What would it take for Harvard Med to drop off #1 in USNWR?

HMS has had a choke-hold on the #1 spot in the USNWR research medical school ranking since the rankings existed. However, most people are not aware of the ranking methodology and, therefore, unaware why HMS is ranked #1. They might see the ranking as a reflection of its reputation among doctors, their number of Nobel prize winners, the size of their hospitals, their MCAT average, their Step 1 average, whether their girlfriend’s grandmother has heard of it.

T and I have wondered and suspected for a while about HMS’ consistently high ranking. I want to answer the question, why is HMS always numero uno under modern USNWR research medical school ranking methodology? (USNWR changes its methods every once in a while just to shake things up, but I won’t go into those differences.) More specifically, is there a primary factor to which I can attribute their dominance?

To address this question, I got a hold of USNWR’s raw data and tried to re-create this year’s rankings. Unfortunately, the magazine is not specific as to how the data are normalized, so my attempt is only an approximation, and its usefulness is limited to analyzing large hypothetical changes. I took the raw data, normalized each measure by dividing by the maximum value of the measure among the top 20 schools (except I inverted this logic for acceptance rate), multiplied each by the weighting USNWR gives (13% to MCAT, 6% GPA, 1% acceptance rate, 15% total NIH funding, 15% per faculty NIH funding, 10% to faculty-student ratio, 20% peer rating, 20% residency rating), and then re-normalized by expressing the score as a percentage of HMS’ score (which USNWR does, too).

re-calc ranks

HMS is just so much higher rated, more than 10% above second place, it’s ridiculous.

I hypothesized that HMS is #1 because it has academic affiliations with so many big hospitals in an arrangement where all their physicians are Harvard faculty (including tons of ‘instructors’) and their money is gets counted as Harvard money in USNWR’s book (but not in Blue Ridge‘s book).

I looked up some figures on the faculty of HMS and the NIH funding granted to its hospitals.

In an alternate universe, let’s say HMS never moved from its location next to MGH in 1847. It never settled into the open farm and marshland of the Fenway on Longwood Avenue, it never spurred the creation of several new hospitals that now make up the Longwood Medical Area. Let’s say the entity known as HMS was only made up of Massachusetts General Hospital and the preclinical departments of HMS. Today they can tout having the #1 hospital in the country, a recent Nobel prize winner, a famous humanitarian, a two-time TIME 100 stem cell scientistcase reports that go straight into the New England Journal. You’re not missing anything (not even a little children’s hospital-within-a-hospital).

That means no Brigham, Boston Children’s, Beth Israel Deaconness, Dana Farber, MEEI, Joslin, MEEI, Schepens, CHA, McLean, Pilgrim, Hebrew SeniorLife, Spaulding, Forsyth (what???). Say HMS got rid of the faculty affiliations of physicians who worked only at those hospitals and could could not lay claim to their grants. That’s okay; you still have a reasonable top medical school.

HMS+MGH has “only” $548.4 million in NIH grants in 2012, and “only” 3332 full-time faculty. Under this scenario, I’ll adjust the faculty:student ratio and grant figures in my ranking re-calculation accordingly and assume factors like reputation rating and student selectivity don’t change. With this grave hypothetical handicap, HMS is no longer #1!

re-calc MGH

Thus, the perennial edge of HMS could be attributed to its promiscuous affiliations with huge research hospitals. HMS is ranked #1 because it has the benefit of holding its crimson umbrella over 9000 grant-hungry physicians and biomedical scientists spread across several of Man’s Greatest Hospitals.

Note that HMS certainly doesn’t pay all those docs. Despite its large endowment, it can’t afford to. The hospitals are all independently owned (e.g. by “Partners”). Word on the street is that Harvard and its affiliates actually pay significantly worse than many peer institutions (because it’s Harvard, because it’s in Boston).

Note that HMS certainly can’t handle 9000 docs clogging up its promotion ladders. In fact, about half of the clinical faculty that contribute to Harvard’s enormous numbers are “instructors,” an entry-level position that seems neither tenure track nor non-tenure track, an amorphous basic starter pack position that few places use to the extent that Harvard does, a directionless space that Harvard could keep a junior faculty at until s/he were 40 if it wanted to.

The effect of Harvard’s vast resources on students is smaller than you might assume. Yes, you (hypothetical HMS student) do get more selection when deciding where to do research. But eventually you pick a lab or mentor and your work in that one lab is then your world and future success; the hundreds of other labs you could have chosen are irrelevant. Yes, you get to have more tutorials than lecture. But the jury is out whether this has a meaningful impact on a student’s learning compared to traditional methods of instruction. Yes, you could get exposure to lots of top hospital environments. But since 2008, your clerkship experiences will only be at one hospital.

Yes, the school looks like it has a lot of money on paper, but actually it doesn’t. The hospitals that are responsible for HMS’ ranking are independently owned and operated and do not have a standing agreement to feed HMS any money. HMS largely relies on grants and endowment funds to operate its research and educational mission. At other schools, clinical activities at the hospital contribute significantly to the school’s operating revenue (accounting for half at my school!). When economic times get rough, like in 2009, the endowment takes a big hit, and the dean of HMS has to call in favors and ask hospitals to chip in a couple tens of millions to keep the students well-fed (with knowledge and occasional free food). This means you’re not overflowing with cash to guarantee full funding for student summer research, to fund student travel to conferences, to throw lavish formals, to give merit scholarships.

My final impressions (TL;DR): HMS is rich, but only about as rich as other top 10 schools are. HMS is clearly a tip-top school, but not as far and away the best as USNWR might make it seem.


One thought on “What would it take for Harvard Med to drop off #1 in USNWR?

  1. I just want to say that I stumbled across this blog from sdn and I’ve spent every extra minute of the last two days reading! I really enjoy both of your blogging styles! You guys provide insightful stat-based information about residency matching and moving emotional pieces about Long Distance relationships during medical school.
    Thanks for all the amazing posts! Keep up all the good work!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s