Russ Roberts

Eric Topol on the Power of Patients in a Digital World

EconTalk Episode with Eric Topol
Hosted by Russ Roberts
Continuing Education... Michae... Continuing Education... Eric T...

We're in the middle of a healthcare revolution but it's about more than marvelous life-saving and life-enhancing apps on our smartphone. Eric Topol of the Scripps Translational Science Institute and author of The Patient Will See You Now argues that the digital revolution will give us more control of our health information and data. More powerful patients will transform the doctor-patient interaction. Topol talks with EconTalk host Russ Roberts about his new book giving us a glimpse of the changes coming to medicine from the digital revolution.

Size:30.8 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast episode

Related Readings
About this week's guest: About ideas and people mentioned in this podcast episode:


Podcast Episode Highlights
0:33Intro. [Recording date: May 1, 2015.] Russ: Now, your book is about the potential 'democratization of medicine', as you call it. What do you mean by the democratization of medicine? What might be changing in the doctor-patient relationship? Guest: Uh, totally changing. That is, it's a flip. It's a great inversion of medicine. I mean, democratization is available to all. But this is about getting each individual's information to himself or herself. It's about having the same medical capabilities, anywhere there's a mobile signal, anywhere in the world. So it is democratization at many different levels. But it's really putting each individual in charge of their health care. Russ: What has changed so far? And we're going to talk a lot about what is in the process of changing. But, you point to the flow of data and information itself. What is changing? What has changed? And why is that so important? Guest: Well, there are several moving parts right now. So, it's actually pretty extraordinary. But the smartphone is center to all of this. And that is that it's now capable through a smartphone to generate high-grade medical data. And it includes lots of sensor, vital-sign data. But it--virtually any physiologic metric of a human being. And add to that doing all your labs through your smart phone, and doing a good portion of the physical exam through the smartphone. And no less ability to summon a doctor--through your phone, even to come to your house. Or wherever you are. So that is, I think the principle thing is this mobile medicine, remarkably high computing power. And then supported by a digital infrastructure with cloud supercomputing and pervasive connectivity. So, all these things coming together, this technological quantum leap, is what's really the underpinning of the change in medicine. Russ: And you really focus on, and hammer on, this idea that in the current world, up until very recently, the medical establishment had my data and often wouldn't share it with me. And this technological innovation is potentially going to reverse that. Talk about what has, in the past, been the doctor-patient relationship, with respect to information, how that can change. Guest: Yeah, it's a really important issue, Russ. I did a lot of research, because I didn't know where this all came from. I said, 'Why do doctors and hospitals own your data?' And in 50 states, in the United States, 49 of them, except for New Hampshire, hospitals or doctors own your data. I said, 'Well where does that come from?' So, it basically, I traced it, did a lot of research. It really is due to paternalism. And it goes back of course to even before Hippocrates, well, 400 B.C. and long before that. So, doctors have thought that patients would not be able to handle having their own data; and therefore they should own it--doctors, hospitals, health systems. And that's a completely wrong model, because today you can generate terabytes of data on your own, which is going to markedly override whatever little data there is in your electronic record. So, ability for people to generate their own data is driving force for why the system that we have today, where people pay for medical services and don't even have any records; and they have to beg to get records, or pay to get their records--that is not going to be sustainable in the future.
4:55Russ: I want to focus on the paternalism a little bit. We've talked on this program about end-of-life issues and the issue of withholding information from a patient for, "the patient's own good." And you talk about how really bizarre that is. And how, in your experience, and many people's experience--relatives, loved ones, or yourself--weren't told information about yourself that the doctor had, because they thought they knew better than you did about what to do with it. Guest: Yes. Exactly. And I think this is something that not all people would want to give that up. I mean, some like to be suppressed and controlled and have full doctor autonomy. But as it turns out, every consumer survey that's been done in recent times has suggested that 80%, even up to 90% of people want their data, and want to have a lot more say in how their health care and medicine moves forward. And they feel fully capable of having their notes--which, say, 70% can't get today--it's amazing, get their office notes-- Russ: You are talking about stuff that the doctor jots down while you are talking about your problems. Guest: Yeah. But today, of course, mostly it's keyboards and not jotting, but rather typing. Russ: Yep. Clicking. Guest: And not even making eye contact. That, almost, more than 2/3rds of doctors still today in the United States will unwilling to give those notes to the patients. Because they feel that that--for lots of reasons. One is that they created the notes, for which there are now another medical, legal issues. Another is, well there may be some things in the note that the patient would be offended by, or get anxious about. But there have been studies, Russ, and particularly a big one called Open Notes--they completely overturned these pre-conceived notions about how patients couldn't handle having their notes. And in fact that whole project is going on with our notes, because all the doctors and patients love having the patients getting their notes. And that's one of the many pieces of this change, this democratization. It's an unstoppable, inevitable, and very profound change in medicine. Russ: Well, as an economist I look at that note-ownership issue through an economist's eyes. And I see that the doctor likes feeling important. When you keep the notes, it gives you some power. It protects you potentially from legal issues. It protects you challenges--just an unpleasant social interaction, like: 'Why did you write that down about me?' Even if it's accurate, of course, there are things that will make a patient uncomfortable and they don't necessarily want to have that be written down. But it is my body. And you'd think people would be open to that possibility. But it's a cultural change that's very hard for doctors. As you point out. Guest: Yeah. And the other thing is, patients should be editing their notes. Okay? This is a partnership. And there's a lot of mistakes, as it turns out, in notes--whether it's medication or factual things that are just wrong. And so, when you do get to your notes and your records, you start to realize: 'Oh, my gosh, there's all these things that are wrong in here.' And that's another part of the story going forward, is editing. And that will happen as well. Russ: Well, as a teacher, professor, anybody who has taught, or coached, or been a doctor, I think knows--or a patient knows--that it isn't important what you say. It's what they hear. That's a quote from some coach that I heard. Obviously they are not the same thing. We can't understand that. When you first start teaching and someone says something they claim you said, you say, 'But I didn't say that.' But that is what they heard. That is what they heard. And so, you are right: in that patient-doctor interaction, errors must be incredibly commonplace. Guest: Oh, gosh, yes. And the more it's looked at, the more serious the problem becomes. And so this is just part of the--really whole different look. Up until now these things weren't even questioned. But because this information flow that patients are going to be generating more information themselves--it's now--for example, it was, up until very recently, you had to get a doctor's order to be able to get a lab result. To go to a lab. And now, just last week, Labcorp, one of the two largest central labs in the country, has re-fit the whole thing by saying, 'You can order your own labs.' And that's a big deal. But again: you have your labs; you can not only go to a Labcorp facility to do that, but very soon you'll be able to do it yourself, in the comfort of your home with a little kit that goes, pop into your smartphone. That ability to do your own labs, to do your physical exam of yourself or your family member or you child--this is really valuable information that's being generated, outside of the usual hallowed halls of medicine. Russ: Yeah; and the other part of this, of course, that the economist thinks of is the ability to get a second opinion at a relatively low cost. My mom actually confessed to me at one point recently that she didn't want to get a second opinion because if might hurt her doctor's feelings. And I said, 'Mom, it's your body. And it's a serious condition--we're not talking about something minor here. You've got get a second opinion.' I think it's very difficult in the current cultural environment for some people to challenge doctors' expertise, to demand ownership of information. And it's going to require some kind of--cultural revolution is too strong a phrase; it's got bad overtones--but a change in our culture with how we interact with medical expertise. Guest: Yeah. That cultural change is fundamental, no question about it. I think that because there are so many things that are happening so quickly now, this is a time--I've been a physician and a student of medicine for 3 decades and I've never seen anything move this fast, this disruptive. And so I think--in many ways, it's exciting, but of course for so many it's anxiety-provoking, too. Russ: Yeah. I asked my doctor at a physical two years ago for some particular test--I can't remember which one it was, now. And he said, 'Well, why would we do that? What would be the point of it?' And I said, 'Well, I want to look at the level of it because I've been doing some reading lately and it seems to me it should be having an effect on that level, from what I've been eating and I've been exercising. I'm curious about whether it's working.' He was intrigued by that, but he was not up on it. And that kind of--I think the--and you are right about this--the empowerment of patients through Google and medical access online to find out stuff, that people of course desperately do when they are in trouble, is changing that culture already. Because people are taking, they are challenging wisdom in ways the didn't before. Because they can. Guest: Right. In the book I get into this whole smart-patient story, that so many patients now are diagnosing complex things about themselves. And that's for many reasons. And it's not because they can look things up on the Internet, because that's not new. That's been around since the 1990s. The difference is having your own data. Not looking up something and thinking you have it, a diagnosis, because you saw it on the Internet. So, you have your own data, and that could be people who have diagnosed your own genetic disorder from being able to have your own DNA (deoxyribonucleic acid) information, and it could be with centers[?] making important diagnoses. This is having your own data. And going back to another point you made previously, it's not even about getting a second opinion. You can get a first opinion, at any time of the day or night, for the same cost as a co-pay that most employers have. You can summon a doctor, whether it be to your house or right onto your screen of your phone. And you can get a very good consult for $30 or $40. And not only are you just talking to a doctor, but exchanging this data, like you were talking about--the lab information you had or the physical exam. This is a whole different look[?] because you probably know that in the United States the average time to get an appointment with a primary care doctor is 2.6 weeks. Now you can do that in seconds. So you can get another opinion. So it isn't just you interpreting all your own data, but now you are talking to a licensed physician immediately.
14:43Russ: I want to get back to that. Before I do, though, I want you to give some examples you tell in the book of how people are using smartphones to gather those data that we're discussing. Because I think, for me, when I first started reading about it, I was thinking--well, my phone: it can get me my pulse rate; I can put my thumb on the sensor. But of course people are adding things to smart phones. So talk about some of the sensor and devices that are being added to smartphones. Talk about your experience on the airplane with a couple of patients. Because I think those are really near-miraculous. I was telling an emergency room doctor about it, and he said, 'Well, how would he get that information?' So I had to go back and re-read the book to remember what it was. So, tell about some of those stories and how smartphones are being augmented by technology. Guest: Yeah. That's the key point. It isn't the smartphone itself. You have to get, typically, some kind of hardware sensor. So, like for the electrocardiogram: there's now a credit card where you just put your thumbs on the back of this credit card; it's got two sensors on it, makes a circle with your heart to get a cardiogram. And of course that's just the beginning. It isn't just that you can get a cardiogram. There's also software embedded in the app reads the cardiogram. So, some of the instances you are referring to is being on a plane and a person having chest pain. Unclear if it was just bad gastrointestinal distress, or was it a heart attack? And doing a smartphone electrocardiogram and seeing absolute unequivocal evidence of a heart attack, leading to an emergency landing. But it isn't just that. That was one example, but it's prototypic, because it means not just having the sensor, but it also is about getting the interpretation. So here you have a validated algorithm that--it gives you an immediate answer as to what you are looking at. So you don't have to be medically savvy or qualified. You have this computer, this machine, that's giving you the answer. So, electrocardiogram is one example, but there are so many others: that you can do all your vital signs, blood pressure, now it's technically feasible to do blood pressure with every heartbeat, or you never have to press Start--you wear your phone with a watch. Or there's a device that we're testing, [?] you hold it to your forehead and it gets your blood pressure, oxygen concentration in your blood so you can diagnose whether you have sleep apnea during a night of sleep. So many diagnoses can be made through a variety of sensors. Now, another great example is the number one reason why children are seeing pediatricians is because maybe they have an ear infection. But there's an add to the phone, pop on the phone, you can look in a child's ear. It's idiot-proof and you can't hurt the child's ear, but you can get a beautiful, exquisite picture of the eardrum. And that can be read through an algorithm, whether the child has an ear infection or not. And the only reason you need a pediatrician is to get a prescription for an antibiotic. Russ: Maybe we can fix that, too, by the way. But that's a long--that may be the last part of the democratization of medicine--my own access to my own drugs to put in my own body. But that's another story. You didn't get into any of that. Guest: Yeah; outside the United States, by the way, Russ, you can go to the pharmacy and just buy the drug. You don't have to get a prescription. But in the United States it's highly regulated. And I don't know that that's wrong. I'm just pointing out that it's not the same. Russ: Correct. Good point. I talked about the fact--I was once in a drug store in Chile and I realized that what's over-the-counter there isn't over-the-counter here. And you have to be a more informed consumer. And we've deadened that awareness in the United States. Anything you can get in a drug store that you can buy over the counter is pretty weak and pretty safe. It's got plusses or minuses.
18:47Russ: Before we leave medical paternalism, I just want to tell a story, get your reaction, that your book reminded me of. I was in the High Country of Yosemite, near Tuolumne Meadows with my family and we were waking up our first morning there, and my daughter, coming out of the bathroom, stumbled, lost consciousness, and fell down and hurt herself. Not too badly, thank God--she didn't hurt herself in the fall, which she could have. But she lost consciousness for a brief time. So the medical people came. And their recommendation--this was our first stay on our vacation in the High Country, and we're at 9500 feet, and my suggestion was that she was having altitude issues and that we had to take it easy that day and that seemed to be the right thing to do. Their recommendation was that she may have had an epileptic or some kind of seizure, some kind of fit or seizure, and we should take her an hour and a bit into, I think it was Truckee--I can't remember where the nearest medical facility was with some serious diagnostic tools and make sure that it wasn't something more serious. And my reaction was, she doesn't have any history of that; there's no history of that in our family; we're at 9500 feet; my best reading of the evidence is she's got altitude sickness and we should just either go down a little bit or just not do anything very active that day. They made me sign a horrifying form that basically said I don't care about my daughter and I'm giving up all responsibility and it's all on me. Guest: Oh, wow. Wow. Russ: And as I was signing the form--which I did--I signed the form happily--because they had I guess some power to force me to take her there or something, I don't know what it was--I turned to the medical person, and I can't remember what the title of the person was, and I just said, 'If you were her father, would you spend the hours it would take to get this taken care of with the possibility of other tests being done that might lead to other things that are probably not relevant? Or would you just keep an eye on it, as I think I should do?' And that was a very uneasy question for that person. But to me that's always the question, to the paternalist: Okay, you are acting like my father. If this were you, what would you do? So, here was a person being paternal for me about my own daughter, and I just thought 'This is nuts.' And so we stayed; she was fine. She had altitude issues. But that was where I felt that hand. Guest: Well, I think this example with your daughter brings up several important points. One is that, you know, we should start--premise is: Patient knows best. You have context. And oftentimes the people evaluating that patient don't have context. And this is even more emphasized when you are having a lot of your data in the real world. So, everything today in medicine has been essentially one-off, where you go and you get a blood pressure or a lab test or whatever--there is no context. But now you have all this data--it could be real-time streaming in the real world. And in this particular example, you didn't require real-time streaming. You just had the context of, you know, this traveling, high altitude and all the other features. Where, the person coming in to evaluate it doesn't have any context, really. So, we're going to--and not only do we have contextual computing; add to that that this can be done through the data that is being gathered, which we didn't have years ago. So a lot of different things. But the premise is that patients are smart. And today, unfortunately, they are regarded as the Rodney Dangerfields, as I wrote--'Don't get no respect.' And not only that, what you brought up, is this whole 'against medical advice' thing, this intimidation. And I gave an example with my mother-in-law when she was in the hospital, in the book, about--she was almost killed in the hospital. And I asked my wife to get her out of the hospital. My wife and the mother-in-law, father-in-law, they all felt they couldn't do it, because that would be challenging the doctors. And that isn't right. This isn't right. This is why the call for medical emancipation--that's where we need to be headed. Russ: I'm going to ask you a question; I want you to reflect on it personally. I'll lead you with another story. I knew someone once who was at Stanford who was visiting to get a heart valve replacement. This is in the mid-1980s. And she was given a--at the time, maybe it's still true, there was a choice between getting a pig's heart valve and I think a plastic one or some synthetic one. And this was her second one, because the first one, which was the pig valve, had worn out. And of course, having a heart valve replacement is not a lot of fun. So she, with trepidation, asked the doctor: Maybe instead of the pig valve this time she could get the plastic one? Which would last longer. And he said, his first response was, 'Mrs. So-and-so, I've held your heart in my hands. I know what's best for it.' Now, I've always loved that story, for two reasons. It's very powerful. It's very poetic. Right? But how are you going to answer that? And so I'm thinking about you on that airplane, saving a person's life--this was an extraordinary surgeon who had a high self-esteem. Which was merited. He was an extraordinary man who had saved many lives. And I'm thinking about you on that airplane. You told two stories in the book, one about where you diagnosed a heart attack and said 'We've got to stop the plane.' And the other was, 'Nah, the guy ate the wrong thing and he's fine'--whatever it was. That's intoxicating, isn't it? Guest: Well, I think that what I tried to get at in the book was that you don't need a doctor to do this now. I mean, the individuals--I've had these three flight experience. In fact, people think, well maybe they shouldn't get on a plane with me because they might end up having to have [?]-- Russ: No, no. I think it's the other way. I think I'd like to just have you along. Guest: But you don't need me-- Russ: it's expensive-- Guest: Because anybody can do it. Another passenger. One other really key thing here, is we are talking about really cheap chips. Moore's Law, 50 years' anniversary a week ago--this is stuff that you can make for pennies. A dollar. This is really cheap stuff. So the stuff that can be used to diagnose anywhere, 30,000 feet. And this includes many different devices. This is cheap stuff with algorithms that are software that will basically give the answer, so that another passenger, or a flight attendant could do this same thing. And so, I would never want to be coming out with 'I've held your heart in my hand,'--or those kind of--it's crazy, because the stuff we're talking about, today, is about the fact that it is democratized. It's for anyone. It's for, because we have this thing called machines. Computing. Immense computing power and cheap chips and you've got a recipe for a whole different look of medicine.
26:27Russ: One of the things I loved in your book is your thoughts on the stethoscope. Which I think if you asked people to draw a doctor, you'd get a white coat and you'd get a stethoscope. Talk about the stethoscope. Guest: Well, that is symbolic of the whole change in[?] medicine, because we're at the 199 years of the stethoscope's invention, and it's the icon of medicine. And it is a total relic. It shouldn't be used. I haven't used a stethoscope in almost 5 years now. And every patient, I examine their heart; but I use a high resolution ultrasound. Which fits in the pocket. And it is so much more informative, because I can see everything. You know, I can do the entire screening of a heart in a minute, the same time it takes to do an adequate exam with a stethoscope. And here's the difference: Not only do you see every part of the heart--not 'lub-dub'--but you see everything. So you see the valves and you see the heart muscle and the thickness and the aorta and whether there's any fluid in the sac around the heart. You can even track the blood flow. Everything. So you see all this stuff. And it's recorded. It's digital. You are digitizing the patient's heart. Whereas the stethoscope--it's a stethophone. It doesn't scope into anything, doesn't look into anything. And it's analog. There's nothing recorded. And whatever--if it was recorded--isn't worth much. So, the reluctance to switch to this era of new stethoscope, if you will, a real stethoscope rather than the stethophone--it's all part of the unwillingness to change in medicine and moving to a digital era. Russ: But you are right that some doctors have mocked you for mocking the stethoscope, correct? Guest: Oh, sure. And I get into in the book--they feel that the stethoscope is sacred and it's sacrosanct and nothing is ever going to be better than the stethoscope. Well, I have to also remind those people that when René Laënnec invented the stethoscope, almost 200 years ago, it took 20 years for doctors to accept it. Twenty years. And they--it was a revolt-- Russ: It was cheating. Guest: Imagine having a war about this? Well, they said it was going to interrupt with the intimacy of their exam by having this equipment. It was going to be a terrible tax on doctors to have to learn what these sounds were all about. I mean, just amazing stuff. Not only that, Russ, but since as an economist you can get into this: In the United States alone we do 130 million ultrasound studies a year. And that number keeps rising. And as I mentioned, in the hospital, most people while they are in the hospital, they might well get an echocardiogram. It's the indication is it's for the presence of a heart. So, you are talking about $100 billion dollars a year of charges for ultrasound studies. Well, we've done work in research that show, at Scripps, where you could probably get rid of 70% of these ultrasound studies by just incorporating this as part of the physical exam. So here is a tool that's been around now for some years but it isn't being used to any significant degree in this country, because it is a challenge to the incentives and the way that billing and free imbursement[?] proceeds in the United States. Russ: One of the other incentive problems which you talk about which was alarming and informative for me was the medical imaging industry generally. Not just ultrasound but MRI (Magnetic resonance imaging) and various other forms of mammography, etc. Things that look inside us. I've always thought how glorious those things are. When you have a headache and you get an MRI, and they tell you, you don't have a tumor, it's extremely comforting. And when I had my first one I was excited to see that economics part of my brain was in fact slightly larger than the other parts, so that was, you know, I always wanted to feel good about that. But on a more serious note, a lot of the imaging that gets done, is done for defensive reasons, I assume. For legal reasons. And I never realized--and you point this out dramatically, the amount of radiation we get when we get those x-rays or other types of scans. And you argue, I think convincingly that patients ought to be told, when they get imaged, what the radiation exposure is. Guest: Right. Well, this is another one of these doctor orders, where 'I want you to go have this nuclear scan.' Well, nowhere, currently, is it routine nor less practiced to say, 'I think, Ms. Jones, you should have this scan, but I want you to know that it's the equivalent of having 2000 chest x-rays.' You know what Mrs. Jones would do? She would typically, once she gets that information, she'd say, 'I don't want any part of this.' And so, if you want to help people avoid radiation, your first step is give them the information about what they are exposed to. But it's amazing to me, the lack of [?] of patients to demand this information. It's just unfathomable to me-- Russ: Well, it doesn't cross our minds. I never think about it. I think about it at the dentist, when they put that lead sheet over you. And as you point out, getting a lot of pediatric x-rays for teeth is probably a bad idea. Guest: Oh, gosh, that can be downstream. That can lead later to have serious implications. But experts in this country--and I'm certainly not an expert in radiation hazards--but medical scans are thought to have induced 3-4% of the cancer that we have in this country. And we in the United States abuse medical scans. I mean, compared to anywhere else in the world. It's, you know, far greater use, for some of the reasons that you've already mentioned. So, one of the first steps that could be done--if there was a willingness--and that's again, it's all part of this, never came up before because of deeply ingrained paternalism--but every patient should demand--if they are going to have anything that's ionized radiation, and that I mean, CT (Computed tomography), nuclear, PET scans (positron emission tomography), angiograms--any of those things, they better know how many equivalent chest x-rays or the actual radiation, in terms of millisieverts, how much they are going to get exposed to. And just knowing that, a lot of people wouldn't have the scans. Or, they'd have one that didn't have radiation imposed.
33:48Russ: And of course we don't give medical advice on this program. You should consult your own physician when making these decisions. But you should also be pro-active. And I think the other part of this, and we've talked about this many times on the program, is just the paternalism in the medical profession and the way that power flows right now, is: The doctor will often do a test on you and not tell you. Forget whether they are going to tell you whether there is radiation involved, or other--that's just one dramatic example of potential future costs. They just start doing the test; you say, 'What are you doing?' 'Oh, I'm doing a test for such-and such.' 'Well, I didn't give you permission?' But that happens all the time. It especially happens to pregnant women. I've talked about that on the program before. But I was sitting at the dentist a few years ago--and I quit[switched?] dentists after this experience; I quit this dentist. But the dentist started doing something in my mouth, and I said, 'What are you doing?' 'Oh, I'm checking you for--some kind of--cancer.' I said, 'I don't want it.' [?] Guest: I didn't see anything happen to me. I went into the dentist, you know, maybe I was a little dehydrated, a little cotton-mouthed, and you know, I didn't even know, but he did a brush biopsy of my tongue. And then I get a bill from the pathologist for, you know, what was nothing. A couple hundred dollars. I just couldn't believe it. I switched dentists, too. This is--that kind of stuff is atrocious. It shouldn't happen. And, you know, that's part of this kind of civil rights, medical civil rights, you know. Russ: But, as you talked about in the book, in a way, it's up to us. It's not always up to us. There's legal issues. There's sometimes regulatory issues. But sometimes just a little bit of backbone can make a difference. Staying on this theme of paternalism a little bit: When you were first on the program, which I think was about 2 years ago, we talked about 23 And Me, which had started fairly recently at that point. They have since had some issues. Talk about what happened to them with the FDA (Food and Drug Administration). Was the FDA right? Guest: Well, yeah. So they had a collision with the FDA. I actually think 23andMe does a good job, and it's bargain. It was, for $99, the one thing that you could get that was unique, was your genetics, with 30 different drugs that are in common use that have an important genetic signal as to whether you would respond or whether you would have a serious side-effect. It's hard to get that information otherwise, and it's very expensive. So, I thought, point[?], was great for that. I mean, there are other things that were provided which perhaps were not terrific. But nonetheless, because 23andMe started getting into an aggressive marketing campaign, and at the same time they weren't responsive to FDA correspondence and communication, they were shut down. Not a good idea, to have a multi-million marketing campaign and not to respond to the FDA. So, now they are getting back up. They had 800,000 people with their saliva, DNA, that they've analyzed. It's the largest genetic, or [?] repository in the world. And they could have had millions. In fact, they had their sights on 10 million people. That would have been right by now--having been held up by the FDA now for, you know, at least a couple of years--they probably would have been into many millions. But nonetheless, I actually think direct-to-consumer genomics is particularly fine, as long as the data that's being generated and returned to the patient--and ideally owned by the patient, is accurate and is giving proper, validated interpretation. There are a lot of really shaky companies that are doing consumer genomics. That's a different story. But 23andMe was very forthright about how they analyzed, genotyped, their data. And now another company, just last week, was announced: Color Genomics. I don't know if you saw that, Russ. They are doing cancer sequencing of almost twenty genes, like Braca (BReast CAncer genes 1 and 2, BRCA1 or BRCA2) cancer genes 1 and 2 and many others that have familiar cancer. For $249. Which is an incredible bargain, because just BRCA1 and BRCA2 sequencing, through a myriad, was almost $4000. So, that's a new consumer-genomics company, which is kind of following in the footsteps of 23andMe in some respects. And we're going to see a lot more of this. And so, 23andMe was the frontrunner in the democratization of genomics. And what's interesting is--and as I pointed out in the book--is that the medical establishment in the book did not like this. Not only the FDA, but the American Medical Association. All sorts of entities went against it, because it was challenging the fact that people could get their own DNA data. By the way--their own data. Without having to go to a doctor. Oh, my gosh. Oh, my gosh. They could actually [?] learn about themselves, get genetically oriented, without a doctor's involvement. Isn't that something! And so, fortunately, I think 23andMe will get back up. And we'll not just see these two companies that I've mentioned but many more going forward. Russ: Talk about Theranos, which is a company I've been keeping an eye on with fascination because it's--well, there's going to be regulatory issues that crop up everywhere in these kind of innovative situations. Talk about what Theranos is trying to do and what you experienced when you went there. Guest: Yeah. Well, it's a very innovative way--although the actual technology hasn't been still disclosed, what it is. I'm assuming it's some type of microfluidics. But at any rate, you basically can get with one droplet of blood, hundreds of routine assays. Very inexpensively and quickly. And Theranos is signed on to put this capability in all the Walgreen's drug stores around the country. Although so far it's only in the Phoenix area and then one in Palo Alto. So it hasn't really gotten throughout the country, so far. It hasn't dissipated. But that is, I guess, in the works. So this is kind of the creative destruction of lab medicine, that you could do it at your drug store. But I actually think it's just a way station to smartphone labs. And I've talked to Elizabeth Holmes, who is the CEO (Chief Executive Officer) of Theranos--that, why do we have to go to a drug store when you can do this through a simple attachment, inexpensively, accurately, through your phone? And she actually acknowledges that that is going to happen in the future. So there may be complementary ways for people who don't want to do this on their own. But whether it's at the drug store or whether it's wherever you are, through your phone, that you are attached to, there are going to be new ways of doing labs. The old lab medicine, which has been around for 60, 70 years of going to the lab to get all this blood drawn-- Russ: Give up a bucket-- Guest: at very high expense-- Russ: They take a bucket--it's expensive and it's really unpleasant, because they have to take a lot. Because they do a lot of tests sometimes. Guest: They put in a big needle in your arm and it doesn't feel good, but they take all these tubes of blood and then you can't get your results. You know--what is this thing? So that will--that is an anachronism. It's going to go away quickly. And you saw that, with LabCorp's announcement, where they are--you know, you can order your own labs. That's just step one. How about always getting your results? If it's your blood, or whatever your test is, urine, whatever, why aren't you getting your results back? Aren't you entitled? Who paid for this? Russ: Well, not me. That's the problem. That's another issue, the third party paid for it. So therefore there's a presumption. Guest: Well, you really did pay for it.
41:59Russ: Well, I paid for part of it. All in different ways. Physically and sometimes financially. No, I just had a physical a couple of months ago. And there's always the blood work part of the physical. Which I have to confess has gotten more pleasant. The needle is sharper. It's not as painful as it was when I was--in 1960, when I was a 6-year-old, needles were different then. They really were. It's not just you get more mature at dealing with pain. But it's no fun. And then, what inevitably happens is I get no call from my doctor. That's the good news. They didn't call me to say I've got diabetes or whatever it is. I just didn't get a call. And in doctor's office--he's great, I like him a lot, I'm not complaining. But it's just interesting that they are trying to--they've got a Portal. And I spent a lot of time the last time I was there trying to work with the Portal. I haven't used it since I went there. It's complicated. They don't make it easy. I didn't think about it. But it should have been. I should have gotten an email, at a minimum, with my lab results. Even if it was two weeks later. But Theranos can give you--how long does it take? Guest: Well, they say 20 minutes. Russ: That's an average. Guest: [?] Russ: Should be 10. Guest: I got that. You can get them much faster. But unfortunately they are not the stores they have in Arizona--you have to send the sample to Palo Alto. So that takes days. But you can--you can get your glucose immediately, if you are a diabetic, and you can do that through your phone, now. And pretty soon you won't even have finger sticks for diabetics, over the next few years. The point is that you can do any routine lab test through your phone. I mean, a great example, Russ, is in Rwanda of all places, 100% accuracy of diagnosing HIV (human immunodeficiency virus) and syphilis through a smartphone, which costs $.50 to add into the smartphone, and the results are back in 20 minutes. So, if you can do that in Rwanda for pathogens like HIV, you can do you-name-it in the United States for cheap, quickly and accurately. Russ: So, Theranos, as you point out, has not revealed how it does it's test, which is a natural impulse. But that has opened the door for Labcorps and others, people who are trying to creatively destroy, to say, 'Oh, it's not reliable; we don't know.' And it really reminds me of the sharing economy, Airbnb, where the existing special interests say, 'Oh, this isn't safe, it needs regulation,' etc. Of course that's what they are going to say. And they could be right sometimes. I'm not suggesting it's always true that the upstart, the new business is better. But it's pretty obvious that Theranos is cheaper, quicker, and will totally destroy those companies if they are successful. And the thought that comes to mind is that for some people, just like some people use Uber--I was just in Israel; it's illegal to use Uber in Tel Aviv. I'm told there are 15 Uber drivers anyway. There should be probably 1500. But there are 15 people who have--you press the button, and you hope that they come. Because they are taking a chance that you are not a government agent or a policeman trying to capture you. And similarly, it will be interesting to see if some of this technology, even if the existing competitors can stifle it or slow it down--people want the knowledge. And some of it may happen anyway. Guest: Right. And this sharing economy, and this Uber example is coming to medicine big time. Now you have 5 different apps, companies that you can touch the app and you can have a doctor come to your house. And one of them was started by an Uber co-founder. And the doctor comes in an Uber car. And it costs between $49 and $99 for a visit; and it could be suturing--it's not just 'Hello, how are you?' It's anything you need, short of having to go an emergency room or a regular medical consult. So, we're seeing the same, on-demand, I want what I want when I want it mentality. Which is across the board, because it's a mobile device that has driven that. Why would you want to try to hail a cab or call a Yellow Cab, when you can just touch your phone and then you have this much better experience with dealing with an Uber driving? So this is coming to medicine in a major way. It is of course tele-medicine, where you don't have to have the doctor come to your house--you can just see the doctor and have the exchange. And it's not just as I mentioned earlier an exchange of talk, talk, talk. It's actual data exchange. So, 'Here are my blood pressures. Here's my exam of my child's eardrum.' Or whatever it is. So, this is a very different look. This is the epitome of how medicine gets democratized. Again that mobile device is center stage. It's the driver of this. Russ: Yeah. It's fascinating. A question I had while I was reading that part of the book is: my dryer in my house was making a horrifying noise this week. So I called a repair person to come out. And he charged me $59--just to come out. Guest: Oh, yeah. Sure. Russ: And then, fortunately, he had the part with him and the repair wasn't too onerous and the total fee was pleasantly reasonable. But how can a doctor come out for $49? What kind of doctors are interested in tele-consulting? Are they doing it out of--is it the profit motive? Is it, they are excited about the technology? Is it they can't get a job in a regular practice, so this is what they are doing? Guest: Oh, no, no. These are amazingly well-qualified doctors. None of them do this full time. They are doing it on the side. What's interesting is a lot of them get paid--a lot of times they are getting paid by the hour. And this is just the beginning, when they are getting [?] and they want to start to get the buzz out there so they are taking the hit financially. These prices won't last. But they are the way to get the word out that, 'Oh my gosh'-- Russ: Get reputation. Guest: Yeah. These are a lot of leading medical center docs that are doing this. Some of them are doing it illegally. It's kind of like those Uber drivers in Israel: they are not supposed to provide medical services through their contract with their health system. But nevertheless, they are lured to this because this is where medicine is going. It's out of traditional [?] and brick-and-mortar. It's actually the return of the house call. And all these things that you can do when you are there, that you couldn't do before. And by the way, you don't need all of this stuff where you go to the clinic and you wait in the office for an hour on average and you have someone come in the room to do your vital signs, and all these extra people, all this expense. You cut all this stuff out. So, economically it's very attractive. Russ: Yeah; but the way, I don't really know if there's anybody in Tel Aviv driving an Uber car. It's probably just a folk tale. I don't want to get them in trouble. So, forget I ever said that. Ahem.
49:49Russ: Now, I want to shift gears a little bit and talk about something that comes up in a lot of detail in the book, which is the promise and perils of big data. And listeners to this program know that I'm skeptical to some degree about big data and its potential. Obviously it has potential. It's easily misused. I've been quoted on the program--maybe I shouldn't say this, but epidemiology is an intellectual cesspool. And people get offended when I say that. But I say that based on the idea that most--I think most findings of epidemiological studies relationship in a population between say drinking and health, drinking and cancer, various fat, carbohydrates in your diet--they are often studies on both sides; we don't really have a lot of information about what's reliable. A finding about coffee being good for you will be reversed 6 months later by a giant study saying it's bad for you. Then it's good for you. And it reminds me a lot of macroeconomics, where we have a lot of variation across the population; we have a lot of complexity; and we don't have access to all the data--we have the data that comes in the survey, say; or that the government collects in the case of economics. And the result is highly unreliable sorts of information. The revolution you are talking about, where smartphones sample people's genetic code or their vital signs in real time, which is so much better than saying 'I have this pain' and then you go in and it doesn't manifest itself so they can't diagnose it. But if we have a lot of data about what's going on and a lot of time, there is the potential for actually discovering things, not just about the human body and its relationship to other things generally, but also about my body. So, what do you think the prospects are for that revolution? And talk about the privacy issues that you do in the book, which are the peril, among the perils. Guest: Yeah. I think the privacy is my biggest concern for holding up this revolution in medicine, because if people's medical data are put out on the Internet or brokered or sold, that's not going to work, with reidentification concerned. So, that's why I think the critical step is that each individual has to own their data--has to shift the ownership model. All their medical data. And by the way, wouldn't that be a flip--when you talk to the doctor and you say to the doctor, 'Would you like to see my data?' That's where we are eventually going to go. I already have a lot of patients, who, by the way, say, 'Would you like to see all my blood pressure, my heart rhythm?' I've already seen where this is headed. So, privacy, security of data, where we are doing nothing to advance medical-wise today, has to get on track. But as far as the big-data thing, obviously there's a lot of noise and how do you get signals, and I have a whole chapter about that. But, what I want to get into, the promise of that, assuming we can come up with the right privacy, security, using technology in many respects and legislation that is critical--if we can get over the privacy hump, then we have the ability to prevent illness. Because it's not big data--I like this point you made about epidemiology at the population level. It's big data per individual. We're digitizing human beings. So, if I am getting real-time data about you for a condition--let's say, asthma--and I'm getting your lung function; I'm getting your nitric oxide in your breath; I'm getting all your vital signs continuously, your air quality, pollen count--I mean, you name it: I'm getting all this data about you. Without your knowing it. That is, you are electing it through your phone--you know you are doing it but it's seamless. So, you know you are doing it because you don't ever want to have an asthma attack again. Or you don't want your child to ever have an asthma attack again. And then, this big data about you is getting all integrated, and processed and analyzed, and telling you--a text or a voice message: 'Russ, you are going to have an asthma attack if you don't take a couple of puffs of this particular inhaler. Because we can detect that you are starting to get early signs of bronchospasm[?] that you don't know, you don't feel it, you don't hear it.' And so that's where we're headed in medicine, where you have big data and we're having it properly processed. And that has enormous potential. And that has nothing to do with the population level. But when you have so much data about any condition for hundreds of thousands, millions of detail, then you get even smarter in the machine learning on the individual basis. Russ: That raises the question, which you allude to in the book--you talk about the driverless car. Which we've talked about here a number of times; I very excited about it; I think it's inevitable; I think it's coming. And of course it threatens the livelihood of a bunch of people who currently are driving taxis. Or who make cars. So they are emphasize the dangers of that. Similarly these algorithms and machine learning you are talking about threaten to put doctors out of business--interpreting x-rays, as you said, correlating events with the onset of an asthma attack. What are we going to need them for, down the road? Guest: Well, that's the doctorless patient concept that I have tried to advance. Of course, it doesn't go over too well with a lot of doctors. But it isn't--it's semi-autonomous. It's not like the driverless car, where who would have ever thought that would advance so quickly and be so safe, as I'm sure you've seen, Russ, or even call for the banning of human drivers because of the safety of several of those cars. Russ: It'll get to that eventually. Guest: Yes, some day. Russ: Who would be so reckless as to teach a child to drive a car? It's dangerous. Just like giving them a drink. Only bad things can come from it. Guest: There you go. And so now: what's the parallel in medicine? And what I see is: Doctors are essential to treatment. They are essential to have oversight of the data, whether it's diagnostic data or the monitoring data. That patients are going to be largely in charge of their diagnosis and monitoring data--for most common, non-very-serious conditions. They are going to be kind of ruling the roost for collecting that data. But they still are going to need doctors who have the experience and the wisdom to provide guidance for them. To communicate, and help them; provide empathy, too--a lot of times it's going to be someone who has an ill condition. So, that is the morph here of the doctor's role, in a partnership role where the patient is generating a lot of this data, the bulk, dominant portion. But the doctor is critical to put that in context for what to do. Does it need treatment? Do we need to have additional whatever? So that I think is going to be a reset going forward. Russ: And many specialties presumably would not be compensated at their current level in such a world. Guest: Well, actually, I think this is the problem today. Doctors are very busy, and they are doing a lot of things in the diagnosis and monitoring mode that they don't need to do. And computing can get much more active in this arena. This is where algorithms and cognitive computing can make a big difference. But if you were just being the guidance person as a doctor and you spent your time doing this thing, which is so vital--this is--ulcer, with diagnosis, diagnosis, diagnosis, you know, in the future it's treatment, treatment, treatment--this is the welcome change. Because it's going to decompress doctors' busy life and shift more responsibility to the individuals who have the most vested interest in their own health. Russ: That would be you and me. Guest: Yeah, you and me, and the wisdom of the body. We have known about the wisdom of the body and how good the body is for regulating glucose and sodium and thirst and hydration. But there's a new, external wisdom of the body--the data on your smartphone. And that's going to make a lot of people a lot smarter about how to deal with their health. And they still are going to need doctors to help advise them, too. Russ: Well, it's interesting because I wouldn't say that empathy is necessarily the trait that currently is at the front of the list of traits that gets people into medical school. So it would be--there would be a lot of changes, some of them cultural, some of them practical, like we're talking about. I suspect in, say, 30, maybe 20 years from now--I hope EconTalk is still available--people can look back on this. I hope I'm still here. Guest: I hope so, too. Russ: People will look back on it--I'll be 80 in 20 years. I'll still be the host. But at least I hope it's still online. Guest: Eighty will be the new 50. Russ: I'm hoping people look back on this and say, 'Wasn't that interesting? They actually thought that--or whatever it is.' But I suspect people going into medical school 10, 20 years from now will have different expectations about their role and their compensation. Although I do think that the AMA (American Medical Association) and others will fight very hard to keep the system as similar to what it is now. It's just the nature of-- Guest: Oh, you're right about that. That's just the nature of it. And the sort[?] of thing where you don't get a fix for this from the inside. It has to come from either the patient's unrest, the public. Or particular forces out there that have muscle, like large employers, with hundreds of thousands of employees, saying 'We're not going to take it any more.' But it won't happen--this change that is in the cards, that is unquestionably going to occur over time--any time soon, if it's up to the medical community itself.
1:00:36Russ: Do you see a world--a good friend of mine is an emergency room doctor, and he sees everything from sore throats to horrible, tragic situations. Some of those, by the way, will disappear--with driverless cars, I hope. Somebody was telling me the other day that the biggest challenge of a world of driverless cars is there won't be so many organs for transplanting because there won't be so many fatal car accidents. We'll figure out how to make those, I hope and assume we will. That would be great if that was our biggest problem. Wouldn't that be wonderful? But I think about--so much of what he encounters is not amenable to an algorithm. It's not just diagnosing whether this person is having a heart attack or not. It includes serious surgery that's not replicable. But I wonder if there is other surgery coming that could be done robotically. My fantasy--science fiction--version of this is: You have a toothache; you go into the booth; the booth diagnoses, the smart, the machine learning, the AI (artificial intelligence) in the booth tells you that you need a root canal. You open your mouth; the machine does the root canal and you walk out. And it does it perfectly and you walk out and you're fine. Similarly, if you are going to have your prostate removed and if we haven't figured out by then how to genetically alter people so that they don't have a prostate or that it's replaced in utero with a plastic prostate that doesn't get cancer, that your prostate could imaginably be removed without surgical intervention. Is that ever going to happen? Guest: That's a good question. I think that's pretty far off. Some day. But that, as opposed to so many things we are talking about that are potentially here and now or are already in existence, that's decades away. You are bringing up some really good ideas there. I mean, having just seen the movie Ex Machina, you let your imagination go wild. But yes, there's certainly a lot of artificial intelligence and of course robotics going to be coming into medicine as well. It's already happened. But so far, a lot of the robotic stuff, like the Da Vinci robot for prostate surgery and gynecologic surgery and whatnot--it hasn't really been shown to be better than conventional surgery. But it's marketed heavily; it's very expensive to buy the equipment. So, all these things, everything we've talked about here, Russ, has to be validated before it will be accepted. But, would it be nice for men not to be born with a prostate? Sure. Can we do genome editing now to do all sorts of corrective things that we never would have envisioned would be possible years ago? Sure. So, eventually these things will take hold in many shapes or form, but they've got to be, at some point, clearly proven that they are better than the way things were done, the state of the art at that time. Russ: So, let's close with just some optimism or pessimism. You're an optimist, I would say. As am I. But you've also been confronted, when you write a book like this, when you go out and speak about it--and you write about this in the book some--people not only are skeptical: they are angry. It's threatening. Their natural self-interest encourages them to see you as dangerous, not just misguided--that you are harmful to patients; that this revolution that you are talking about is going to be bad for people; that they are not capable of dealing with these issues that you are trying to give them power over. Etc., etc. In the short time--and time moves quickly today--that you've been speaking and writing about this, do you see the beginnings of a change toward sympathy toward your views, or is it mainly hostile? And how do you feel the future's coming? Guest: No, I see a lot of sympathy. I am amazed actually that so much is moving towards democratizing medicine. Which is all I ask for, and researched to get the book done--which is that we have this unique capability. Although, there are naysayers. And although there are challengers, they are not so hostile. They, I think, realize that so many of these things are the past or the future of medicine. And there is resistance for various reasons, such as financial ones and entrenchment and even lack of knowledge in many years like genomics that we touched on, or the use of sensors and things. But I don't think there's nearly as much adversity as one would anticipate. It's certainly much less than I had expected. So I'm gratified for that. I think that most people realize that this is the best thing that could happen. We're in an economic upheaval in American medicine. We've got a solution here that is tenable, and it's putting people in charge. They want their data; they should own it; and they can function so much better with the amount of data that's going to be flowing through their devices. So, let's open things up. And I don't think there's going to be nearly as much of that type of resistance or adversarial response as you might have anticipated.

Comments and Sharing

TWITTER: Follow Russ Roberts @EconTalker

COMMENTS (22 to date)
Glenn Mercer writes:

Hey, Russ, we were digitizing doctor's notes back in the 18th century! (grin)

Okay, well maybe I overstated that.

Good to see some conditions persist over the centuries:

Glenn Mercer

JLV writes:

Good episode.

The discussion at the beginning of the episode made me think of signaling models (or at least the equilibrium concepts in signaling models). Suppose there are two types of patients, a "low" type for whom getting their data will either give them no benefit or will actively harm them (e.g. Jenny McCarthy), and a "high" type who will be able to use their data to obtain a benefit. If both the high types and low types believe in patient autonomy and do not have a way to signal their type, then it should be obvious that we'll end up in a pooling equilibrium (in this case, at no information given).

Greg writes:

In the UK there is through the Data Protection Act the right (for a nominal fee, about £20) for you to request a copy of any personal information held on you by a company/organisation that they have to provide so you know exactly what information they have on you. This includes doctors notes and other medical records which means even though most people don't realise this doctors have to assume anything they write will be read by the patient.

On the downside this has pretty much killed the amusing doctor's anacronyms and short hands for terrible patients and so on.

Mark Wonsil writes:

Two thoughts:

  • Doctor visits might provide some insight into some illnesses if the physician can see the environment that the patient lives in.

  • There are some people trying to encrypt parts of medical records so that the patient can control the demographic and result data shared (one I saw recently is called the SEED protocol but there are others.

Brian W writes:

Great episode, really exciting developments. The story about your daughter at altitude is similiar to my experience as my wife and I are going through our first pregnancy. I had a doctor once tell me that a test was needed even if no symptoms for the issue existed because "she had heard of it happening before with someone". When I asked for data for these claims I was told she didn't have any but that she was the expert. When I went to decline a third performance of blood tests for my wife in the same week when the first two had found zero issues I was told I would be made to sign a waiver. I am lucky enough to have a doctor in the family who called and got it straightened out but few people are that lucky.

Mort Dubois writes:

Can't wait until my new iPhone comes with integrated rectal and nasal probes for full-time health monitoring.

I agree with JLV - just because people have access to information doesn't mean they'll make good use of it. On the other hand, I have no love for doctors and the existing system. It will be interesting to see how this plays out.

John L writes:

As a physician finishing fellowship training, I enjoyed your podcast this morning. I see increased patient ownership of data and decision-making as a welcome change.

Regarding the form you were required to sign saying that you were going against medical advice: It seemed you interpreted this as a paternalistic challenge to your choice as a parent. I think that's one valid interpretation, certainly. I think another reason for the form, however, is that the medical team (very likely under the direction of their administrators) probably sought protection from the legal system that currently holds physicians liable for outcomes in a way that assumes a degree of paternalism and control over the patient's behavior. I think it would be better to transition to a system in which the doctor bears less responsibility, but there will have to be less liability for doctors in this new system as well.

Relatedly, I think it's worth conceptually separating doctors from hospitals when thinking about these issues. Most of us don't run most medical systems and have little direct influence over access to medical records. They are generally controlled by the health system/hospital in large part as a way to try to keep patients within a system and to compete with other health systems.

My 2c

Shawn Barnhart writes:

On paper, electronic health care records are a great idea. In practice, I think there are a lot of problems.

Most of the clinics I've used have deployed elaborate systems (computers, software, wireless networks, tablets) to allow doctors to work with the systems. These cost significant money and have significant overhead costs associated with them, especially the tablet PC versions. Is the immediate data entry value worth the 100x cost over paper charts and information later entered by a lower-cost data entry person?

I had a boss who made a wise remark regarding bureaucracy at work -- if you require a lot of administrative data collection by employees, you turn them from their primary task into bureaucrats. If we push data entry and collection onto doctors, are they becoming data entry employees instead of doctors?

I also question the impact of form-based data entry on doctors. If doctors become form-driven, does it reduce their diagnostic skills and abilities as they stop making cognitive effort and just check off boxes? Getting doctors to do their own data entry requires the data entry to be convenient to them and meaningful to the record-keeping system which is going to be focused on the kinds of checkbox-oriented data collection that can be analyzed by computer based reporting. It goes against the free-from comments that fall outside the bounds of forms and checkboxes doctors may be inclined to add to paper charts.

Security and privacy are HUGE problems. Banking and finance have used computers for decades and have a huge financial incentive to get security right yet financial theft and fraud remain constant problems. What makes us think that medicine can solve this problem with far less direct accountability and direct financial risk and much less history with using and collecting this data?

And what is the motivation for collecting this data? Is it being used to help me, or is it being used against me? How much is collected just to satisfy insurers vs. improving my health outcomes? I'm reluctant to share some information with my doctor at all, lest he enter it into his computer and I find it used against me, out of context, later on. How do I know what information has been collected about me and correct any errors (and how do I know it's even in error)?

If the credit reporting system is any indication, medical records will be fraught with errors and the system will be stacked in favor of those with a major financial stake -- doctors and insurers -- and against patients.

Dr. Duru writes:

I am thinking hard on how to answer some of the questions in the continuing education section. In the meantime, I have a bunch of immediate reactions.

First disclosure is that I am married to an OGBYN. She has opened my eyes to what medical practice is REALLY like and how the public's perception of medicine can be so misguided and overly influenced by a few bad experiences with a few bad practitioners. I am particularly sensitive now to pieces and people who question doctor's motives.

So, this is my bias when I listened to the first half of this podcast and thought it was over-the-top anti-doctor. I am so glad in the end, you started listing off SOME things that doctors are good for and noted how they can still be positive and functional supports for the revolutionary, personalized medicine that is coming. You also at least made passing note to the dysfunctional institutions that force doctors to behave in certain ways.

I 100% agree with John L. on that form you had to sign. My wife has had to ask non-compliant patients to sign similar forms. It is not that she wants to force anyone to do anything. Seriously. She is all good with someone making medical decisions for themselves. She just needs the legal protections to avoid getting sued when outcomes go wrong. For example, protection from the patient who returns and asks well "why didn't you tell me that XYZ could happen?" Or even better a lawyer or some such asking "Well, why did you allow the patient to leave when you KNEW XYZ..." We even had this happen to us verbally when my wife decided on a particular course of treatment for our daughter that went against protocols. We didn't take offense or assume paternalistic attitude - we know the drill: the doctor needs total confirmation that the patient understands he/she is taking full responsibility. We had a good laugh because we sympathized with the doctor's legitimate concern.

I also mostly agree with Shawn. I used to think of doctors as Luddites. But now I understand better a whole host of obstacles that make medical administration difficult, expensive, and challenging. I understand a lot better why we end up waiting so long in a doctor's office as administrative tasks need to be handled between and during visits. The pressures doctors are under to get patient through the protocols faster because of insurance edicts or reduced compensation for certain care.

Yet, I would love to understand better the quantitative evidence that demonstrates doctors in general would prefer not to share their notes. I think back to a funny Seinfeld episode where Elaine finds out a doctor has put in her record that she is a non-compliant patient (I think that's what happened). That record spreads through the medical community, and she can't even get a veterinarian in a remote farming community to look at her issue. Now, in real life, should a doctor be subject to lawsuits for providing his/her honest assessment that then can influence the way other doctors treat the patient? No good answer. I just want to point out that the patient-doctor relationship is a lot more complex and has a lot of institutional pressures impinging on it that seemed to be underplayed in this podcast.

Having said all that, I am definitely looking forward to personalized medicine and the ability for doctors to truly customize care. I think doctors will also find it incredibly liberating if the system will allow them to depart from what is considered standard practice and procedure and to instead treat a patient as a very specific case or sub-case. I look forward to a world where we can significantly reduce the cost of medical care because prognosis/diagnosis/care can all be better targeted. For the vast majority of doctors who got into the business to help people, they too will find the new technologies incredibly empowering.

And maybe, just maybe, we can even reduce the cost of medical education along the way...

Daniel Barkalow writes:

There's a great chart here showing how different sources of radiation compare, including natural sources, nuclear disasters, and medical procedures, which is great for considering radiation exposure in terms of things from everyday life. (For example, a chest x-ray is 4 times a dental x-ray, or twice living above ground for a day, or half taking a cross-country plane flight. Living in Denver is like living at sea level and getting dental x-rays every four days. A chest CT scan is two years of radiation from natural sources.) I think if people were given that chart and a measure of how big a dose a scan would be, they'd tend to get the scan, but they'd be making an informed decision.

Hi, Daniel Barkalow.

Isn't, to which you link in your comment above for a chart, a website for comedy and publication of comics?

I'm not saying there's nothing there of interest. Comics can be very pointed and accurate. But how can I possibly sort out what's comedy and what's substantively legitimate data reported at your link? Sometimes when facts are mixed with fun, it's hard to figure out.

Jeffrey Withington writes:

Hey Lauren,

The chart linked by Daniel was released by XKCD in the wake of the Fukushima accident. I can't find the original posting right now, but it's meant to be 100% educational.

And Daniel, I thought of that chart, too!

Retired Physician writes:

The early part of this episode seemed to suggest that the practice of medicine involved nothing more than evaluating a lot of patient data, lab tests, imaging studies, etc. It seemed that anyone with a smartphone would be able, in the future, to manage their own care once he/she had ordered their own tests, had their smartphone evaluate them, then purchased their own medications.

Fortunately Dr. Topol corrected this oversimplified view of medical care late in the episode. Knowledge and experience are required to know what tests, if any, are needed, and to put results in perspective, before a wise course of action can be chosen. Perhaps medical training is good for something, after all.

It sounds like Russ has had some bad experiences with his own physicians. My doctor's portal has lab results posted within hours of a test, along with an email from my doctor with his interpretation of them. My care is collaborative with him. I value his training and experience but ultimately I choose what is done for me. All of his patients are treated this way. Look around, Russ, there are doctors like this.

Daniel Barkalow writes:

Lauren: That particular chart cites its sources, so you can check whether they seem legitimate and match the chart. If you're a doctor considering giving this information to patients, you should probably go through that exercise yourself. Honestly, you should probably check sources on any medical advice you give professionally, rather than just trusting it, whether it's xkcd or FDA guidelines.

Hi, Daniel, Jeffrey.

Thanks for your excellent responses. For the record, I didn't really mean to cast aspersions on xkcd. I actually love that the chart is so visual. It deserves the attention it has received!

I couldn't find all the source info Daniel refers to being at that link. All the same, I did some rough checking online and with a few friends who know more than I do (and I know a little just on my own) regarding millisieverts for various activities. While some of the exact numbers don't match up--mostly because there are wide ranges possible for many of the activities in the chart--I have to admit that the chart was fair enough.

It really helps to paint with a broad brush, so that the chart mostly illustrates orders of magnitude. Details like the possible ranges are small by comparison to most of the main (green) items.

The (blue) sidebars with details are a super and illustrative idea toward the original aim of showing where the Fukushima event stands generally. But, looking deeper, the ranges in those smaller medical x-ray details can be very large. Some of the sidebar ordering based on point estimates would change if the sidebars were further elaborated on.

And it's in those sidebar areas where personal, daily, medical decisions are often made. I think a lot of folks don't understand that the issue is that x-ray exposure is cumulative over your lifetime. A choice to do annual dental bitewings in your 20s--and an ordinary reader can't really tell from the chart what is meant by "dental x-ray"--can impact later choices to get a CT scan. (Yeah, dentists don't like being reminded of the cumulative effects.)

It's exactly right that people should be checking their data themselves, and also asking their doctors for more info all the time--in advance of tests and afterwards. And, it may be a reason why individuals should own their own medical data and have access to it throughout their lifetimes. But it also brings to the fore one of the tangential matters from the podcast episode: What if people own their own data but don't realize that this or that particular piece of data matters to the doctor they are seeing at the moment for a specific medical condition? Doctors have to be trained to ask for pertinent data the patient may not realize he has. And to explain why they are asking, because patients may just assume it's a frivolous and unrelated question and may respond inaccurately as a result.

One of my concerns when I see visually enticing charts like this--isolated web pages that get as popularized by social media as this chart has become--is that the info can get locked into the public's minds or perpetuate urban myths. Otherwise, this xkcd chart is a gorgeous, appealing, informative graphic illustration, and evidently accurate for it's original intended purpose of illustrating where the Fukushima nuclear event fell in the overall exposure range.

Gav writes:

With more and more data being easier to access, I'm surprised the topic of over-diagnosis wasn't brought up during this episode. The more we look for problems, the following problems arise:
- more false positives
- smaller and smaller problems will be found, leading to over-treatment

This leads to finding cancers that would never have metastasized, prescribing drugs in situations where harms outweigh benefits (eg. slightly elevated cholesterol or blood pressure), etc.

Dr. Duru writes:

A NYT great article from last year sheds light on the various forces infringing on doctors's decision-making. A quote particularly relevant to this podcast on paternalism:

"Medicine has been appropriately criticized for its past paternalism, where doctors imposed their views on the patient. In recent years, however, the balance of power has shifted away from the physician to the patient, in large part because of access to clinical information on the web.

In truth, the power belongs to the insurers and regulators that control payment. There is now a new paternalism, largely invisible to the public, diminishing the autonomy of both doctor and patient."

"How Medical Care Is Being Corrupted"

Robert Swan writes:

It's a seductive idea. I picture someone like Star Trek's Dr McCoy waving around a little gadget, a few moments' deliberation, and out comes an unerring diagnosis and clear path to health. The reality will not be quite so good I'm afraid.

Besides my built-in cynicism, this bold prediction is based on what a mixed bag hi-tech has been for cars. On-board computers have done amazing things for power and economy. They also have blessed the car with the "diagnostic socket". This is able to give an immediate diagnosis, and it's usually right. Our cars already have their own Dr McCoy: built in, no less!

But there have been consequences. For one thing, the diagnostic plug is a great leveller of mechanics' talents. A skillful mechanic can hardly stand out from the crowd when it comes to plugging in a gadget and reading its screen. For another, the computer isn't always right. Who hasn't heard tales of people having to return again and again to the garage having one expensive component after another replaced and never getting the problem sorted out? Perhaps this is where a skillful mechanic might have a chance to shine.

Bear in mind that cars are fully understood, and these diagnostics are part of their original designs. How much harder is it when working with something as complex as a human body? And (a hobby horse of mine) who decided that the air pressure in inflatable cuff having various effects on the blood circulation at my elbow gives you a pair of numbers that allow my health to be meaningfully compared with Arnold Schwarzenegger's? Or a ballerina's?

There is an another cost with this too. The presence of the diagnostic plug in cars has changed the training of mechanics (at least in Australia) such that some general practical things have been removed from the course to make way for instructing the students in these hi-tech gadgets. What will be jettisoned from medical training to make way for high tech? Or are we headed for a world where medical graduates are going to be in their 50s before they can go into practice?

All that said, there is plenty wrong with medicine today. In my view its two biggest problems are dodgy abuse of statistics by pharma and academics, and the arms race between lawyers and medical defence insurance companies.

Technology would be well down my list from those.

Another entertaining and thought-provoking podcast.

Arde writes:

I have read an interview with one of the leading obstetricians in my country (I am not from the US). She complained that nowadays women are reading so much on the internet about pregnancy/birth giving and that internet is full of wrong information and that this is very damaging. She said that if her patient starts a sentence with "But I read on the internet that…", then she replies: "Then go to your internet to help you, why do you come to me?".

Jim Vandiver writes:

This was a timely discussion for me because I see a cardiologist for the first time in two weeks to explore some heart irregularities. I hope he uses ultrasound equipment instead of a stethoscope. If not, the first visit may turn confrontational! I tried Googling for offices/cardiologists in my area who use ultrasound equipment and only found links to training for and jobs in the use of the equipment. Lastly, I was intrigued last year by some inexpensive self-monitoring equipment from in Italy, but lacked motivation to buy and try it.

Charles Mann writes:

Interesting episode but I can't help but be reminded of the old joke about self-representation and having a fool for a client.

How applicable would that joke be in regards to self-diagnosis?

Peter Wei writes:

I enjoyed your recent podcast with Dr. Topol - as a medical resident interested in technology it was great to hear about the stuff he's been working on.

At the same time, I wanted to push back on the easy assumption that these gadgets will be adopted by the population as a whole, so long as doctors' paternalism doesn't stand in their way. I suspect that although these self-monitoring technologies are useful for what Tyler Cowen has called "infovores," they will be much more confusing and hard to use for the population as a whole.

Note that many patients are not using already-existing health monitoring technology, even when they have the opportunity to. Many insurance plans subsidize the purchase of a home blood pressure cuff, but relatively few Americans have them. Among diabetic patients, a substantial minority do not regularly check their blood glucose even when machines and test strips are provided.

One of the fascinating things about practicing medicine is that it deals with something close to a cross-section of the population. Professors tend to deal with professors and students; lawyers with other lawyers, and so on, while in medicine the job intrinsically involves interacting with a cross section of the population - and one of the surprising effects of medical school was challenging my assumptions of what patients need most and how I could best help them. If I had diabetes or hypertension, it would be awesome to be able to check my blood sugar or blood pressure on the go, and run little experiments to try to optimize my health. And I'd be quite excited to open a clinic that treated EconTalk listeners as empowered patients! But in fact for many of the patients I see, merely taking their medications as prescribed is a challenge. Quite often the best use of the limited time in the office is to reinforce the basics of taking medicine regularly, making lifestyle changes, and (ironically) reassuring patients not to get worried about small anomalies or false positives in their labs.

This is certainly paternalistic, but quite often it's paternalistic in the best sense of the word - taking an interest in the well-being of the patient and counseling them appropriately to encourage them to take better care of themselves. Admittedly doctors could discriminate more effectively rather than blanket-paternalizing everyone, encouraging more educated patients in their own researches. But the overall optimal amount of paternalism is not zero, and is quite a bit higher than it would be for the average EconTalk listener!

Comments for this podcast episode have been closed
Return to top