A Conversation With Robert Wachter, MD: Medicine's Bumpy Ride Into the Digital Age

Interview by Peter Wehrwein

Robert Wachter, MD, is a professor and associate chair of the Department of Medicine at the University of California–San Francisco. A prolific writer and blogger, his most recent book is The Digital Doctor: Hope, Hype, and Harm at the Dawn of Medicine’s Computer Age. In April, Wachter topped Modern Healthcare’s list of the 50 most influential physician executives in the country.

You’re often given credit for hospitalists becoming part of health care.

When people say that, I say it sounds just like Al Gore invented the Internet. I coined the term, and I think people generally see me as the academic leader of the field.

In the mid-’90s in the managed care era, I had just been given a new job to run the medical service at UCSF, and my boss, a very thoughtful, strategic guy, said to me, “Come up with some new way of delivering care that’s better and less expensive and trains people better.” And I began sniffing around and found this new model emerging in early-adopter places, where you had a separate doctor in the hospital. Nobody had written about; it didn’t have any standardized name.

As I was out there talking about the benefits of the model, many people yelled at me and said, “This is terrible! You don’t get it!”

What, according to them, didn’t you get?

“You’re screwing up the doctor–patient relationship” or “My patients need me.” There were actually more complaints from doctors than from patients. I think patients accepted the notion that their primary care doctor was overwhelmingly busy and couldn’t be there for them all day long.

It was primary care doctors who said, “You’re saying we’re not good enough to take care of our own patients in the hospital.” And I said, “No, I’m not saying that. What I’m saying is that the organizational model for hospital care, which may have made sense in the days of Marcus Welby, doesn’t make sense anymore. And it’s not that these new doctors are going to be any smarter; it’s just that they’re going to have the professional focus on hospital care in a way that you don’t.”

Is there anything comparable between the hospitalist and the computerization of health care that we’re seeing now?

I think the IT transition is a much bigger issue because it’s much more ubiquitous and will ultimately have a much larger impact on the world.

But in other ways, they’re the same. These are big changes, and we always underestimate how hard they’re going to be, and we underestimate the pushback from people who, for various reasons, don’t want the change.

You mention several things that motivated you to write this book. What was the driving force?

I think the driving force was disappointment, and the disappointment was because I’ve been doing work in patient safety for 15 years. We’ve been waiting for this day—when computers would finally enter health care and fix everything. We’ve come to believe that if you computerize something, it makes it better, faster, slicker, and cheaper—and it’s all good.

And then one day about two years ago, I was sitting in a meeting at UCSF and we were discussing this case where we gave a kid a 39-fold overdose of an antibiotic. In a completely wired system. As I began to hear about the nature of the case, my jaw dropped, and I realized that something fundamental had changed in the way we practice medicine, the way we talk to each other, the way we trust automation.

Nobody had written about it before. When I looked around for literature, it was either technical or too futuristic and hype-y for my taste.

You write about the loss of communication in medicine—the electronic siloing. The few times I’ve been a patient, when I’ve had younger doctors, I felt like they were getting sucked into the computer. The older doctors, not so much. Do you see the current era of electronic medicine as interfering with communication in a lot of different ways?

Yeah, I don’t think there’s any question that it is. That doesn’t mean that we should pull out the plugs and go back to pen and paper. Some of it is just an extension of the rest of our lives. Is Facebook interfering with communication? Your kids used to learn to talk to you and their friends, and now social media is how they communicate. Is that a good thing or a bad thing? Who knows, but we’re not going to turn back the clock.

Computers are helpful but there are times when medicine is a human-to-human experience, and the machines can get in the way.

What struck me was the degree to which technology profoundly changed the way we communicate with each other—doctors to doctors, doctors to nurses, doctors to patients, patients to patients.

What we’re trying to do in health care these days is figure out how to deliver better, more satisfying, more patient-centric care—and do it at a cost that doesn’t bankrupt the country. Where are computers helping? Where are they not? When they are not, how do we mitigate that? Is it with different electronic tools? Is it a matter of making the ones we’ve already got better? Or is it something that’s pretty old school, where we’d say, “We’ve got to schedule meetings or informal gatherings that bring back person-to-person communication in an electronic environment.”

It seems like you would probably land on the last thing you said—more person-to-person communication in an electronic environment.

You know, it’s interesting you say that, because I try to be thoughtful and open and a bit agnostic about where I’m going to land.

But let’s take the radiology department as an example. There’s no question in my mind that the decrease in interactions between the frontline clinicians taking care of the patients and the radiologists compromises the care of the patients and compromises the education of both parties.

I would approach the question pretty agnostically about how we fix that in the most cost-effective way. And it might be that we recreate old-style radiology rounds. Or we might then say, “Really? We’re going to make all these busy teams schlep down to another floor and wait for the prior team to go through?”

So maybe it shouldn’t be exactly as it was. Maybe, we teleconference, along with some fancy new tool that allows the radiologist to point to a spot on the film that we’re all looking at on our computers.

I don’t know the right way to do it, but I think sometimes the solution needs to be a new or better form of technology.

Let’s talk about the government’s role in health care and this IT transition. You’re pretty hard on meaningful use.

Many people call it meaningless abuse!

You’re talking about phases 2 and 3, right? You think phase 1 was a good idea.

I think the first phase was a politically astute and necessary act that served its purpose quite well. The adoption curve for health IT was unbelievably slow until the federal money kicked in. You know, we were at 10% adoption in doctors’ offices and hospitals in 2008. We’re at 70% today.

If the economy had not imploded in 2008 and if there hadn’t been $700 billion in stimulus money, then there’s no way we would have found $30 billion for health IT adoption. We wouldn’t have found $3 billion. I’m not sure if we would have found $300 million.

So is the EHR the Great Recession’s WPA?

There is an analogy there. It’s a public good that would not have happened on its own.

But the analogy—it’s a little tenuous because I think we would have gotten there eventually. With the ACA, with the pressures now on providers—clinicians and hospitals and health care systems—to deliver high-quality, safe care at a lower cost: Those pressures would have ultimately led to computerization. But there’s no way in a million years they would have gotten us from 10 to 70% adoption in five years.

And the opposition?

The pushback—there are parts of it I agree with, there are parts that I don’t. One criticism is that the computer systems and the EHRs were not ready for prime time, so the federal government subsidized the adoption of mediocre systems. That doesn’t pass the sniff test because, I believe, the systems would have remained mediocre if nobody was using them. These systems only get better when you get to version 14.0.

But once the feds get deeply in the weeds in the world of technology, you have problems. Just think about if the feds were designing your iPhone. It can’t work. Not that they’re bad people. Even forgetting the politics of blue states and red states. It’s just not an appropriate role for the federal government. And that $30 billion put the government in a position where they created a pretty powerful regulatory apparatus. Of course, all the interest groups jumped in to have their say, which is a natural phenomenon.

During my research for the book, when I spoke to frontline doctors and nurses, or to CIOs or CMIOs trying to grapple with the federal regulations, or to vendors, they told me we are spending so much of our bandwidth now on just meeting the next meaningful use requirement. We could do better and more innovative stuff if these things got dialed back.

Would you junk the second and third phase?

I would declare victory and then pull back massively on it. Not even call it “meaningful use” anymore. There’s no more money to give out. And now you’re talking about penalizing doctors for not having the right computer system.

I think the world in 2015 looks very different from the world in 2008. We now have incentives on value, so if you are a health care delivery system or a doctor’s office, and if you deliver good, quality, satisfying care at the lowest cost, you will do well—and if you don’t, you won’t. With those incentives in place, I don’t think you want the federal government intervening very much.

There are two exceptions where I think there is a role for the federal government. It has to intervene on privacy and security because nobody has figured this out yet.

The other is interoperability. It is a public good if all these systems talk to one another. And the federal government is the only entity that can bash people’s heads together and say, “Folks, either you link these systems together yourselves, or we will make sure you do it.”

You end your book with the story of a patient in the ICU and turning off the machines. I think what you’re driving at there is that at a certain point, we want machines out of the picture.

Yes, I think that’s right. There are times when the machines are helpful and wonderful, and there are times when medicine is a fundamentally human-to-human enterprise, and the machines can get in our way. We have to be thoughtful about when those moments occur and address them.

Despite all the problems, you’re upbeat about the future of medicine, EHRs, and digital communication.

I interviewed nearly 100 people, and they each gave me very different views on a lot of stuff. But then I said, “What’s the end game here? After we figure out the policies and the politics and the new designs, what does this look like?” And they all pretty much painted the same picture, and it was pretty terrific.

You sort of go all Eric Topol there.

Where I differ from Eric is I don’t think he places the advances in enough context or addresses the challenges well enough. I wrote The Digital Doctor partly in reaction to his first book, which I thought was kind of over the top. There are probably a few people in La Jolla who have sensors in their underwear or who wear gizmos on their head when they sleep. That’s all really fun. It has very little to do with the day-to-day practice of medicine.

Do you think we’re at a point with health care IT where there’s all this friction, but like primary care physicians learning to accept hospitalists, people are recognizing that EHRs and computerization are going to work out?

I think the difference is that the complaining—it’s not just about losing something. It’s actually quite legitimate. I think what we’re hearing from many docs is that the systems are not very good, and they’re making it difficult for them to do their work. That doesn’t mean you turn back the clock. These systems just have to get better. I’m hoping the book makes a contribution to doing that.