Q&A: Yale’s Zack Cooper on rampant surprise billing and how it might be stopped

As legislatures across the country tackle the issue of unexpected or “surprise” medical bills, two researchers at Yale recently published a new report detailing the surprising frequency of these charges. The paper, published in The New England Journal of Medicine, used a private health insurance company’s data on 2.2 million emergency room visits. Even though patients visited an in-network hospital, nearly a quarter of them were treated by an out-of-network doctor, exposing them to potentially major costs.

The authors suggested hospitals should sell insurance companies a bundled ER package that includes all doctors’ services, thereby eliminating out-of-network charges for physician care.

The response from the American College of Emergency Physicians was swift, with the group’s president questioning the study’s data and arguing that the unexpected bills are not physicians’ fault, but the result of inadequate coverage from insurance plans. “Most emergency physicians prefer to be ‘in-network’ as long as insurance companies pay fairly,” the group’s president said in a statement.

We caught up with co-author Zack Cooper, an assistant professor of health policy and economics at Yale University, to discuss how he became interested in this issue, the study’s methodology, and his response to the industry group’s sharp criticism.

Q: What are the most important points of your recent research?

A: The first is just that this happens: You can go to an in-network hospital, see an out-of-network physician and get a really big bill. The second that it can happen with pretty alarming frequency, up to 22 percent of the time. Third, it can have pretty devastating financial consequences for individuals. These sorts of surprise bills can tally up into the hundreds or thousands of dollars and really wreak financial havoc on people’s lives.

Q. Tell us more about the geographic variation you found and what it means. Why is it important to look at this?

A: The fact that surprise billing is zero in certain areas means this is a solvable problem. We now need to step up the pace and figure out why and what we can do about it. That’s something we’re trying to figure out in future research.

Q: How did you decide to look into this? What’s next?

A: I read some of journalist Elisabeth Rosenthal’s work in The New York Times that highlighted some of these egregious cases where a person went to an emergency department and came out with massive bills. The question I was really curious about: Was this an isolated case she found or is this sufficiently common that we need to take notice? Once we found out that this was as common as it was, we decided to put out a quick article and document the frequency with which it occurs. Now, we’re trying to move forward and really figure out why. Hopefully, we’ll have some results out in June that begin to flesh out why it happens and different strategies to fix it.

Q: How do you figure out why this happens?

A: One way is talking to a lot of stakeholders: physicians, hospitals and insurers. Our paper generated a lot of chatter, and that’s helpful to generate hypotheses. The first thing is to come up with different theories that could explain it. Now, we’re doing machine learning, seeing what data can explain the pattern and testing various hypotheses.

Q: Tell us about the challenging process of obtaining data from private payers. Do such agreements ever pose a conflict of interest?

A: We have no idea what the world looks like for the privately insured because all of the publicly available data is from Medicare, but that’s a small part of the overall population. The challenge is getting data on the privately insured. Particularly for the research about pricing, if you identify the name of the firm, the firm could face financial consequences when the research is published. If I put their data out there, then their competitors are going to know contract rates and that puts them at a competitive disadvantage.

I went around to different insurance companies and said, ‘Hey, can I use your data?’ and a lot said no. I managed to get one of the large payers to give me their data, covering tens of millions of lives, and $7 billion in emergency room activity. The data usage agreements that Yale signs requires that the party giving me the data has no input on what I do. They don’t have a right to edit my research. That’s why you get a data use agreement.

Q: But if they don’t like it, they might not give you access next time.

A: That’s completely possible, but they have no input on your research. Universities put in so many checks to make sure it’s free of outside influence.

Q: When you look at the “surprise billing” story, two sides emerge: the doctors blame the insurance companies and the insurance companies blame the doctors. It seemed like your paper came down harder on the physician side. Why is that?

A: I don’t think any party has a monopoly on virtue or vice.

One of the things we looked at was the average compensation rates for ER physicians versus other specialties. In our data, an in-network emergency room physician was getting 297 percent of Medicare (reimbursement rates) compared to an average of 178.6 percent for an orthopedist doing a knee replacement or an internist getting 158 percent of Medicare for an average office visit. What that tells me is there is a higher mark-up for ER physicians. The reason they are going out of network is not because their in-network rates are low.

I also went to markets where our data was thick and thin, meaning in some areas my insurer has a lot of covered lives, and in other counties much less. If you think the story is a bargaining story — that the power of the insurer matters — then what you’d expect is that areas where I have lots of data, you’d see lower rates. But if the rates look identical in areas where I have a lot of people and not a lot of people, it tells you it’s not something on the insurance side. Given that, and the high rate of physician payments, this is probably a physician story.

Q: The backlash included accusations that the data wasn’t accurate — such as the $19,000 bill mentioned in the publication. What are your thoughts on this?

A: I went back and looked at that, and it’s legit. I posted the table that the journal didn’t publish for space and what you see is the mean and the median out-of-network physician charges look nearly the same so this isn’t a case of some outlier observation driving this. Another recent report from the Federal Trade Commission matches our data almost perfectly.

There’s a lot of stuff wrong with health care that’s really hard to solve and this is one of those things you can solve. I think it’s going to be a lot harder if we impugn data. I’d rather see insurers and doctors come forward and say, “Hey, this is a really big problem, patients are getting hurt. Let’s forget who is at fault and fix this.”

Q: Can you tell us more about the policy solution you propose?

A: There’s three existing flavors of reform out there.

One flavor is the insurer pays everything the doctor asks for. In other states, the state sets the reimbursement rate for out-of-network services as a percentage of the Medicare rate. Each of those are set by a regulator, not by competition, and they’re subject to lobbying. So in a state where doctors have a lot of power, they’re going to force the insurers to pay everything. And in states where insurers have more power, they’ll force the physicians to take a low rate. That’s not how I think prices should be set.

The other version is what happened in New York. If the doctor and insurer can’t reach a settlement, each puts forth one bill, and an arbitrator picks which one they’ll pay. That forces both sides) to make reasonable bids. That’s pretty clunky and I just don’t think the way to set emergency room prices is to constantly be arbitrating.

What we’ve argued is that hospitals should sell a bundle that basically includes physician services for emergency care so that when you go to in-network hospital, you know what you’re paying. And if you go to an out-of-network one, you know what you’re paying up front. It’s up to the hospital to decide how they contract with physicians, or hire them themselves.

Q: What’s happing on the legislative front in terms of surprise medical bills?

A: There are a lot of states who are working on it. The challenge is that states can only regulate fully insured plans, but most people get their insurance from firms that are self-insured. Most big companies pay the claims themselves and they’re paying an insurance company to administer the claim. The risk is not held by the insurance company; it’s held by the firm. That’s why, in my policy proposal, states can regulate (surprise bills) because they can regulate hospitals.

Q: Given the data challenges, how should journalists cover these topics?

A: The best way is for journalists to figure out what’s happening under the hood. I thought it was a great feature that Reed Abelson and Margot Sanger-Katz did where you could email your story about surprise billing. Academic research is like a 50,000-foot view on the world and the natural complement is the work of journalists, which can fill in the pictures. Once you see the pattern, you can zoom in to learn more. 

**