Don’t Be So Sure

By Aaron E. Carroll ’94
How medicine is like an LJST course at Amherst

Aaron E. Carroll ’94

When I was a student at Amherst, I sought out classes that relied on tests, not essays. I was always more comfortable with subjects that had answers, where you could be right or wrong. I especially enjoyed classes in computer science, although it wasn’t my major, because when you were done, there was no question whether you’d done well or poorly. Things worked or they didn’t.

I thought that was how medicine was going to be. In medical school I spent countless hours memorizing facts about the human body. I learned how cells worked, how organs functioned. After years of study, I was finally released into the hospital, ready to practice.

illustration of medical caduceus with speach bubbles from the 3 snakes

That’s when it all began to fall apart. I found that the way we treat disease is incredibly doctor-dependent. Some doctors are interventionists, ordering lots of tests and procedures. Others are minimalists, preferring to let the body heal itself. Yet for many illnesses, I learned, people had the same health outcomes no matter what we did.

What stunned me—though it will surprise no one who has ever encountered a medical professional—is how sure we always sounded about each treatment plan.

Some doctors know how to say, “I don’t know.” But their ranks are too small.

Doing medical research is hard, time-consuming and expensive. The number of things we haven’t studied dwarfs the number of things we have. Often, we have to take what data do exist and extrapolate them to cover wider areas. Experience guides us. But we rarely express this “best-guess mentality” to patients.

I questioned everything my attending physicians said. Why did we charge $150 to build foam sleep wedges for babies with gastroesophageal reflux? Why did we put coughing babies in croup tents? This did not endear me to my supervisors, and I soon realized I’d be better off helping to answer the questions myself. When I shifted to research, I learned that many treatments don’t work or have no evidence to support their use.

These days I spend most of my time talking about health policy, often health care reform. Again, we have few real answers; that prevents almost no one from speaking as if they are sure their opinions are correct.

This is a problem. Our system is an utter mess. I know people exist who believe the U.S. health care system is number one in all ways, and I hear from many of them daily, but on almost any metric you pick, we’re woefully behind. We are one of the few systems in the developed world that fails to provide universal access to its citizens. We are unbelievably expensive, in terms of both insurance premiums and out-of-pocket costs. More people go without needed care in the United States than in any comparable nation. And our quality, while excellent for certain conditions in certain hospitals, falls short when considered for the country as a whole. Our life expectancy and our mortality rates are surprisingly middling-to-poor when measured against those of comparable countries.

Ideas for reform are rarely rational or based on evidence. In the same breath, Americans will demonize Canada’s insurance system while lionizing Medicare. Single-payer health care is evidently American-as-apple-pie if you’re 65, but tyranny-and-the-end-of-freedom if you’re 64. Conversely, people will rail against for-profit companies in medicine while personally enjoying the many benefits of drugs, medical devices and private insurance.

Such surety in the face of so little data can lead one to despair.

In those moments, I think back to one of the few classes I took at Amherst that was essay-based, “Legal Institutions and Democratic Practice,” taught by Professor Lawrence Douglas. I swear, I can’t remember why I chose that course. It covered the history of the courts in the United States, highlighting important cases along the way and how legal decisions changed the way the country worked over time.

Often, I was sure I knew which side of a case was “right.” Yet many in the class believed the opposite. And Professor Douglas didn’t tell us the answer, because there was no “answer.” How do you proceed when you don’t know the right thing to do?

Years later, I decided the answer lay in the class itself. The point of discussing the cases in class was not to reach the truth. It was to teach us the importance of the arguing. We have to keep challenging, keep refining our thoughts. Over time, if we allow the argument to continue, we might learn what’s best.

This understanding has made me more humble about what I “know.” It’s led me to write a few books debunking medical myths. It’s led me to hold greater equipoise about what I might find in my research. Most importantly, it’s led me to bring up ideas in public forums, and to listen to what people say back.

I always wanted to be a test-taker, not a writer. But there’s no progress to be made in only repeating what we “know.” It’s not always fun to read the online comments about what I say, but it’s immensely rewarding to be part of the debate. As I learned many years ago, it’s so often the argument that matters.

Carroll  is an associate dean for research mentoring and a professor of pediatrics at the Indiana University School of Medicine. He is a regular contributor to The Upshot at The New York Times.

Monika Aichele illustration