At the end of each episode, I ask you to share your thoughts about the show and send me any questions you might have. Well, here’s a question we got from a listener named Max:
MAX: Hey Bapu, this is Max. So, a few weeks ago, my father had a heart attack and triple bypass. This was a big surprise because he’s never had any signs that his heart was headed in this direction. What gave it away was an angiogram that showed 90 plus percent blockage in three of his arteries. My question is: why is the angiogram not considered a more routine procedure for heart health? Is it just too expensive to make part of a normal physical? Is it scarce resources? And can I actually make it part of my routine physical so I can monitor my heart in a more detailed way?
Max’s question is a good one. And it’s relevant too: heart disease is the leading cause of death in the United States. According to the C.D.C., it killed nearly 700,000 people in 2020, which is about one in five deaths. And it’s costly — by some estimates over $200 billion a year. Heart disease should be on all of our radars because there’s a good chance that it’ll affect us or someone we love. So, how can we figure out if we have heart disease? Do we ALL need an angiogram, the kind of test Max’s dad got? And what is an angiogram, anyway?
From the Freakonomics Radio Network, this is Freakonomics, M.D. I’m Bapu Jena. I’m an economist and I’m also a medical doctor. Each episode, I dissect an interesting question at the sweet spot between health and economics. Today on the show, testing is an essential part of medicine. It touches every specialty. But deciding who needs a test — and when — isn’t always as clear as it seems.
REDBERG: I was a second-year medical student, and I started questioning the tests. Before that I had just assumed that everything that doctors ordered in the hospital, was for a good reason.
We’ll talk with Dr. Rita Redberg about who and when to test for heart disease, and how her views on testing have changed over time. And my friend and colleague, Dr. Bobby Yeh, will help answer Max’s question.
YEH: It’s a really, really important question. You know, we’re talking about screening tests. And what is a good diagnostic or screening test for patients who may be at risk for coronary artery disease?
REDBERG: It started really, I think, as I was growing up in Brooklyn where, both of my parents had to drop out of high school in order to work. That was during the Depression. I grew up in fairly modest circumstances where we didn’t waste a lot. So, I would say it was drilled into me not to waste things.
Dr. Rita Redberg has been a cardiologist for nearly forty years. She’s a professor at the University of California San Francisco, and she’s also the editor of the medical journal JAMA Internal Medicine. Over the course of her career, she’s become known as a champion for the “less is more” approach. In cardiology, that often means focusing more on prevention — things like lifestyle changes and important medications — and less on invasive tests or procedures.
Tests can tell us a lot, and in cardiology there’s no shortage of tests. So, when Rita or any other doctor sees a patient who has heart disease, or who’s at risk for it, they have a lot of tools they can work with. Tests are important because they can often give doctors valuable information about a person’s health. This might lead you to think that more testing is always a good idea — the more information the better, right?
Well, it’s complicated. Some clinicians and researchers, like Rita, think we test too many patients too often — we over-test — and they see it as a real problem. Over the last decade or so, a lot of studies have highlighted the issue of over-testing in cardiology. Earlier this year, the American Heart Association actually published a new scientific statement calling for a reduction in “low-value” care, which are services that provide little or no benefit to patients. Some of that is testing. An estimate from 2019 suggested that low-value care comes at a high cost: close to a $100 billion dollars each year.
But financial costs are just one part of the story.
REDBERG: I think in general, we have a feeling that more information is always good. and information comes from a test. No one is going to say, why did you order this test when the patient didn’t need it. They’re much more likely to be worried about someone asking them for a test result that they didn’t order. And so, there’s very little in our current system that disincentivizes someone from ordering an extra test.
JENA: Part of it’s also that doctors don’t face the cost of the test. So, they may or may not get paid to do the diagnostic evaluation, for example, but if a P.C.P. is ordering a lab test, they’re not typically going to get paid for it, but they also don’t, see any of the costs that the patient would face or the insurance company would face. So, that seems to me like another reason why we might see more testing occur than otherwise should occur. I’m curious as to whether or not culture plays a role here.
REDBERG: Oh, absolutely. I think culture is a big part of why we order a lot of tests. It’s become part of our culture and certainly not enough people asking the question: how does this test help in management? I mean, it’s very hard to change that medical culture of “a test can’t hurt.”
JENA: Can you tell me a little bit about how tests can hurt? Like, what are the various ways that getting a test can be harmful to someone or to the system?
REDBERG: There are a lot of ways that a test can hurt. Of course, some tests are needed and helpful, and I fully support those tests, but it’s important to look at before ordering the test: what is the information you’re going to get from this test and how will it be helpful, in the care of the patient? And if you can’t come up with a good answer, or if you could have made the same decision or choice without the information from the test, then I would consider not ordering the test.And then the harms that can come from a test are — depending on the kind of test, of course, any kind of C.T. scan or nuclear scan has radiation and there is a cumulative effect of radiation, and there is an increased cancer risk. And in fact, the Institute of Medicine report a few years ago on breast cancer identified medical imaging as the No. 1 preventable cause of breast cancer. And then there’s – well, the tests are never perfect. So, you could get a result from the test that isn’t accurate and could be a false positive.And then you’ve gone down some rabbit hole of figuring out, is there something really wrong with you or not? Particularly, I think it’s a problem for screening tests. It’s tests for people that feel perfectly well, but just want to be reassured that they’re perfectly well. And so, then you’ve taken someone who really did feel fine and now they’re worried. They’re not sleeping that well at night, they’re worried about that calcium in their coronary artery, for example.
JENA: if I knew that every time I went to the doctor, he, or she would want to do a test, and there’s some possibility of them finding out something that was never going to affect me, but because they tested for it, they found it out and it would worry me, I might be more reluctant to go to the doctor. And I wonder if this idea of over testing has implications, not just for anxiety that might come about from people, hearing about diagnosis that would never have affected them, but it goes even further than that in terms of the relationship that a patient has with their doctor, with the healthcare system.
REDBERG: That’s a good thought. It could certainly introduce this element of worry into the doctor-patient relationship because it’s a pretty power imbalanced relationship. And if your doctor is ordering this test and you’re going to worry about these results and that is not good for patient’s peace of mind.
REDBERG: It’s really hard to make an asymptomatic person feel better.
JENA: Yeah. That’s true. That is true. What’s sort of your framework for thinking about who needs a test, who should get a test? And let’s focus on cardiology.
REDBERG: First of all, I would say they should have symptoms. Like, if you want to prevent a heart attack, that is great, but I can tell you a lot of ways to prevent a heart attack and none of them are going to involve having a test. I think, we have the most, bang for our buck with testing when there’s something specific you’re looking for and it’s based on, a history and physical and the person came to you with symptoms. There are really few tests that are going to make a healthy person live longer. Like, if I’m going to say, you know, you should eat healthy and exercise more, we hardly need a test to say that.
JENA: A few things that could happen when you have a test — you could have a false positive. But you can imagine if you have a person who is, overweight, maybe they smoke. And so, they’re at higher risk for heart disease than someone who doesn’t have those risk factors and they get a test and the test looks actually okay. You could imagine that would be falsely reassuring. So, that’s almost, a false negative. Have you ever seen situations like that occur where people’s behavior changes because they are, maybe falsely reassured because we’re putting so much emphasis on the diagnostic test, as opposed to all of the other things that we know epidemiologically, physiologically affect the likelihood of, let’s say, heart disease?
REDBERG: I would say I’ve had colleagues who will remain nameless that I know got a coronary calcium test, for reassurance, and their score was zero and gained like 10 or 20 pounds in the next year. And I think that’s because you get this kind of feeling, well, my score was zero, so I don’t have to worry about what I eat or exercise every day.
REDBERG: And of course, it’s not true. And so, that’s a definite downside to the test.
JENA: One thing that we talked about earlier was the reasons why over testing occurs. And there’s certainly physician factors. Why is it that patients decide to get tested? And again, let’s focus on people who don’t have symptoms, right? Because if you have symptoms, you want to figure out what’s going on. So, why is it that patients who don’t have symptoms get tested?
REDBERG: Sometimes somebody in their family may have just had a heart attack or some heart problem. And so, now they’re worried and they want to be reassured that that won’t happen to them. Or I sometimes have patients who’ve just told me their friend had a cardiac C.T. test and told them it was really good and they should get one, too.
REDBERG: Some people have perhaps a mistaken idea of like that family history is a risk factor because when we talk about family history, we’re really talking about premature family history, like getting heart disease before age 45 for a man or 55 for a woman, because, as we started out saying, heart disease is the leading cause of death in this country. Most people are going to die of heart disease and having your grandparent die of heart disease at age 75 is just not a risk factor for heart disease for you.
JENA: Rita, what are the most valuable cardiac tests for screening for heart disease and how would you rank them in terms of their utility, you know, cost, effectiveness, things like that?
REDBERG: Screening for heart disease means taking someone who’s asymptomatic and doing a test to detect something. And like, there’s really nothing, good that can come out of that test. We have a lot of different risk scores, the Framingham risk score, Reynolds risk score. You put in your age and, blood pressure and do you smoke an, cholesterol. And those will give you your risk over the next 10 years. Obviously, that’s not a test so much as a risk assessment, but the screening tests, because they would be done on asymptomatic persons, they really are not helpful. There is nothing that we know of that we can do for someone who’s asymptomatic that’s going to make them less likely to have a heart attack, or likely to live longer based on a test.
What about someone like our listener Max, though? After his dad’s health scare, he’s worried about his own cardiac health. As far as we know though, he doesn’t have any symptoms of heart disease, but it’s natural for Max to feel nervous. Coming up after the break: cardiac tests are more advanced and more available than ever before. Are they actually helping doctors find and prevent heart disease?
YEH: If you did angiograms on everybody over the age of 60 or 70, you’re going to find a large number of people with these 30, 40 percent blockages that may never actually affect the patient in any way.
I’m Bapu Jena, and this is Freakonomics, MD.
* * *
In his question to us, our listener Max wondered whether he should get an angiogram to assess his own heart health, because of what happened to his dad. He also asked why this test isn’t something that people just get done routinely.
An angiogram, also known as a coronary angiogram, is an invasive procedure that uses contrast dye and x-ray imaging to illuminate the blood vessels of the heart, to see if there are blockages or any other problems with blood flow.
Max doesn’t have any cardiac symptoms that we’re aware of, but like a lot of people, he’s anxious based on his family history. An angiogram could help him find out if any of his coronary arteries are blocked. But Max could be putting himself at risk for complications like injury to the blood vessels, an allergic reaction to contrast dye, even stroke or a heart attack. These aren’t common but they can happen, whether you have symptoms of heart disease or not.
But even in people who do have appropriate indications, angiograms aren’t perfect. A 2010 study published in the New England Journal of Medicine found that in patients who were suspected to have heart disease but hadn’t been diagnosed with it yet, coronary angiography revealed obstructed coronary arteries only 40 percent of the time, and that another 40 percent of patients who underwent the test had essentially clear coronary arteries. One takeaway from the study was that a lot of people who were suspected to have heart disease didn’t.
And that brings us to an important discussion. How do we evaluate someone who actually has symptoms of heart disease? How do physicians decide which tests to use, and in what order? And what could be the risks of doing a test if it isn’t medically necessary?
YEH: The first thing we are trying to sort out when we hear that a patient has chest pain is what is the character of that chest pain, and how likely is that chest pain to be attributable to blockages in the coronary arteries or other etiologies? And that can actually be a very complicated undertaking.
That’s Dr. Robert Yeh, but I call him Bobby. He’s a friend and colleague, and he’s also an interventional cardiologist at Beth Israel Deaconess Medical Center in Boston.
Most of the patients that Bobby treats have chest pain, which may be an indicator of an acute coronary syndrome or heart attack. So, the nature of chest pain seemed like a good place to start in terms of how Bobby assesses patients. There are lots of non-cardiac causes of chest pain, like acid reflux or muscle or bone injuries, but because heart attacks are life threatening, doctors often treat chest pain as cardiac until proven otherwise.
JENA: So, someone comes to their doctor or goes to the emergency department. They’ve got chest pain. What’s sort of the workup for that?
YEH: In some patients, heart disease is a stable, fixed, certainly slowly progressing narrowing of the heart arteries. In other situations, heart blockages can arise because there are ruptures of small plaques in the arteries that can cause severe, rapid narrowing associated with clot formation in the heart. In that situation, the clot formation in the heart is what can lead to a heart attack. So, I think the first thing to really separate when one thinks about chest pain is whether or not this is sort of that stable type of chest pain that’s happening with exertion or it’s this acute, unremitting chest pain that is happening sort of unexpectedly, more acutely, not associated with exertion at all. Those patients we need to act quickly on. And generally, what we’re doing is we’re checking an electrocardiogram or an E.C.G. or an E.K.G., sometimes called. And we’re looking for signs of damage or stress on the heart. There are characteristic patterns on the E.C.G. which would say to us there looks like there’s a significant blockage of the heart that needs to be addressed immediately. And when we find a blockage like that, we are rushing those patients to the cardiac catheterization laboratory for a coronary angiogram, which is just a picture of the heart arteries using I.V. dye under x-ray. And then, if we find a blockage, we’re often addressing that with angioplasty or stents.
JENA: And then, in the case where the person has more stable chest pain, what’s the evaluation there?
YEH: There we are still often checking an E.C.G. at baseline because it’s good to know what that looks like., and then secondarily, we’re often sending those patients for what we call a test for functional ischemia, which is basically a stress test. If they can exercise, we get them on a treadmill. We slowly ramp up the amount of exertion that they do, and then check progressive E.K.G.s during the course of the stress test, as well as blood pressure measurements. And sometimes what we find is their response to exercise is entirely normal, in which case our concern is sort of alleviated. And other times, we find E.K.G. changes or E.C.G. changes, which are indicative of the heart undergoing some sort of stressor.
JENA: Suppose you do that test, and you see that there’s evidence of what you called reversible ischemia or reversible pain. So, if you slow down, the pain goes away. What do you then do for those patients who now have documented heart disease?
YEH: It’s a terrific question. And it’s one where I would say that there’s not a clear consensus necessarily, and for which the evidence is changing. I would say historically, what we’ve done for patients like that is we’ve started them on some medications, like a baby aspirin, maybe a medication like a beta blocker and a treatment for elevated cholesterol. And then, we’ve sent those patients to coronary angiography to look at the type of anatomy that we have. But more recently, we are sometimes managing those patients medically. We’re presuming that they have coronary artery disease, and we’re starting those patients on medications without doing further testing. And then, a third test, which I think is increasing in use is doing a C.T. — or a cat scan — C.T. coronary angiogram, which is a noninvasive way of looking at the blood vessels.
Based on Max’s description, his dad’s case was fairly serious. He had blockages in three blood vessels of the heart and required bypass surgery to open them up. Again, we don’t know if he was symptomatic before, but according to Max, the angiogram was the test that gave it away. And you can’t just walk into the doctor’s office and get an angiogram because you want one, so something must have tipped off his dad’s physician that this type of invasive test was needed. But does that mean all high-risk patients need or should get an angiogram?
YEH: You know, it’s a really, really important question, and certainly one that many people are thinking about. We’re really talking about screening tests. And what is a good diagnostic or screening test for patients who may be at risk for coronary artery disease? Right now, the way that we manage those types of things is to look at their set of risk factors. But generally, a high-risk score doesn’t lead one to recommend a coronary angiogram or a stress test. What really drives those tests is the presence of symptoms, symptoms related to exertion. And patients for whom they have no chest discomfort, even those who have risk factors, but no chest discomfort whatsoever, nothing to tip you off to think that they have cardiac disease — continuing down the pathway of testing certainly will lead to a lot of potential unnecessary testing. You know, you find somebody who has no chest discomfort, but you’re worried for whatever reason. You do a stress test — these are not perfect tests. They have false positive rates. It shows up as a positive test. And now, this patient is sent for coronary angiogram. Now, on the angiogram, it is not a benign test. It’s an invasive test that has complications associated with it. Maybe you find there’s a 30 percent blockage or 40 percent blockage in the heart artery. In all likelihood, that 30 to 40 percent blockage won’t actually lead to a heart attack. If you did angiograms on everybody over the age of 60 or 70, you’re going to find a large number of people with these 30, 40 percent blockages that may never actually affect the patient in any way. And then, you’re left with a question. Well, should I stent that 30 to 40 percent blockage? By and large, our answer is no. Because our evidence suggests that stenting those blockages actually doesn’t prevent heart attacks. So, the right treatment for those blockages to prevent future heart attacks is actually giving statins and controlling cardiac risk factors. But that was the treatment that we would’ve recommended at the start before we went down this cascade of diagnostic testing. What you ultimately find is really unlikely to change your treatment decisions, unless you were to find something that is quite extreme.
JENA: In Max’s case, absent any symptoms, but just understanding that his family history. there would be then no reason for him to get an angiogram or C.T.A. or stress testing. Just thinking about his particular question, none of those things should be really on his radar.
YEH: I don’t want to be too flippant about my recommendation here. But that is true on the population scale, that there would be no reason in our current understanding of the evidence to do further testing And that’s because although Max’s father was extraordinarily unlucky to some extent, that if we did testing on all patients like Max’s father, the vast, vast majority of them would have no evidence of something on their coronary angiogram that would lead us to make a different decision. And some of those patients might suffer a complication or an unnecessary series of further testing, that would not change their further management. So, there is this conundrum that I think science is trying to figure out. Certainly, we’re actively trying to figure out if we can sort of prophylactically treat blockages that we find, if we can find blockages that are “vulnerable,” more likely to rupture, even in patients who are asymptomatic. But as of now, we haven’t found that sort of holy grail to prevent future cardiac events.
So how do we find the right patients, without overtesting the wrong ones? We may never get it perfect, but we can try to get closer. Here’s Rita Redberg again.
REDBERG: I think there are a few things that physicians and organizations can do to improve the problem of over testing and overuse. First of all, talking about it, putting it in our undergraduate medical education, our graduate medical education, our quality measures, our boards, is really important because it really is a quality measure. Changing the culture and sort of talking more about, well, why did you order this test? When I am on service, I explain to the house staff how I think about testing. And then say, I realize this, isn’t maybe what you heard last week from the last attending, but I’d like to talk about tests that you want to order so that we can be sure that these are tests that our patients will be better off with.
REDBERG: Another thing that I think increases testing particularly in the U.S. is that a lot of these tests are what we call supply sensitive. So, once you’ve bought an expensive piece of equipment, like a cardiac C.T. scanner, you want to use it.
JENA: I’ll tell you, you used the word supply sensitive, I say Dorito sensitive, because I have a bag of Doritos in the house they’re going to get eaten.
REDBERG: The more of these pieces of equipment that are sold, the more they’re going to be used. And so, some change in policy that disincentivizes that would be helpful.
JENA: We actually had a study you probably remember, years ago in, JAMA internal medicine where we looked at doctors who focus on inpatient medical care, and we try to profile each doctor according to their just general costs. So, some doctors tend to order more tests and procedures. They order more consults from other physicians, they get more imaging. So, there are some higher cost doctors and there are some lower cost doctors. And we didn’t find any relationship between, how much money they tend to spend on each inpatient on average, and the mortality rates of their patients.
JENA: And so, you could look at that and you would say, “Well, that suggests that the higher spending doctors, aren’t getting any better outcomes. But, you might then be tempted to say, “Well, we can then make those higher spending doctors spend less without having any impact on mortality.” And that next jump isn’t technically correct, because if those doctors who are spending more and getting the same outcomes are deficient in some other way, then if you eliminated the extra consults, the extra imaging, they might actually have worse outcomes. So, this question of how much healthcare use can we reliably reduce without having impacts on patients — it strikes me as a difficult one
REDBERG: I think if we and the F.D.A. demanded to have evidence before we start doing something it would be really huge for the profession because once something becomes established in the culture, it’s a lot harder to change practice, even when we know in our hearts we are not helping patients with this procedure or test.
If Hamlet were a physician, he might’ve said “To test, or not to test. That is the question.” There are a lot of things that push us towards overtesting in medicine, but it’s important to remember that under-testing is a problem too. It’s easy to imagine a situation where someone might be harmed if they didn’t get tested for a condition that could’ve been treated, if caught early enough. It just makes sense. It’s harder to picture a situation though where a test caused more harm than good. But it happens.
During our discussion, Rita told me about a case study that was published in the Archives of Internal Medicine in 2011. A 52-year-old female patient with a history of high blood pressure and mild obesity went to the hospital because she was having chest pain after starting a new diet and exercise routine.
REDBERG: She noticed the sharp pain. And she went to her doctor because she had heard that women can have, you know, funny pains and that could be heart disease. And the doctor examined her and it was normal
REDBERG: But then the doctor said just to reassure you, we’ll do a coronary, C.T. test. Well, the C.T. had calcium, so they couldn’t really see her coronary arteries and they decided, to take her to the cath lab and do coronary angiography. So, they did the cardiac cath and had the unfortunate, but known complication of a left main dissection. So, she had a dissection, a tear in her coronary artery, which required emergency surgery, because that is a life-threatening complication. By the way, her coronary arteries were normal. But now she had to have bypass surgery to repair that tear that happened during the invasive test. And the story got even worse. The bypass graft went down, it closed. She had to have a stent. She was on dual antiplatelet therapy. So, all the risk of bleeding and the stent closed. And she ended up with a heart transplant. And it all started from that C.T. scan that the doctor ordered just to reassure her.
I hope this story sticks with you all, because it really stuck with me.
That’s it for today’s show. I want to thank my guests, Rita Redberg and Bobby Yeh, and our listener Max, whose questions inspired the idea for this episode. And thanks to you, as always, for listening.
* * *
Coming up next week: When the Supreme Court overturned Roe versus Wade earlier this year, it changed the status of abortion in some states. It also caused a lot of confusion about what was legal for doctors to do in certain situations.
KING: We hear of these reports of delays in care for ectopics or miscarriages when there’s bleeding and trying to reach some particular moment in time where it’s life threatening truly in the moment so that you can intervene, which is not how we practice medicine.
Many factors can influence how, and how well, doctors practice medicine. Including, sometimes, the law.
MELLO: We generally define defensive medicine as engaging in clinical behaviors, primarily because you would like to reduce the risk of a lawsuit and not primarily out of clinical judgment.
We’re going to talk about defensive medicine: what it is, how much it costs us, and why it’s on the minds of many physicians and health care policy experts these days. That’s all coming up next week on Freakonomics, M.D. Thanks again for listening.
* * *
Freakonomics, M.D. is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and People I (Mostly) Admire. All our shows are produced by Stitcher and Renbud Radio. You can find us on Twitter and Instagram at @drbapupod. This episode was produced by Julie Kanfer and mixed by Eleanor Osborne, with help from Jasmin Klinger. We also had help this week from Lyric Bowditch. Our staff also includes Neal Carruth, Gabriel Roth, Greg Rippin, Rebecca Lee Douglas, Morgan Levey, Zack Lapinski, Ryan Kelley, Jeremy Johnston, Emma Tyrrell, Jacob Clemente, Alina Kulman, and Stephen Dubner. Original music composed by Luis Guerra. If you like this show, or any other show in the Freakonomics Radio Network, please recommend it to your family and friends. That’s the best way to support the podcasts you love. As always, thanks for listening.
JENA: As you were talking, the first idea that came into my head was, “Gee, I wonder if I looked at all the residents and the medical students who worked in the cardiology service when you were the attending and I followed these people out two years or three years or 10 years — I wonder if they order, fewer lab tests and more appropriate imaging than their peers who had a different cardiology attending?”
REDBERG: I would love to see those results.
- “Strategies to Reduce Low-Value Cardiovascular Care: A Scientific Statement From the American Heart Association,” by Vinay Kini, Khadijah Breathett, Peter W. Groeneveld, P. Michael Ho, Brahmajee K. Nallamothu, Pamela N. Peterson, Pam Rush, Tracy Y. Wang, Emily P. Zeitler, and William B. Borden (Circulation: Cardiovascular Quality and Outcomes, 2022).
- “Variation in Physician Spending and Association With Patient Outcomes,” by Yusuke Tsugawa, Ashish K. Jha, Joseph P. Newhouse, Alan M. Zaslavsky, and Anupam B. Jena (JAMA Internal Medicine, 2017).
- “Environmental Causes of Breast Cancer and Radiation From Medical Imaging: Findings From the Institute of Medicine Report,” by Rebecca Smith-Bindman (Archives of Internal Medicine, 2012).
- “Left Main Trunk Coronary Artery Dissection as a Consequence of Inaccurate Coronary Computed Tomographic Angiography,” by Matthew C. Becker, John M. Galla, and Steven E. Nissen (JAMA Internal Medicine, 2011).
- “Low Diagnostic Yield of Elective Coronary Angiography,” by Manesh R. Patel, Eric D. Peterson, David Dai, J. Matthew Brennan, Rita F. Redberg, Vernon Anderson, Ralph G. Brindis, and Pamela S. Douglas (The New England Journal of Medicine, 2010).
- “Projected Cancer Risks From Computed Tomographic Scans Performed in the United States in 2007,” by Amy Berrington de González, Mahadevappa Mahesh, Kwang-Pyo Kim, Mythreyi Bhargavan, Rebecca Lewis, Fred Mettler, and Charles Land (JAMA Internal Medicine, 2009).