• Perspectives on Safety
  • Published November 2012

In Conversation With… Abraham Verghese, MD


Editor's note: Abraham Verghese, MD, is Professor for the Theory and Practice of Medicine at the Stanford University School of Medicine and Senior Associate Chair of the Department of Internal Medicine. He has written 3 bestselling books, including Cutting for Stone, which has been on the New York Times bestseller list for more than 2 years. He has written about the importance of humanism in medicine and has become a passionate advocate for the primacy of the physical examination.

Dr. Robert Wachter, Editor, AHRQ WebM&M: Tell us a little bit about what got you interested in the demise of the physical exam.

Dr. Abraham Verghese: Where I trained, the physical exam was a very important part of the training. You had a high-stakes exam at the end of medical school and it didn't matter how well you did on the cognitive portion, you had to appear before the examiners and examine a long case and a short case. There was a premium put on being able to feel the spleen, use the right technique, and sort out rheumatic valve disease. What surprised me when I first moved to the United States in 1980 was how unimportant much of this was here. Even though many of the attendings that I trained with were quite capable of this stuff, it certainly wasn't high up on their list. I always thought the physical exam was quite fascinating. I thought it was the fun part of medicine and easy pickings. It took a while for tests to come back and very often you could tell volumes with a good physical exam, especially in neurology. The CT scan cannot tell you the patient's functional status. You have to figure out that this is not moving and that's not working and then correlate that with the CT scan.

RW: Did you have any theories at the time about why there was such a difference between the approach in the United States and the approach elsewhere?

AV: Even as far back as 1980, I think it was already clear that in the absence of a high-stakes exam like the one I took at the end of medical school, there simply wasn't going to be a premium put on physical diagnosis. In the US, my board exam consisted entirely of multiple choice questions, with no assessment of my physical examination skills. And, to arrive in America was to have this wonderful experience of actually using all the technology that I had read about in Harrison's textbook. But, the tradeoff seemed to be that the availability of this technology (things like angiograms and echocardiograms) came with a drop-off in physical diagnosis skills—people didn't think it mattered. But, I'm convinced it does.

RW: As you've seen this evolve now over 30 years, it sounds like you think it's become even more exaggerated. Is that your sense, and what are the forces driving the demise of the physical exam even more now than what once existed?

AV: There may have been a sort of agnosticism when I first arrived in the 1980s, but more and more I think it's almost as though there's a big void there. One populates the physical diagnosis part of the medical record by filling in dropdown boxes, and no one actually does the stuff that they claim to do. Or nothing that's said there has actually truly been done on the patient, I feel, more and more. It just is a pro forma thing. So over time physical diagnosis skills have really dropped off because the generation of attending physicians who valued, celebrated, and taught that no longer are teaching that. The folks who are now teaching never saw it valued, perhaps never had occasion to use it greatly.

We're studying physical diagnosis errors—errors that have had consequences for patients in terms of unnecessary testing, unnecessary surgery, or some other misadventure because a fairly obvious physical finding was overlooked. It's the sort of study that you can only do by collecting anecdotes. We have hundreds of anecdotes, and it's clear to me that this stuff is consequential. The study is showing me something that I've always supposed to be true, which is that if you don't do this stuff right and if you misstep—someone has neurofibroma and it's all over their skin anywhere and you don't happen to feel any other skin but the belly and you decided that this is an intra-abdominal mass and then send the patient for a CT scan of their chest and abdomen—you've done a real disservice to the patient who didn't need any of that because they have a very obvious diagnosis of neurofibromatosis. That type of error is rampant and I would guess is becoming more so, although it's hard to prove that.

RW: Is your main concern that it has safety consequences or that it has cost consequences, or is it something about the interpersonal aspect that bothers you the most?

AV: I think they're all important. Most important to me is the issue of safety and basically the disappointing thought that we might actually be worse off than physicians 100 years ago who had none of the technology we have. We might be worse off in our ability to spot and diagnose simple things and use technology to answer a second level of questioning or get a better read on something we already know. That disappoints me, and I think it has both cost and patient safety consequences.

But, lately I've also become more and more cognizant that examining patients is just about the only way we as physicians make contact with patients. If you view medicine as largely a cognitive exercise of getting certain numbers, figures, and facts and plugging them into algorithms—none of this matters. But, if we still view medicine as art and science, much of the art is in listening, in finding the nuances in the story, and deciding what they mean. Examination of the patient represents an important time, at least from the patient's point of view, where something is transpiring between patient and physician, and not to make too much of it, but it's an important event in the patient–physician encounter. I sense from many anecdotes that patients find this very valuable even if we think it's not that important. I remember one time at the VA I was demonstrating an exam on an old veteran. He had a pleural effusion. We were percussing it and this man hadn't said very much during the whole exam. He suddenly piped up as I was percussing him. He said, "My physician used to do that when I was a young boy. He was a good physician." In his mind he had equated the two things. I think that the laying on of hands, the handling of the body, has an important function for the patient. It's a way of saying, I'm taking all your symptoms and validating them on your body. I'm giving them weight. I'm localizing your story on your body, and then I'll get the images and do the other things. It's an important ritual and more and more I'm sensing that it's a ritual we have given up, and we've done so with some consequences. We give it up at some peril.

RW: I guess no one argues against the physical exam, but the main pushbacks that you hear are first that the test characteristics of the physical exam wouldn't stand up in a scientific validation study, and the second is it just takes too much time. Can you address both of those?

AV: I think it's fair to say that many aspects of the physical exam, when looked at, many signs and maneuvers simply don't hold up. But many of them do, even very simple ones. Decreased breath zones on one side definitely means something. Some signs clearly have very good operating characteristics, which suggest that we should do them and make the necessary inferences. I actually think that by not doing the physical exam we often lose time. There are many anecdotes where someone comes in with terrible chest pain and it creates this whole flurry of activity—leads are hooked up, troponins are drawn, there's worry about dissection, and an ED physician finally comes in and says, "Well, let's take the guy's shirt completely off," and sees the rash of shingles. I often worry that, in our effort to save time, we don't see the patient and we order a test. We get a call from the nurse that the patient is complaining of such and such, and we go to see them but in our mind we've already decided we're going to get a CT scan of the belly. Examining them takes just a few minutes, might spare them the test, and might actually allow you to order a better test or ask a better question of the test. So I think that's an unfair argument that the physical exam wastes time. I would be the first to say that the routine annual physical, for example, is a low-yield procedure. But, for someone coming in with a very specific symptom and being hospitalized, the physical exam is a fairly high-yield procedure in combination with a good history and allows you to be further ahead than someone who doesn't bother with it.

RW: Have we entered a kind of death spiral phase where nobody knows how to either do a high-quality exam or teach it anymore, and therefore it really is not valuable in their hands?

AV: Well, it's interesting that you should say that because I think that the physical exam has many parallels with other rituals and crafts that are handed down from generation to generation. Typically, if they are not valued they will die down. If you look at a ritual like, I don't know, Indonesian sword crafting, there are very elaborate ceremonial swords. You need a relationship with a master craftsman. You need a long apprenticeship. There's specialized terminology. There's a specialized workspace. There's specialized clothing. Every aspect of the physical exam can be compared to these crafts. One of the truisms about such crafts is that if they are not taught or valued, they will in fact die out. I'm hoping that we don't get that far. Out of embarrassment I'm hoping that things will reverse themselves. In fact, I'm looking to the American Board of Internal Medicine to turn the ship around—to make it the law of the land, so to speak, that anybody with an MD behind their name has a pretty good chance of finding the spleen if the spleen is what's causing the person to feel full after meals. Yes, they can confirm it with a CT scan and an ultrasound, but there are basic skills that I think the public expects of us and we should expect of ourselves.

RW: You've taken it upon yourself to promote this agenda within your own institution and to some extent nationally. How has it gone? Has the reaction to that surprised you in any ways?

AV: Actually one of the big surprises is that I thought I would get a lot of pushback from people saying this is unnecessary, we don't need this. Or from my own house staff saying they have better things to do. I get some of that from time to time, but overwhelmingly young physicians recognize that having these sets of skills that allow them at the bedside to make a very sophisticated neurological diagnosis and the MRI will take another 4 hours if it comes at all that day, is a good skillset to have. So we find that our house staff are actually eager and hungry to do this. When we have our training sessions, we call them the "Stanford Medicine 25" sessions, we teach people how to do a set of skills, for example to do all the reflexes. It turns out that in theory most everyone knows what a hammer does and knows generally where you need to hit it and what's supposed to happen. But, in these sessions people begin to really appreciate how important technique is. That it's not enough to abstractly know you've hit the patella and this is supposed to happen. How do you position the leg, particularly in a bedridden patient? How do you best enhance your ability to get a result that you can rely on? I'm finding that young physicians recognize that this aspect of their skill, their technique, is something that they clearly want to have and they're investing in it. In fact, the Stanford Medicine 25 has been taken over by a generation of young hospitalists here who are working closely with me, and they value it and are bringing to it their own attributes. For example, John Kugler has me now carrying a portable ultrasound around. The other day I examined a lady with a carcinoma who had shortness of breath. She had big neck veins, and one of our worries was did she have a pericardial effusion. I had the ultrasound right with me and I could see that she had a big effusion, both pleural and pericardial. I didn't have the sophistication to say she had cardiac tamponade; it took the cardiologists and the big echo machine to do that. But, it's clearly not a one-way street where we are showing them only old-fashioned methods. As we come to the bedside more, we will in a sense come together from both ends. It will be us demonstrating simple physical diagnosis techniques of old, but it will also be technology coming to the bedside and the two melding together in a good way.

RW: That's an interesting idea. In some ways I think people simplistically dichotomize the laying on of hands versus technology. It sounds like you believe that these things will mesh together into something that's different from your father's physical exam but ultimately achieves some of the same goals that you've been articulating.

AV: Absolutely. Very often people who don't know my work well or don't know me presume that I'm some sort of a Luddite who is holding out for the old ways—and I'm not a Luddite. I work at one of the most sophisticated medical centers in the world and I love technology. But the reality is that we simply cannot be ordering tests willy-nilly the way we do right now. At some point, we're going to be held accountable, and costs are going to be important. Alan Garber, my colleague who's now provost at Harvard, described American health care as having a menu without prices. You know, it's filet mignon every night if you like. I think that's going to change. We're going to have to ration some of the tests we order and justify them. And what better way to justify them than to have some skill at the bedside where you can take a history and a good physical and say, "This is a high-yield test that I want to get." Not, "What the hell is going on, let's get everything." The former approach, I'm hoping, will be rewarded by changes coming down the pike.

RW: You've written a lot about the role of the computer more generally in changing the doctor–patient relationship. Can you describe your thinking about that and how that dovetails with the physical exam? Is it a parallel issue in some ways or is it fundamentally different?

AV: It's a slightly different issue but it's clearly related. The computer is obviously a great advance. Medical record keeping prior to it involved hunting down old dusty volumes and hoping that one wasn't missing or locked in somebody's office. The fact that you can now have it all in one screen is just amazing. It's just wonderful. I'd never go back. Because of our documentation requirements, because of the great focus on quality, appropriately so, we are spending more and more time in front of the computer screen and not enough time dealing with the patient. We have to find ways to minimize that computer time, which is very hard to do since that's become the nexus in which all the work is transacted, but the great danger is that the patient in the bed can seem almost secondary. The patient in the bed can seem almost as an icon for the "real" patient who's in the computer—who I have called "the iPatient." Clearly patients are left wondering who's in charge, who's coming to see me, who are all these people? There is no sense on the patients' part that they are at the center of all these activities. We have to find a way, I don't quite know how, to bridge that divide. Part of it is by taking the computer into the room, by making the computer secondary. Maybe even by hiring people whose only job is to be the gatekeeper to the information portal in some way. Maybe by having voice-activated devices that quickly transcribe all our bedside comments for us, so we're not sitting there in a very ineffective way writing out things that were written out the day before and cutting and pasting stuff that nobody reads, all for the purpose of documentation and quality, but at the expense of patient satisfaction.

RW: The demise of the physical exam—the embrace of more testing to get at some of the same information, and the fact that people are spending more time at the computer terminal and less time with the patient—might it be that clinicians don't either enjoy or prize their time with patients?

AV: I think clinicians still enjoy their time with patients. If anything clinicians resent the time that they have to spend in front of the computer. The computer is not something that is drawing them out of joy. This is not like my teenage son and video games. It's only because that's where the information is. That's where we have to sit and put our notes. We're there somewhat reluctantly, imprisoned usually in a room with seven other people also staring at monitors. I think that we have to find a way to bring back the social interactions between physicians, and between physicians and patients, that somehow got lost in this conversion to electronic medical records, again without losing the best part of the record. Somehow restoring the collegiality of rounds, restoring the joy of running into colleagues in the workstations, and having informal discussions. We miss a lot of that because so much of the consultation is happening online so to speak, not face to face.

RW: You've made the point several times as we've spoken about anecdotes. How important are the stories and how important is it going to be to have cold hard data, for example, that the time, energy, and money you're spending teaching people physical exam pays off in fewer tests, more satisfied patients, or fewer errors?

AV: Data is important. I would be the first to say that. I don't think I'm setting data up as being in opposition to everything else. There's an interesting guy by the name of David Orr at Oberlin who's written a lot about the contrasts between what he calls "slow knowledge" and "fast knowledge." Fast knowledge implies that more data is better than less data. It doesn't matter if the data is filtered. It doesn't matter if we don't know what it all means, the more the merrier. Whereas slow knowledge celebrates ritual, celebrates the acquisition of knowledge that's more directed. We're at this point in medicine where that debate between so-called wisdom and knowledge is something that we need to engage in more. People come to us, after all, they don't come to us as data points. They come to us with unique stories. In fact, we take a history from the patient. The word history has the word story embedded in it. Very much the art of medicine is listening to that story, extracting the medical elements from it that allow us to make a diagnosis, the temporal elements. But more than that, the story represents who this patient is. We all create these elaborate fictions, both conscious and subconscious, that we try to sell ourselves and others as who we are and this is the real story of our life. Sometimes that is who we are and sometimes it's completely different from what we, as physicians, perceive to be the real case. To see medicine as all data is to be very simplistic. You and I have just finished stints attending and it was so much more than data. It was so much more about people, emotions, and stories and trying to figure out where people were in the arc of stories. I think we'll always need both: we'll need anecdote, we'll need data, and we'll need to continue to examine both and get better at what we do.

RW: In some ways you're swimming against a very rapid stream. Are you an optimist or a pessimist when you think about where all of this is going to go in the next 10 or 15 years?

AV: That's been the story of my life, swimming against the stream. Although I actually see the pendulum swinging back in my direction in the sense that we have had the luxury of pretending that we didn't need to listen to the patient or examine the patient and the luxury of ordering tests that are hugely wasteful. If you really sat and examined how many of these tests and angiograms and CT scans were really needed, many of them were clearly unnecessary. Not to mention all of the vested monetary interests that drive all the kinds of testing we do. I think the pendulum is swinging back. Because if we in fact are going to insure more and more people and if we're going to be able to afford to care for them, we will have to be more thoughtful about the ways that we diagnose and treat diseases. I think that plays right into my hands. It involves listening more carefully, examining more carefully, and choosing tests more judiciously. So, I feel like I'm not swimming against the tide as much as I have been. Although it's not a bad place to be: Interesting things float by you.

Previous interviews can be heard by subscribing to the Podcast

Subscribe to the Perspectives on Safety Podcast now

View Related Perspective

Back to Top