It was recently reported that medical errors are the third-eading cause of death in the United States after heart disease and cancer. The estimated number per year is over 400,000. This is not new information, and unfortunately, this will not be a simple problem to fix.

A portion of these deaths result from health care providers lacking resources and being overextended, both contributing to making errors. Some will also result from systems issues such as handoffs between providers during shift changes or call. But some simply result from bad health care decision-making and even worse, negligence and indifference on the part of providers. Let’s say this portion accounts for 10 percent of these deaths. That would mean 40,000 deaths per year.

Does anyone find that acceptable?

The truth is that there are bad health care providers out there. And we health care professionals do a poor job of policing ourselves. How many of you remember CNN’s report last year on the pediatric heart surgeon in Florida who had a mortality rate 3 times the national average? The general public was appalled. But some of my surgical colleagues blamed CNN. Their refrain was, “That reporter ruined that guy’s life! He will never be able to get a job. He’s going to go broke!”

This sort of irrational solidarity, particularly among physicians, is rooted in a fear characterized by the idea that “That could’ve been me.”

It is a fact that most physicians are competent, caring and conscientious.  The majority of negligence is a result of a subgroup of bad apples. And most of us know who they are. Unfortunately so do the medical schools and residency programs that train these bad apples.

So rather than leaving the messy, expensive and time-consuming process of removing a bad apple to hospitals and medical licensing boards, we really should be thinking about how to re-direct these people before they enter, and especially leave medical school.

In order to do this, medical schools and residency programs will need to develop metrics that predict who will be (or is) a conscientious, caring and effective physician as opposed to someone who scores really high on standardized exams, has a straight 4.0 grade point average, and not a single shred of empathy in their body.

These new metrics will need to be accurate and their results reproducible across institutions. Don’t think this approach can work? Au contraire, it has been successfully in employed in another high-performance field where lives routinely hang in the balance: selecting paratroopers for the Israeli Army.

One way to persuade medical schools and residency programs to get on board with this idea is hold them accountable for the performance of their trainees after they leave the nest. Currently, residency programs are accredited based on their graduates’ pass/fail rate on board certification exams. After that, their accountability mostly ends.

What if they were instead held accountable for their graduates’ clinical outcomes over a 10-year period?  Ten years is a reasonable number because most physicians have to recertify in their field through a written exam at year ten. By that point, they will have matured into self-sustaining learners, and the bad apples will largely have been weeded out.

Medical schools will also need to be held accountable for the performance of their graduates. Failure to meet a minimum standard for outcomes in a particular field would lead to probation and possible loss of accreditation.

With this kind of pressure, medical schools and residencies will quickly develop the necessary metrics that are predictive of high quality, compassionate clinical care. Shifting to this approach will not change things overnight. However, through knowledge sharing and demonstrating the reproducibility of these metrics across institutions, a profile of the optimal physician candidate will evolve.

Will it be perfect? No! But it will be a vast improvement over what we do now. It will also go a long way towards rooting out at least one cause of those 400,000 deaths every year.

Peter Nichol is a surgeon.  This article originally appeared in the Medaware Systems Inc. blog.

Image credit: Shutterstock.com