and MCAT scores >= 30.

Rather than aggregate medical school acceptance rates which are highly correlated with innate student abilities that differ markedly by school, a better measure of pre-med performance might be how programs fare with their acceptance percentages in the individual cells of the AAMC table. Schools whose percentages exceed the comparable AMMC figures can make a reasonable argument that they're adding value to their students.

I propose that pre-med programs report their annual acceptances/rejections in the buckets formed by the intersection of MCAT and gpa categories consistent with AAMC report. Schools can provide the cuts as granular as they'd like – first-time applicants/committee recommended, re-applicants/not recommended, etc. – as long as, in total, all applicants are included so the entire experience can be aggregated. Pre-med prospects looking to compare school performance would then have all the performance pieces needed to make their assessments.

Based on my admittedly limited research, I cannot reject the hypothesis that it's the capability of the student, rather than the pre-med address, that's far and away the most important factor in determining admission to medical school. The main reason top colleges have gaudy med school placement rates is that they attract outstanding students who, predictably, are successful in their pre-med studies – gravitating to the upper right quadrant of the AAMC table.

Billionaire Harvard dropouts Bill Gates and Mark Zuckerberg are business marvels because of their individual talents, not their abbreviated Harvard educations. If a high schooler has the talent/capability to gain acceptance to a prestigious private college, she certainly has the talent/capability to assure herself an 80 percent or more chance of getting accepted to medical school, regardless of where she ultimately goes for undergrad. If there are indeed school differences, they're likely small. Much more significant is the difference in capabilities between students at elite and less elite schools. Student ability trumps school cachet.

So I told my neighbor if her daughter gets accepted at one of those prestigious schools and the family can afford to send her, by all means take the opportunity. If, on the other hand, shelling out $60,000/year would be a financial burden, the talented daughter can still guarantee herself 80 percent-plus acceptance odds by performing up to her capabilities at the much-more-affordable University of Illinois.

For reprint and licensing requests for this article, click here.