Transcription
I would like to welcome you to the 2016 Picard Lecture. And the timing of this lecture is absolutely ideal. And you're probably wondering why that's the case. Well, just yesterday, if I'm not mistaken, Justice Ellen Picard retired. Uh Justice Oh, okay. I didn't even need a prompt a a round of applause. Um she started the Health Law Institute in 1977. Uh she really Many people in this room perhaps don't even know this. Uh Justice Picard uh really went a long way to starting the field of health law, not just nationally, but internationally. Uh she's done a lot for me personally. Uh Ellen Can we now officially call you Ellen now instead of Justice Ellen Picard? Uh so I would like to personally thank her, but not just for what she's done for me personally and for the institute, but what she's done for health law, what she's done for legal scholarship, what she has done for the law generally, and what she's done for Alberta. So thank you, Ellen. And I'm so glad that we have such a spectacular speaker for this event, Ellen.
Uh we have uh I think I will now call you my friend, Gilbert Welch, who is an extremely well-known uh academic uh physician. He is a professor of medicine at Dartmouth. Um and one of the things about Gilbert is that I think, and I actually uh asked him if I could characterize him this way. He is really one of the international leaders in a field that's been called sort of choose wisely. The idea that we should re-examine the nature of the of the treatments that we provide, of the screening technology that we use, of really the how we think about the value of our health the things that we have in our health care system. So, he is really an international leader. He had he's had two very well received books that I love he's had more than that but the two that I'm going to refer to I absolutely love. The one first one is overdiagnosed. Now, I actually have a a little bit of a story about that. Uh Gilbert didn't know this, but I had the exact same editor for my book that Gilbert had for his wonderful overdiagnosed. And when the editor was assigned to me, she sent me a copy of of the book and I was flattered I think I'm I'm being put in this rank of Gilbert Ross. And then I realized it was actually an insult. It was like write more like this, please. That was clearly the message that was being sent to me. That book had a huge amount of influence in his in his latest one which you're going to hear about today is equally fantastic. So, this is an extremely knowledgeable individual. This is an individual that has had a huge influence not just in the United States but really around the world on how we think about the value of of health care of all the things that we do. I know this is going to be a spectacular presentation. He's going to speak for about 40 minutes and then we'll have time for Q&A. So, Gilbert again, thank you for coming to to Edmonton and I look forward to this presentation. Thank you very much.
Yeah, that's great. It's great to be here in Saskatchewan at um I'm an American in case you didn't know. By the way, the one other thing that happened yesterday you don't know but uh we're actually uh planning to build a wall on the northern border. You know who's going to pay for it? You guys are going to pay for it, all right? That's what it is. All right, I'm going to start. 60 years ago, I was born affluent American. It is what it is. I can't change it. Um here's a picture of me um as a child. My mother took this, old technology, it was called Kodachrome. I don't know if there's any uh clinical experts in the room, but uh some of you may notice immediately there's some things wrong with that child. The child's clearly overweight. Probably has toddler hypertension. Hyperlipidemia for sure, glucose intolerance, I wouldn't be surprised. And then I want you to look at that face. Uh the pediatricians have a name for this, it's called an FLK, a funny-looking uh kid. I undoubtedly have a bunch of single nucleotide polymorphisms, uh a couple of genetic variants, and maybe even a frameshift mutation. I'm I'm not sure about that. But the real question is, what is going on in there? Look at that thing. That protuberance, that's kind of concerning, isn't it? Could that be an early abdominal aortic aneurysm? Whoa. Boy, I'm glad I was born 60 years ago today.
I want to talk to you today about seven assumptions that drive too much uh medical care, particularly in my country, but it's a worldwide problem. And a disclaimer before I start, uh and maybe it's not important to say this, but I want to be clear, medical care can be very important in selected settings. It can do a lot of good. I'm a conventionally trained uh physician, and and there's a lot of medical care I believe in. But at the same time, there's a growing recognition that the conventional concern about too little medical care needs to be balanced by a concern about too much. In fact, about half of American primary care practitioners worry that their patients are getting too much medical care. So, the challenge I'm going to ask you to think about is can we communicate the nuances of medical care that some is good, but more is not necessarily better. Uh the disclaimer, I have a book on the topic. It's a great book. And it has a really cool cover with a What is that? A CAT scan of an apple, and it's called Less Medicine, More Health. And if this is beginning to sound like another medical advertisement, um I want to be clear I have no financial interest in the book. All the royalties of all my books have gone to charities in the upper valley of Connecticut of the Connecticut River, which is my home states of Vermont and New Hampshire. It's because of all the money in medicine that I did this. I didn't want to be making money off of selling an questioning the medical profession.
So, let's deal with some assumptions. The first assumption is all risks can be lowered. And certainly medical care can reduce some risks, but the effort is most likely to be useful when the risk is high. Consider something where the chance of something bad happening in the next 5 years is really high, upwards of 50% without any intervention. Medical care can be very effective sometimes at reducing risk, may even be able to cut it in half, as is the case with treating really, really high blood pressure. But most of us aren't at really high risk. Our chance of something bad happening the next few years is a matter of 2, 3%. And it's hard to lower an already low risk. And all of our interventions pose some harms. And that's a very important thing to know. That's more relevant to a person that doesn't have that much to gain in the first place. And the importance of where you are on that baseline risk first became evident in the treatment of high blood pressure, hypertension. And on this graph, I'm going to show you the effect of treatment across the spectrum of diastolic hypertension. That's the bottom number. In terms of the annual rate of death or end organ damage. End organ damage is a fancy word for really clinically important events like heart attack, stroke, renal failure, and congestive heart failure. And I'm going to show you the untreated groups first. These are all randomized trials. People randomly allocated to treatment or no treatment. In very mild hypertension, this is what it looks like. In mild hypertension, this is what it looks like. Moderate and severe. And what you see here is not something that surprising. Who's at really high risk have something bad happen to them? Someone with very, very high blood pressure. I told you these were all randomized trials. This is the spectrum of baseline risk sort of from very mild to severe. And let's look at what happens in the treated group. There is the treated group. And you might say the yellow bars are treated. Why aren't they all exactly the same level? Because these are real data. Real data never perfect. But the main point of showing you this is to consider the benefit of being treated. And the benefit of being treated is moving from the red bars to the yellow bars. That difference, the so-called absolute risk reduction, is the benefit of treatment. Little bit of a noisy slide, so I'll take it away and let's just look at treatment benefit, that difference between the two bars. And what you see, who stands to really benefit from treatment, are the people at the extreme end of the spectrum of baseline risk. Well, this is a very important relationship in in medicine. I'm going to just sort of draw it as a curve, upward-sloping curve of increased benefit as you go up the spectrum of baseline risk. And we're talking right now about the spectrum between borderline hypertension and severe hypertension. And I told you that all of our treatments pose some harm. So, there has to be a harm curve uh here. And I'll start by drawing it here. And I'll I'll be clear, there's a little bit of uncertainty there, so I'll make it wiggle a little bit because we're not sure exactly what its slope is. And uh uh clinician in the room might point out, "Well, Gilly, you've got the wrong slope on that harm curve. If anything, it should be going downwards because who's most likely to experience an adverse event from starting blood pressure therapy? Someone whose blood pressure isn't that high to begin with." Well, we don't have to worry about that. My main point is treatment harm is more a fixed effect of treatment. It's less dependent on the spectrum of base baseline risk. It's more a function of what the treatment is. Well, if that's true, then clearly there's an area where there's net benefit, where treatment benefit exceeds uh treatment harm. But at the same time, we have to worry about the other extreme, places where there might be net harm, pri- primarily because there's so little benefit to be had.
Now, the importance of baseline risk is obscured by our tendency to describe the benefit of intervention in relative terms. And let's just see how smart this group is. You guys know how to subtract? This is going to be tough. I know it's Canada, right? Okay. Uh division, right? Okay, let's try it out. Let's say you're buying a car for $50,000. I won't ask why you're buying a car for $50,000, but you get it for $25,000. How much did you save? If you do the subtraction, how much did you save? $25K, that's right. If you do the division, how much did you save? How would we normally talk about it? You got 50% off. If you did the division, you did 50% off. So, the subtraction is the so-called absolute change. The division is the so-called relative change. Now, let's consider you're buying a Popsicle that lists for $2, but you get it for $1. All right, do the subtraction. How much did you save? You saved $1. You do the division. How much did you save? 50% off. And therein you see the problem. Right? You see the problem of relative change. That relative change provides no information of what you got 50% off from. While the absolute change tells you, it's a dollar versus $25. Now, I want to be clear. We don't offer a lot of 50% off sales in medicine. There are just not that many of them. We're lucky if we're talking 10 to 20% off. That's the typical relative risk reductions we're getting. And the question I'm asking you to ask your physician is, am I getting offered 10 to 20% off from here, or am I getting 10 to 20% off from here? It really matters. By the way, I'm going to add one more thing to this already complex curve at the top, which is the number of affected individuals. Turns out there not that many people at the very severe end of the spectrum, but as you move down the spectrum of baseline risk towards the more average, you involve a whole lot more patients. There are a whole lot more people here. By the way, pharmaceutical companies understand that curve. That's why they have such interest in expanding into moderate and mild risk populations.
Now, this is absolutely true not just with hypertension, but with diabetes, whether you're a borderline diabetic or have severe diabetes. And when I was in training at the University of Utah, our goal for managing diabetes was a hemoglobin A1C between 7 and 8%. That's a measure of blood sugar over the last two or three months. That was our goal, our treatment goal. But then a few years later, someone said, "No, no, no, no. We should make diabetics look like normal patients. We should get their hemoglobin A1C goal under 6%. That would be person without diabetes." Now, what is the effect of that kind of recommendation? Well, the first thing is you got a whole lot more patients. Right? If you used to have a hemoglobin A1C of 6.5, you didn't need to be treated, but now you do. Who are those new patients? Well, those new patients are the ones who arguably stand the least to benefit and are at the highest risk for net harm. That doesn't sound good, and it wasn't good. Here's the NIH 2008 for safety. NHLBI changes intensive blood sugar treatment strategy. That's not the right language for the press release. They stopped it. They stopped it. Why did they stop it? They stopped it because it increased the risk of death 22% absolute from 60 to 73 per thousand. It was statistically significant. It's not good to try to turn diabetics into having normal blood sugars. The assumption is all risks can be lowered. The disturbing truth is risk can't always be lowered and trying creates risks of its own.
Assumption two is it's always better to fix the problem, but that's not always true. Sometimes it's better to manage the problem of heart disease with good food, regular movement, and a few good medications. Was that a clap for movement? Does someone want to move? You can stand up if you want. It's better to do these simple things than it is to insert wires, blow up balloons, and place stents, which we do a lot of in my country. This goes under the language of PCI, percutaneous coronary interventions. Now, I want to be clear. If I'm having a heart attack, I'd like to have the problem fixed. Now, why do I say that? Because we have considerable amounts of data about the value of this intervention in people with acute heart attacks. This is 23 randomized trials, almost 8,000 patients, and this is a type of plot that is basically telling you the relative risk of death in the group that is undergoing PCI versus people that are not. And here's the summary estimate, and the way we would report it is the RR is 0.73, and we'd interpret that as a 27% reduction in the risk of death. Now, I just told you 27% fewer, that's a relative statement. You say, "Well, give me the absolute numbers." I'll tell you the absolute It's 7 versus 9% died. Now, maybe that doesn't sound that much, but let me be clear. This is in the next 30 days after your heart attack. I'll take the PCI, thank you. I think it's a good treatment for that. 65% fewer second heart attacks immediately following not a second heart attacks 3 versus 7%. Plus, PCI led to fewer strokes and fewer bleeds. I think this is a good treatment for people who have an acute heart attack. It's a better treatment than medicine to dissolve clots. But, most patients undergoing PCI in the United States have not had a heart attack. They're something different. 2/3 are elective procedures. Largely for patients with angina or chest pain, reversible chest pain. The trial that told us about this group is the so-called COURAGE trial, which part of it was done in Canada. It's about 2,300 patients with stable angina. That's chest pain with exertion. When you stop, it goes away. The intervention was immediate PCI versus optimal medical therapy. Optimal medical therapy is basically about controlling heart rate and blood pressure. And the findings at 5 years, 19% of patients had died in the PCI group compared to 18.5 in the medicine group. If anything, medicine looked better than PCI. That's not statistically significant. Angina at 5 years, it was 26% versus 28%. Again, not a statistically significant difference. Well, this should sound familiar to you. It comes back to this same curve. I think PCI for person actively having a heart attack is at one extreme. That's at the high end of the baseline risk, while PCI for angina is somewhere in there. And probably about a third of PCIs in the United States is for something less than that.
Anybody know who that is? It's George. How's he look? He wasn't born in Canada, was he? I shouldn't say that, should I? Okay. No, it's President George Bush and he's a very active man. He he clears brush on his Texas ranch. Here's a nicer picture. He's an attractive guy and he's a big he's a big mountain biker. I mean, he he runs. He's a very very active guy. He felt fine. He went for a checkup. He got a balloon and stent. Wow. What's interesting though is some American doctors were willing to say exactly what they thought of that. Here's writing for the Washington Post. It's worth noting that at least two large randomized trials show that stenting these sorts of lesions does not improve survival. Because Mr. Bush had no symptoms, it's impossible that he felt better after these procedures. Instead, George W. Bush will have to take two blood thinners, aspirin and Plavix for at least a month and probably a year. While he takes these medicines, he will have a higher risk of bleeding complications with no real benefit. And then this from the head of the Cleveland Clinic. This is really American medicine at its worst. It's one of the reasons we spend so much on health care and we don't get a lot for it. In this circumstance, the stent doesn't prolong life, it doesn't prevent heart attacks, and it's hard to make a patient who has no symptoms feel better. I couldn't say it any better. The assumption is it's always better to fix the problem, but the disturbing truth is trying to eliminate a problem can be more dangerous than simply managing one.
Assumption three is sooner is always better, and we've all been taught the value of early detection, particularly in a feared disease like cancer. But our understanding of cancer is changing. We're now beginning to understanding it's an incredibly heterogeneous group of diseases. And ironically, we've learned that because of the side effects from early detection. In the past, doctors treated a population, and those are represented here in those little red dots. And what we did is we waited for problems to develop, and then the fraction that developed problems, we made diagnoses and instituted treatment. The early diagnosis ideal was to take that same population, advance in time that group with the expectation that their natural course of events were what we found would be those destined to develop problems. And it seemed to make very good sense on the surface. But the reality was quite different. So, whenever we looked early, all of a sudden we found more patients. And now the so-called natural history, the natural course of events, must be more complicated. Hopefully, we've gotten those destined to develop problems, although we may miss them because some of the worst problems occur so quickly. But at least we've got a new fraction here. We've got those not destined to develop problems. And they are the so-called over-diagnosed and needlessly treated fraction. People treated for a problem that was never going to happen. Which brings me to the idea that we have to think about cancer differently. It's not just one thing. Not all the cellular abnormalities that get labeled cancer by our pathologists will ultimately matter to the patient. And that brings us to the barnyard pen of cancers. A way to think about cancer. And there're three animals in the barnyard. They're the birds, the rabbits, and the turtles. And the goal of early detection is to try to catch them early, to fence these animals in. But the birds have already flown away. These represent the fastest growing cancers. The cancers that have often spread by the time they're detectable. Early detection doesn't help those. Then there are the rabbits. They're hopping around and if you build enough fences, those are the cancers you can catch and may be a benefit to catch them. But then there the turtles. They're the cancers that aren't going anywhere anyway. You don't need any fences because they're not going anywhere. And the unfortunate reality is screening is really good at finding turtles. And the sad thing is doctors can't distinguish which animal's which. We tend to treat everybody and that means we're treating people for cancers that were never going to bother them.
Now here is an article from Life magazine. It was a really interesting article. It was called The Plea Against the Blind Fear of Cancer. Also written by a doc from the Cleveland Clinic by the name of George Cryle, and I just want to read one paragraph to you. In clinical practice to say that a person has cancer gives us little information about the possible course of his disease as to say he has an infection. There are dangerous infections that may be fatal. And there are harmless infections that are self-limited or may disappear. The same is true of cancers. Cancer is not a single entity, it is a broad spectrum of diseases related to each other only in name. Some of you don't even know what Life magazine is. Uh those of you who do remember, um would remember that it's been a long time since they ran black and white uh photographs. Uh this was written in 1955, which just blew me away. Uh that's the year I was born, um which is a rude reminder that I've never had an original idea in my life. Uh um Cryle was really ahead of his time and was probably the origina- originator of the uh birds, rabbits, and turtles analogy. It's an uncomfortable reality for us. There are a lot of turtles out there. We know that from autopsy studies of people who have died from something else. It may not sound like a lot, but 1 to 3% of adults harbor unsuspected kidney cancers. That's far more than we'll ever die from kidney cancer. At least a third of adults harbor small thyroid cancer. About a third of women age uh 40 to 49 harbor small breast cancers. And over half of men over age 60 harbor small prostate cancers. Cancer's just a part of life. And so when you see this amazing kind of headlines like this, molecular breast imaging nearly quadruples the detection rate for breast cancer, we've all been taught to think, "Boy, that's a really good thing." But now I want you to think about something else. Is that just finding more turtles? The assumption is sooner is always better. And while early detection may make sense in selected settings, particularly people who are at very high risk for the disease, its widespread adoption has been a recipe for making people sick. The disturbing truth is early diagnosis can needlessly turn people into patients.
Assumption four is it never hurts to get more information. And the value of information seems to be a central tenant of the 21st uh century. But it's a tenant that confuses data with useful knowledge. And it's really important to keep that clear because there's all sorts of data out there. There's no limit in how much data we could collect. Right now, uh we there are already thousands of bits of metabolic data we could measure on every individual in this room. And there's some companies like Theranos who would like to do just uh that. Uh there are millions of pixel of imaging data if we did cross-sectional imaging in every uh uh diameter that we could conceivably do, there's just millions of pixel on a single individual of things we could be looking for. And there are over 3 billion data points in your genome. There is no shortage of data we could potentially collect on patients.
I want to share with you a a case, um and this is a true story. It's about a friend of mine. Um and Bruce is a colleague who is an intensivist. That sounds like a personality disorder, but um it's not. It's a medical specialty. It is someone who runs intensive care units. And um he he understands how too much data can be a problem. The intensive care unit is probably the most uh data intensive setting in medicine where we're monitoring all sorts of things. You get a lot of immediate uh feedback. When Bruce had turned 50, uh he decided to be screened uh for colon cancer. Um and his primary care practitioner was happy to oblige, but he also wanted to do some other tests. And so he said to Bruce, um I think we should get a chest x-ray. Well, Bruce also happens to be a pulmonologist. He knows that's utterly useless. And the crazy thing he said, "No, no, you're not getting a chest x-ray on me. That's there's no indication for that." Well, let me do a cholesterol check. And Bruce said, you know, he had no heart disease. He felt fine. He said, "No, no, I don't want to, you know, I'm here to get colon cancer screened. Don't sell me on a lot of other tests." And then the primary care practitioner said, "At least let me get an EKG, electrocardiogram." Now, Bruce had a pretty good idea that uh he was in good cardiovascular shape. Uh he lives in Anchorage, Alaska. Uh he rides his bike to work every day. He does about 100 mi a week on his bike until the snow flies and then he's a big Nordic skier. He's a bit of an animal. I mean, he's a pretty exercise oriented guy. And he said, "Oh, I'm sure I'll have a normal EKG." Plus, he did not want to be labeled as a difficult patient. Even doctors don't want to be difficult patients, you know. All right, if you really feel you got to do it, all right, you know. Oh, okay. So, he gets his EKG and of course it comes out and goes flying off the chart up here. And they said, "That's it. Oh my, you got to get an echo." And the echo showed a thick cardiac wall. And now everybody's nervous about something and they say, "What what what you really need is a cardiac MRI." Well, now that's not that simple a test. This is an MRI of a beating heart. It's got to be synced. There's all sorts of decisions to make and then there's a whole question of how to interpret what you're seeing. So, Bruce wanted to get um you know, wanted to go to a place that did a lot of cardiac MRIs. So, he left Anchorage and he went down to the Mayo Clinic. And have any of you been to the Mayo Clinic? Yeah, it's a very shiny place. It really is. Um and it's um it's got a wonderful atrium and I don't know if you guys have an atrium here. If you don't, you're not a good hospital. You got to have an atrium to be a good hospital. Do you have one? Good. Okay. Cuz that way we know it's high-quality facility. And I'm making a little fun of Mayo and I know Bruce would want me to say one thing right away as an Alaskan physician, he really appreciates the support Mayo provides outline physicians with difficult cases. And as a patient, he thought everybody was incredibly uh kind and efficient and they were indeed efficient. They in the space of about 2 days, he had his cardiac MRI. He had a whole bunch of blood tests. He had his echo and his EKG repeated. He had overnight oximetry to see whether he was desaturating or not. He had this incredible exercise stress testing with physiologic testing which he passed with flying colors. So, Bruce sat down with the Mayo cardiologist after these 2 days of evaluations to learn the final diagnosis. Non-obstructive hypertrophic cardiomyopathy. Wow, that's long. It barely fits on the screen. And so, Bruce had a pretty innocent question. And his question is what should I do? And the Mayo cardiologist looked down, thought a little bit and said, "Well, why don't you come see us in 3 years?" Now, that was like not a very satisfying answer. Although, it's probably exactly the right answer. In fact, the American College of College of Cardiology and the AHA write this in their guidelines. It is recommended that comorbidities that may contribute to cardiovascular disease, e.g., hypertension, diabetes, hyperlipidemia, and obesity be treated in compliance with relevant existing guidelines. How many of you are lawyers? Okay, can you translate that for me or should I do it? It It really says do whatever you're going to do anyway with the patient. That's really There's no new information. Treat him like you would any other patient. So, there was really nothing for Bruce to do except worry. Now, he has this new um diagnosis. The guideline goes on to say, uh very encouragingly, most affected individuals probably achieve a normal life expectancy without disability or the necessity for major therapeutic interventions. Of course, it doesn't just stop there. It goes on to say, on the other hand, in some patients, hypertrophic cardiomyopathy is associated with disease complications that may be profound with the potential to result in disease progression or my favorite, premature death. Well, all of a sudden, biking wasn't quite the same. And this is why he related this story to me. Biking wasn't quite the same. Any sensation in the chest is that heartburn, nothing to worry about? Is it just a muscle spasm, nothing to worry about? Or is it cardiac pain? Hard to know. information. Oh, the Mayo cardiologist did have one suggestion, genetic testing. It turns out non-obstructive hypertrophic cardiomyopathy is associated with 1,400 mutations on eight genes. About half of them he could be tested for right now, and I suspect in another 2 weeks all of them could be tested in about 10 minutes. But then what? If Bruce tested positive, what could he do? He could worry more. He could worry both about himself and his four children. And of course, his children could be tested. So, they could worry, too, along with Bruce and his wife. Bruce said, "No. That's this is this is this is too much information. This is too much info I don't want that." And it's a perfectly rational choice. It wasn't going to change what they did. So, the assumption should be really rewritten. It's it never hurts to get more data. And to be clear, Bruce felt fine. And maybe all of you can see there's a reason not to just seek unspecified data while you're well. But certainly, you would always want more data when you're sick.
Well, let's consider that. This is a chest x-ray. You're looking at someone who has a pacemaker or defibrillator in. And it's a relatively easy measurement to say from this point on the pacemaker or the defibrillator to the lead, how well does electricity flow? Well, there's lung in between there. And if there's air there, electricity shouldn't flow very well. The impedance would be high. But if there's water there, electricity would flow well. Impedance would be low. A very interesting randomized trial of pulmonary impedance was done about 3 years ago. And the objective was to reduce readmission rates by real-time impedance monitoring. Ongoing minute-by-minute impedance monitoring. The patients were 335 outpatients with heart failure and implantable defibrillators. These are not well patients. These are sick patients. They already have an implantable defibrillator. The intervention is simply to turn the defibrillator's impedance monitor on. Everybody has the hardware. The only question was whether to turn on the software or not. And then patients with abnormal impedance were, and this is a quote from the article, "alerted with an audible tone." Now, how would you translate that for most people? "Alerted with an audible tone." You're a communications guy, what Siren? That's an interesting one. Siren. The word I would have thought of was was alarm, but I you know, I don't know, yeah. Your lung is now filling with fluid. Oh my god. Thank you, Siri. Your lung is now filling with fluid. All right, let's see how good you guys are. What do you think that did to clinic visits? What happened to them? If you had the monitor on, three times more. Three times more clinic visits in the group that had the monitor. Readmission rates, that's the outcome you're trying to drive down. What happened to them? Yeah, they went up. Statistically significant, 36% versus 22% after 1 year. Mortality? There was no change in mortality. In fact, honestly, there's a slight increase in mortality. Now, I shouldn't say that because it wasn't statistically significant, but it was going the wrong direction.
It's not just about impedance. How about searching for cancer metastases? Another study, very recent, looking for effort looking for cancer metastases early in patients who have cancer. This is not screening. And hopefully reduce their mortality. It was 1,200 men and women with colon cancer. And the intervention was intensive diagnostic follow-up, specifically looking for metastatic disease in the chest, abdomen, and pelvis. Chest, abdomen, and pelvis. It was basically a total body CT scan with a carcinoembryonic antigen. What do you think? Can we find metastases early if we scan people aggressively? Yeah, of course we can. What do you think that does to surgery rates? It increases them. What do you think it does to mortality? Absolutely no effect. In fact, it was actually a slight increase. More surgery. And by the way, these are the same findings for a similar trial looking for early breast cancer metastases 20 years earlier. More information is not a uniform good. Seeking information about early signs of metastatic cancer can be very scary for patients. You have a new cough? Some doctors say, "Well, maybe you have a cold. Come back and see." Others might say, "Wow, that could be a met. Let's get a chest CT." New ankle pain? Could you just started running? Well, some doctors say, "Well, let's just see what happens." Others might say, "Wow, that could be a met. Let's get a bone scan." Seeking information about early signs of metastatic cancer doesn't seem to help cancer patients live longer. But it does scare many and leads a few to live longer with the knowledge that they have incurable disease. They are subject to additional therapies and their toxicities earlier at a time they would otherwise be asymptomatic. The assumption is it never hurts to get more information. But the disturbing truth is data overload can scare patients and distract doctors from what's important.
Assumption five is action's always better than inaction. Now, in the interest of time, I'm not going to say a lot more about this. I usually talk about back surgery at this point. But all I'll say is our interventions may interfere with the body's ability to heal. And the disturbing truth is action is not reliably the right choice, and sometimes doing nothing is the right thing to do.
Assumption six is newer is always better. Really? How many of you want to beta test new software? Raise your hand. And then oh, you want to Yeah, okay. How many of bought a bottle a brand new car in its first model year and regretted it? I I bought a 1994 Explorer. What a dog. Um Anyway, um let's talk about a new procedure quickly or was a new procedure in 1990. This is autologous bone marrow transplantation for women with metastatic breast cancer. Early enthusiasm here, the Los Angeles Times, breast cancer survival rate may double study finds. Uh 10 years later there was a sobering reality. Uh to a reasonable degree of probability this form of treatment for metastatic breast cancer has been proven to be ineffective and should be abandoned. Now, a lot happened in between. Over 30,000 American women who had metastatic breast cancer went through one of the most difficult therapies with extreme side effects including uh death before we realized it was not a good therapy.
How about a new device? Well, here I'm going to show you um a a a a plain X-ray of of the pelvis. Um and I want to be clear I'm not breaking any American uh HIPAA regulations. This is my pelvis, so it's okay. Um there's my prostate. You didn't think I'd show you that. Um Actually, it's just a little bit bigger than that, but I don't But I'm not here to talk about prostate cancer. I'm here to talk about hips. And talk about a implant, a new device, the metal-on-metal hip. Now, one thing to know is orthopedic surgeons have a lot of hips to choose from. A really lot of really frothy private sector market with all sorts of hips with different equipment needed for them. Really a lot of hips. But the basic thing I want to show you here is what was the standard of care, which is the hip you see on the left with the plastic liner. And the new innovation was the metal-on-metal hip. And the one thing you can see about the metal-metal hip, because it didn't have the plastic liner, the ball could be bigger. And this had some theoretical advantages, increased range of motion, decreased risk of dislocation. And the plastic liner did have some problems. It would wear. So, after years of use, plastic would wear and little bits of plastic would break off and get into the joint. But plastic is pretty inert. They just sort of sit there. Didn't cause a particular problem. But anyway, there are there was a theoretical advantage and so we went in big to metal-on-metal hips. Now, what is wrong with this picture? Do you want a metal-on-metal hip? No. Okay, that's good. Yeah, because it fails at seven times the rate as the metal-on-plastic. Now, you may already know this, but what happens is instead of having little plastic pieces in your joint space, you have metal pieces. And metal is highly reactive. And in some patient, it leads to a very profound inflammatory response, which is called a pseudo tumor. It's not a cancer. It's just a really painful inflammatory uh, response to the metal particles.
Now, I'm a proud American. Um, is that obvious? Okay. Uh, and I was born in the United States. I should be clear about that. And I'm not running for president. Um, and as a proud American, the one thing that really upsets me about this, uh, data is these are not US data. We probably have more experience in putting in metal-on-metal hips than any other country in the world, but the data I'm showing you from is that from this tyrannical power that taxed us without representation. And it's like an embarrassment, some third-world socialist country. Why? Because they can track what's going on in their health-care system. Because they can measure system performance in a way we cannot because we have so many payers involved, patients change payers all the time, some go out of business. We are beyond the ability to actually measure our own health-care system performance. And for me, as a researcher, that's like the number one argument to moving for a single payer is simply to have a single data system. Okay, enough politics for that. Then we got metal ions in the blood. Now, that can't be good. And ironically, on the same day about 2 years ago, the New England Journal and The Lancet both reported some of the problems with having so much metal in your blood. I'll tell you about one. There's a woman who's in late 50s and she started to develop congestive heart failure. Became very severe congestive heart failure. She's only in her 50s and she underwent cardiac transplantation cuz her heart was basically dying. A few weeks later, the transplanted heart began to fail. Well, that doesn't usually happen. This is coming from, like a young person who's killed in a car accident and that heart starts to fail. This really throws off the doctors. That shouldn't happen. What's going on? Until someone went through the charts said, "Boy, she does have a metal-on-metal hip implant." They measured her serum cobalt. It was 300 times normal. Don't miss what's happening here. This is one device causing two hearts to fail in the same patient.
Finally, a new drug. Anybody remember that drug? Incredibly hyped up drug. 10 times more expensive than an every bit as good agent until it's pulled off the market because it increases the risk of cardiac death two to three-fold. I love this on the internet though here. Vioxx, arthritis pain alleviated even if it kills you. I don't know how many of you uh remember Huey Lewis and the news. You remember this? You must be pretty old though. Sorry. I don't want a new drug. I want an old drug. Right? I I don't want to start with a new and I don't want a procedure that's brand new. I want a procedure that's known to work and I don't want a new device. I want a device with a proven track record. The assumption is newer is always better, but the disturbing truth is new interventions are typically not well tested and often wind up being judged ineffective or even harmful.
Well, the final assumption is it's all about avoiding death. And it's important to be clear here. We're moving into value judgment territory. There is no right answers here, but I'm going to at least spend a few seconds telling you about my values. Now, many of us were drawn to medicine because of our interest in saving lives. And this is the second picture of of me I'll show you. This is in 1975. I'm no different. I was an ambulance attendant. I loved running squad in my hometown of Boulder, Colorado. As you can see, I thought I was a pretty cool guy. Some of you may note the pocket protector sort of gives away that you weren't really that cool a guy. Uh and you know, it was all about avoid and I have to tell you cuz I also like music that same summer as I was working this song by Elton John came out. We'd open up the windows and drive around. And then I was all for it. I said, "What a great song." So, this is the way we think about saving lives and avoiding death. But these are words that really are are mislabeling a lot of what's happening or or they represent far less evocative terms, terms like lengthening life and delaying death. And it's probably important to begin to think about some of the distinctions before how we talk about this. And it's important to ask a question whether length of life is the only variable. Often in medicine we act, at least from our studies, that this is the variable that will trump everything else, no matter how big the difference or what kind of life is involved. But it's something that's really important for patients to think about.
It's all about avoiding death. Is that really really true? And if that was true, you know, why don't we lower the national speed limit to 55? Uh this is miles per hour. I'm not suggesting a killing me. That would be pretty uh severe. But I I I know I I know you wouldn't be a fan of it uh in Alberta, right? I know it wouldn't be in Montana, either, you know? I mean, it did it did reduce uh death rates, and uh people clearly would rather have higher speed limits. This makes a difference in in in anything. So, there's a place. We've made a trade-off already. Uh why do we climb mountains? This is the Alaska Range, and one of the best trips I ever took. That's my wife in the You know, but clearly, if it was all about avoiding death, we'd never put ourselves in any sort of uh danger or place we couldn't be helped. And and and and why do we swim in the ocean? And full full disclosure, I I I I I love uh to swim. I love to swim in natural bodies of water. I love getting in any I'll get in almost anything. Um and so, I went to the CDC and and just said, you know, how how many swimming-related deaths, not boating-related, just swimming-related deaths do we have in the United States each year? It's over 3,000, and half of them are in natural uh bodies of water. Now, what should our policy response be? Should we close the beaches? No, I don't think so, because I think you also understand the disturbing truth. The disturbing truth is a fixation on preventing death diminishes life. And so, we have to find a balance, or that's not the only thing that's relevant.
So, I want to conclude again by emphasizing medical care can do a lot of good in selected settings. That's particularly true in people who are acutely ill or injured, what might be dubbed acute care uh medicine. But, at the same time, in many other settings, we've exaggerated the benefits of medical care and understated its harms. So, my question that I want to leave you with is communicating this nuance part of our job. Thank you very much for your attention. I'll be happy to take some questions. We have time for a few quick questions. I know that some people have places to go at 1:00. Does anyone have I have a lot of questions, but I want the audience to have a chance. Yes, please. Hey. Um yeah, um thank you very much for this. I um We do a lot of screening here in Canada. It's publicly funded. Um so at the moment I have a mammogram requisition sitting at home. Mhm. My physician told me it's up to me because there's a lot of controversy right now. Um I'm not a fan of mammograms and I have significant concerns about whether it's a worthwhile thing to fund. Also, occult blood, only 9% of uh colon cancers are actually found. And we we use screening between 50 and 70 years of age.
So, well, uh, yeah, uh, it. So, let me just make a couple of general comments about screening, which has been a large uh part of my work.
Um, I think in general, we've totally exaggerated the benefit of population-based screening, and we've uh totally ignored uh uh uh the fact that it leads to some harm. The most prominent harm, particularly in breast and prostate cancer screening, is the problem of overdiagnosis, that is, identifying cancers that actually were never going to be a problem. But because we don't know which they are, we treat everybody, so we're treating some people unnecessarily. That's a substantial harm, um, and it's more frequent than uh than any doctors would like, and we're we're only beginning to get get a handle on that.
That said, I would say all, almost virtually all screening in general will will have a mixture of effects. It will produce some benefits, and it will produce some harms. And um, we've, in my country, we've sort of made a huge distinction between prostate cancer screening and breast cancer screening. We say prostate cancer screening is bad, breast cancer screening is good, and the truth is they're really not that different um, and that there are real choices here, and the trade-off is a small potential to reduce the chance of a breast or prostate cancer death. It really is quite small, and it's getting smaller all the time as our treatments for the diseases get better uh against the more likely problem of being over-diagnosed. Still not a very, very common, but it is more likely um, and there's no right answer to that question. That's a, that's a calculus that different patients can look at the same data and come to different conclusions about.
So, I've always said we've argued like screening was a public health imperative. That is not right. It is a choice um, and it's a choice that needs to be informed um, and it needs to be informed with data. Part of the issue is psychologically, I feel like if I don't do this, I'm asking for it. Yeah, and and and I, and I'm sure there's a lot of of pressure uh coming from the provider community and the or in the larger health community that sort of communicates that message, whether it's intended or not. And that's a shame. And in my country, because doctors are actually being measured on how much mammography they do, there have been women who decided they don't want the test and they've been told they have to leave the practice. Now, that is just totally crazy. That's coercion um, and that's just so wrong. Th- Th- These are really choices uh, they're they're they're not imperatives.
Few more quick questions and and quick responses, please. Yeah, quick. No, no. That was brutal. Good point though, Tim. I'm I'm with you. I can do it. I'm continuing with the easy questions. Oh, good. As as a British intensivist over here. Oh, oh. Th- This personality disorder, yeah. Yes. Um, in answer to your question, definitely. Uh, verbal dexterity is every bit as important as manual dexterity. It's communication. Yes. You want to follow my job? Has to be. Just thought, but it has to be, it has to be. But giving you an opportunity to communicate in return, presumably one of the graphs you could have put up was costs in the medical system or even number of doctors. Yeah. Oh, sorry, lawyers versus costs. Right. Versus necessary tests. Uh, doctors are guilty of hitting the target, missing the Right. Pharmaceutical companies are guilty of treating the worried well because that's where the money is. Right. But you've got a room full of lawyers. Mhm. Part of this is being driven by the legal profession as well, and it's lovely to sit back and have a smirk at others. What should they do? Yeah, they should clean up their act.
Um, it it, but but let me just in more more seriously say that, you know, what you've described, I think, is absolutely accurate. It's really a complex web of forces at work. Um, part of it is the sense that we are punished for uh under-diagnosis, but we're never punished for over-diagnosis or over-treatment. So, we're not stupid, you know, if we see an unbalanced set of incentives, we we tend to go one way. And then then uh in a fee-for-service system, which Canada and the US both predominantly rely on, it also becomes in your financial interest to always uh do more. And then you have a frothy uh private sector that produces this stuff, whether it's tests, drugs, or devices, and you have a perfect storm uh that that that tends to uh drive utilization. And it's not just about money. I mean, I I'd say if you had all the money in the world, there's still a very important reason for each of us to ask ourselves the question, how much medical care do we want? Because the forces are so lined up to deliver us more that I think it's really important for all of us, not because of money, even if you had all the money in the world, what kind of life do I want to lead and what extent do I want to place myself at risk by getting involved with medical care when I'm fine right now? And that's, these are questions for all of us. And that wasn't short. It was a good one though.
One more. Go ahead, please. Okay. Thank you so much for asking that question because the the real key, and that's why I say in selected settings, the real key here is who's most likely to benefit from an early detection effort? Someone who is at extraordinarily high risk, not just a little bit, not 10%. I'm talking five, 10 times higher risk. That's the group that stands to benefit the most. It It's the more extreme. And so clearly, someone with the BRCA1 or 2 mutation is at somewhere between five to eight times higher risk of developing and dying from breast cancer or ovarian cancer, is probably even more important. They're the people where screening makes a lot more sense. They have a lower risk of overdiagnosis and more higher risk of finding consequential disease. And that's the nuance. It's not like all this is always bad. It's not like it's some Luddite up here saying, "Gosh, I wish we never developed abdominal CTs." No, they're hugely important for helping us decide what's going on with sick people. It's the nuance of just because we have good technology doesn't mean it's good when it's distributed totally widely. It can cause problems.
One sentence on Biden and Obama's moonshot. Oh, I hate moonshots. One sentence. I just did. That's it. I hate moonshots. I hate moonshots. I think that's a full sentence. I hate moonshots. It just so exaggerates what we can do and It suggests that that is is so unidimensional. And and it and it and it makes the public think that it's only a matter of uh of will. And it's it's more a matter of time.
Well, I would like to thank the wonderful Justice Allan Rockard. Thank you for having me, Justice. I would like to thank uh Gil for coming to Alberta and giving a truly spectacular uh presentation that I think has relevance to the health care system. It has relevance to lawyers. It has relevance to the to uh physicians. And really to to uh the Alberta uh citizenry. So, thank you so so much. And I hope you come back.