We’re Not Getting Sicker — We’re Overdiagnosed

Can you have too much of a good thing? Medical care is a good thing. But my guest today, Suzanne O’Sullivan, says that our medical system is over testing, over diagnosing and over treating patients. That’s the case she makes in her newest book, The Age of Diagnosis: How Our Obsession with Medical Labels Is Making Us Sicker. Suzanne is a neurologist who focuses on the treatment of epilepsy, and she’s written four books about her experiences as a physician.

O’SULLIVAN: Many of my patients are in their twenties or thirties. I see people with long lists of medical labels at that age. And it shocks me because I didn’t see it 30 years ago. And I can’t believe that we’ve got sicker, you know, to that degree. Something is going wrong. And I appreciate that some of the subjects in this book will be difficult for people. But I also think that something needs to be said about these issues.

Welcome to People I (Mostly) Admire, with Steve Levitt.

*      *      *

Suzanne not only thinks that the medical system is working on overdrive. She believes that many patients are misdiagnosed. Her first book tackles the complexities of psychosomatic disorders as a neurologist, she commonly sees patients who experience paralysis, headaches, dizziness, and seizures. However, she estimates that for 20 percent of her patients, the cause of their sometimes debilitating symptoms is not neurological, it’s psychosomatic. We start our conversation with Matthew, one of the patients she describes in her first book, It’s All in Your Head: True Stories of Imaginary Illness.

O’SULLIVAN: So Matthew had noticed — I think he had a combination of back pain and tingling in his legs in the first instance. So his symptoms at the beginning were not terribly disabling, but over the course of a few months, he got to a point where he came to my clinic in a wheelchair, essentially unable to walk or unable to move or to feel his legs. As a neurologist, you can generally tell if somebody has weakness in a limb, whether it’s coming from their brain or whether it’s coming from a nerve or whether it’s coming from a muscle, because these types of weaknesses have a very precise pattern that fit with our neuroanatomy essentially. So examining Matthew, his clinical signs were very contradictory, meaning they didn’t fit with neuroanatomy. Certain reflexes were normal in a situation in which his muscles were entirely paralyzed. 

LEVITT: So if you banged him on the knee with the hammer, his leg would kick, but he absolutely was unable himself to force his leg to move.

O’SULLIVAN: Yeah, and that’s acceptable neurologically in some diseases, but in the situation in which he had this kind of very profound amount of paralysis, you would expect the reflexes, if it’s a nerve problem, to be gone, or the reflexes to be very brisk if it’s something in the brain. But neither was present. His unconscious reflexes were all entirely normal. What we’re trying to do as neurologists always is to try and hone down to a point in the nervous system; someone’s weak in their legs, is it a back problem? Is it a brain problem? Is it a muscle problem? And that simply wasn’t possible with Matthew because the signs were not consistent with any single neuro-anatomical place. And that happens in a very specific condition, which neurologists see very regularly, maybe about a fifth of their clinical encounters, and that’s a condition called functional neurological disorder. It would previously have been referred to as psychosomatic. So this was weakness which had a psychological origin rather than being due to a neurological disease.

LEVITT: So psychosomatic is a term we’ve all heard. Can I make sure that the definition that we use in popular culture matches with the definition you’re using as a doctor?

O’SULLIVAN: I think that’s a good idea. A psychosomatic disorder is a set of physical symptoms that are real, that are usually causing significant disability, but that cannot be explained by a disease and are likely to have a psychological origin.

LEVITT: So Matthew is a little bit like me, when he feels something wrong, he sounds like he was the kind of person who went on the internet. He came into your office completely and utterly convinced that he had multiple sclerosis — although he, in your view, didn’t. What do you do when you face a patient who is certain of their own diagnosis and you’re certain they have a different diagnosis? How does that conversation go?

O’SULLIVAN: Well, I mean, that’s a phenomenally difficult conversation. Most people are not aware that something as dramatic as complete paralysis can occur as a result of a purely psychological cause when the brain and nerves are otherwise healthy. So that diagnosis usually comes as an absolutely immense shock. And if you deliver the diagnosis clumsily, people’s initial reaction is, “You’re telling me I’m doing it on purpose. You think I’m imagining it.” And so the first thing I really need to do with patients in that situation is reassure them that this is a real condition. I do not believe it’s within their control, but I believe it is controllable. I usually start by really describing to people all of the really common psychosomatic symptoms that are not enough to be a disorder that we all experience. So I’ll often say to patients, “Well, you know, sometimes if you’re nervous, your heart beats incredibly quickly. You can’t control that, you’re not doing it on purpose, but you know it’s not due to a heart disease. If I told you to stop it, you couldn’t simply stop it.” And it’s the same with motor symptoms or sensory symptoms. Movement is supposed to be automatic. If you’re paying an excessive amount of attention to movement, it stops being automatic. We’ve all experienced that. If we’ve walked on a side of a cliff or somewhere that feels precarious to us, it makes us less efficient in our movement. And that’s the way that we all experience this kind of mind-body connection that makes our bodies less efficient. But for some people, it can become much more extreme and lead to disability. Similarly, sensation: our bodies are awash with sensations that are available for us to notice, but we don’t notice them because they’re not important. And we dismiss them because we couldn’t possibly deal with all of the information we’re being bombarded with all the time. But if I say to somebody, “Well, pay attention now to how the chair feels underneath your bottom,” or, “Pay attention to your hand,” suddenly your hand feels different. And through examples like that, I try to let patients know that these are common experiences, and like anything in the body, they go wrong. And when they go wrong, they can lead to illness.

LEVITT: It’s really interesting and surprising to me that the brain has the power to pull these things off. It must be working through some kind of a feedback loop in which little actions trigger things far beyond the control of what Matthew or someone else is trying to accomplish. 

O’SULLIVAN: I mean, we think about these things as part of what they call predictive coding. So our brains are like prediction machines. They’re not assessing the world as it’s presented to us exactly, they are assessing the world according to its experience and expectations. And those sort of perceptions, unfortunately, can lead us into illness. So if I give an example of, say, someone having a blood test, if you are really phobic about having a blood test and a needle is approaching your arm, people can feel pain before the needle even touches their arm because their expectation of pain is so strong that it creates pain even in absence of the painful stimulus. And in the case of Matthew, the first thing he noticed was the tingling in his legs, and that directed all of the attention towards his legs. And then, as we all do, he developed a very rich kind of inner picture of what he thought was happening in his body. And he had an expectation of how his symptoms would progress. “Well, is my walking okay?” So now he’s paying lots of attention to his walking and something that we do automatically and never think about is suddenly within his frame of attention, and that makes his walking unnatural. So it’s sort of an escalating problem that the more you notice, the more you pay attention, and then it escalates in that way.

LEVITT: Now, you’ve also told the story of a woman named Sharon, and she’d been having seizures for five years before she found her way to you, and twice because of her seizures lasting so long, doctors had put her into an induced coma out of fear that the prolonged seizure would lead to brain damage or death. So to any observer like me, Sharon obviously has epilepsy, but when you see people like Sharon, oftentimes you find the answer lies somewhere very different.

O’SULLIVAN: We all kind of dissociate, which is the basis for these seizures. We all dissociate from time to time. So I am constantly having to replay news broadcasts or podcasts for the fact that I missed. Sometimes I have to do it 10 times in a row because I zone out regularly when I’m listening to things because my brain is distracted or my brain can’t handle everything I have on my plate at the moment. But for some people, it goes so wrong that it produces dramatic symptoms that are disabling. And that’s when it becomes a disorder.

LEVITT: How do you distinguish between someone who has epilepsy, and someone who’s having seizures that are coming from a totally different source, a psychological source?

O’SULLIVAN: Yeah, so I’m a seizure specialist, so I would say that it’s phenomenally easy, but it’s not phenomenally easy for people who don’t see seizures all the time. In an epileptic seizure, there’s kind of a wave of electrical discharge that shouldn’t be there passing through the brain. And as it passes through the brain, it does that in an anatomical way. So it starts in one bit of area of the brain, then it spreads to the motor region on the same side of the brain, and then forward or backward, and that makes intelligible kind of sense in how the symptoms evolve. But in a non-epileptic seizure or dissociative seizure as we now call them, what happens is that the seizures don’t make anatomical sense, so the shaking stops and starts, for example. If you want to tell the difference between these different seizures, if you’re not a seizure specialist, you do an E.E.G. or a brainwave test and in an epileptic seizure in which someone is convulsing, you will see lots of these kind of spike discharges representing unwanted electrical activity in the brain in someone with epilepsy. But if you look at someone with dissociative or non-epileptic seizures, psychosomatic seizures, the brainwave pattern looks like a normal waking pattern.

LEVITT: And in modern Western society, there is a strong tendency to be dismissive of psychosomatic disorders. The word hysteria, which preceded it, is loaded with negative connotations. We just want to be really clear, in no way, shape, or form are these patients faking anything. These are real seizures that are happening. They’re just not being triggered by the mechanisms that we associate, say, with epilepsy. But in general, society is not very sympathetic to these kinds of illnesses, but you are sympathetic to them. Can you talk about your own transformation from, I think, being less sympathetic to being much more sympathetic?

O’SULLIVAN: Yeah, I’ve gone through the same kind of trajectory as most people, which is struggling to believe these are possible, to struggling to believe that my patients genuinely have no insight into what is happening to them. And I’m 100 percent sure they have no insight. As a medical student, I didn’t always find that idea easy to hang on to because these conditions can often show signs that make you suspicious of your patients. For example, in people like Matthew, what can happen is they’ve got this profound weakness in their legs when they’re lying on the examination couch and you say, move your toes, move your foot, they can’t move it at all. And then you go back to your desk and you’re writing your notes, you’ll see little flickers of movement as they’re putting on their shoes. Or people like Sharon when they’re having seizures sometimes they wake up in the middle of the seizure and then go back into it again. And you think, are they doing it on purpose? Are they trying to trick me? So I had those experiences as a medical student. I understand the disorders better now. I realize that the reason someone like Matthew can momentarily move his feet when he’s putting on his shoes is because these disorders are maintained by the attention you’re paying to your symptoms. And in those little distracted moments, when he’s thinking less about his symptoms, his symptoms are not so severe. The reason I became interested was because I wasn’t taught about these as a medical student. I’m not saying I wasn’t aware of them, but it wasn’t really considered to be my job as a neurologist to deal with these issues because they were considered to fall more within the remit of a psychologist or a psychiatrist. So when I became a consultant and patients were my own patients, I found that I was seeing a very large number of these people because I was working for an epilepsy service and I was expected to see somebody who was having 100 seizures a day and say, “Well, it’s not epilepsy. So you no longer can see me. Go home and find somebody else to see.” These people have physical symptoms, but doctors who look after people with physical symptoms are not taking responsibility for them. And then I just found that it was just too difficult to be constantly discharging people who are some of the sickest people I was seeing.

LEVITT: So even though you’re in an epilepsy clinic and these folks don’t have epilepsy, you keep seeing them and keep treating them because they’re sick.

O’SULLIVAN: Yeah, I mean, there’s a limit to what I can do in terms of treatment. So I refer them to my psychiatry and psychology colleagues. It depends on the problem. If it’s a problem like Matthew’s with walking, well, if you couldn’t walk, how would you feel about just being referred to a psychiatrist? You’d think, “Well, I can’t walk, I need to be rehabilitated.” So if it’s a physical problem, I’ll make sure they get physical rehabilitation with psychological support, but I will also very often keep them under my care, because what happens is that they feel dismissed, like their problem isn’t a real problem. And when they’re having lots and lots of seizures there’s not much value in me saying, “I believe you have a real problem, but I’m not going to see you anymore for it.” So I think that kind of ongoing care to support them in the belief of the diagnosis is very important. What can also happen is people with seizures, they end up in emergency departments, and they see a new doctor who’s never seen a seizure in their life before. And that doctor says, “No, this is epilepsy.” So, a very big part of my role is just helping them to keep trust in the diagnosis and to do that, I need to keep them under my care.

LEVITT: Because recovery is not quick. One might naively expect that, well, if it’s psychosomatic and once I come to understand it’s psychosomatic, maybe the symptoms will go away. But it’s not nearly that simple is it?

O’SULLIVAN: No, I mean, that’s the difficulty for these patients very often. You can imagine you’ve got a colleague or a next door neighbor who you thought had epilepsy and now they’re telling you they have these sort of dissociative seizures that were once called hysteria. In many people’s minds that immediately downgrades the level of suffering or disability that they have. But the reality is that patients with epilepsy, 70 percent of them will go into remission and they will live completely normal lives. And epileptic seizures are often very brief, lasting less than a minute or a couple of minutes. Whereas these dissociative seizures only 30 percent go into remission, and often those seizures are much more frequent, dozens in a week or even in a day. Often those seizures are much longer and they’re much more likely to lead to intensive care admissions. It’s ironic that people respect this type of suffering so much less, when in many ways it’s extremely disabling and it’s very hard to recover from. But if you can see a patient who has these sort of problems very quickly, you know, if I met Matthew in the first couple of weeks of his journey before the paralysis had set in and was able to allay his fears, it would not have progressed. Or if I met Sharon three months after her seizures started, was able to direct her attention away from the seizures, you can stop them in their tracks that way. But unfortunately, many people spend years looking for that physical disease, that cancer, that multiple sclerosis before they get to the point of having someone tell them it’s psychosomatic. It’s the delayed diagnosis that I suspect is the reason why so many people don’t get better.

LEVITT: Now, I’m sure you’re familiar with the book The Body Keeps the Score, by Bessel van der Kolk. I’m probably misinterpreting his argument. For certain, I’m radically simplifying it. But I remember him arguing that the body, not just the brain, retains powerful memories of childhood trauma. And it seems, superficially at least, to align with the existence of these psychosomatic illnesses. Do you see things in the same way as van der Kolk does?

O’SULLIVAN: I don’t disagree with that statement, but I also think that these disorders have a much wider range than that. So I think that sometimes these disorders are conceptualized as all due to stress, and all due to childhood trauma, and therefore, for many years, people have gone to patients with these sort of conditions and said, “Something must have happened to you.” And they’re desperately trying to talk the patient into admitting that there’s some major psychological trauma, either in their background or in their current lives. And the truth is that for a large percentage of patients, that’s not the case. It’s actually just a maladaptive response to some physical illness that’s happened to them. So I see these kind of problems often in people who, for example, had an injury that changes their relationship with their body and changes the attention they had with their body, or someone who faints. A faint can often lead to these dissociative non-epileptic seizures. I think it’s just that there are a great deal more causes. And when we focus too much on trauma, psychological trauma and childhood trauma, patients often feel that they have to dig into this position of having a real illness that has nothing to do with psychology.

LEVITT: Naively one might think that a good way to treat psychosomatic disorders would be with placebos. If the brain can, quote, “trick the body into illness,” perhaps the brain can be tricked out of it. 

O’SULLIVAN: I think placebos could have a really useful place in treating these disorders for exactly the kind of reasons you describe. Some of patients who have non-epileptic dissociative seizures are given epilepsy drugs, which obviously shouldn’t have an effect on them, but actually do get rid of the seizures. So placebo would certainly help patients. I think that these treatments have just fallen out of favor and people are a bit nervous about using them because they are perceived as a bit of a trick. I don’t think that they are a trick, because actually, as I’m sure you know, placebos work even if you know it’s a placebo. I think it depends on the patient, of course.

LEVITT: My dad is a doctor, and I still remember vividly a story he told me when I was a child. He was in the U.S. Air Force, stationed in the U.K. in the early 1960s, and he had to deal with a really wide range of illnesses with the soldiers who would come and see him. And my father is very physical as a doctor, he tries to really put his hands on people and make them feel like he’s giving them attention. And when he was convinced that one of his soldiers’ symptoms were psychosomatic, he’d listened very carefully to the story, and then he would send them home with sugar pills, placebos, and tell them to come back in two weeks if they weren’t feeling better. Now he wouldn’t tell them they were placebos, he would just say, here are your pills. Now if the patient came back in two weeks, He would give them another thorough examination and then he would send them home with another set of sugar pills, but these pills look different. They were huge pills, like horse pills, with the idea being that a pill that is this hard to swallow, it must be powerful if the doctor is giving me such a crazy pill. And I think he would say just for emphasis, “Now you need to take two of these pills.” And then if the patient came back two weeks later, then my dad went to his secret weapon. Now, very few patients did come back, but this third set of placebos were incredibly tiny pills. And he would say, “These are so powerful. They’re very tiny. You only need one.” And he said he never had a patient come back again after those pills. Now, I find the psychology of what he was doing fascinating, but also the notion that he was fine to practice medicine that way in the 1960s. I think you’d get into a lot of trouble if you were doing that today.

O’SULLIVAN: Yeah, certainly I wouldn’t suggest doing that today. But I mean, in treating these disorders, the relationship between the patient and the doctor and the trust that the patient has in the doctor, it may not even have been the pills. Your father put his hands on these patients, which is something that doesn’t necessarily always happen these days. And, treating people with these sort of disorders is a very important clinical skill and time spent with a patient — I often find one conversation with a patient can actually make seizures better almost immediately. I am fascinated what he got away with, but not surprised to hear it was helpful. 

*      *      *

LEVITT: As I read your first book, which talked about these psychosomatic illnesses, it was really mind boggling. But nothing prepared me for what was going to be in your book called Sleeping Beauties. Could you talk about a young girl named Sophie in Sweden and what her experience was?

O’SULLIVAN: Yeah. Sophie is a young girl. I encountered her first, actually, in a newspaper article. So I read that there was a group of children in Sweden who were in a deeply comatose state and simply couldn’t be woken up. And she was one of those. I deal with people who are unconscious from multiple different levels, so I recognized something in her story from my own clinical practice. And I’ve come to recognize that when you see “mystery illness” in the newspaper, it very often is referring to a psychosomatic disorder. Because when I drilled down a little more into this young woman’s story, I discovered that she was one of a group of asylum seekers who traveled to Sweden, lived there for a few years, and then when faced with potential deportation because they were refused permanent status in Sweden, lapsed slowly into this, what people were calling a catatonic state. Now, this state was absolutely profound. So this is a group of children all under the age of 18, and many of them 12, 13 years old, who basically become listless and withdrawn, then they withdraw absolutely completely to the point that they lie in bed. Their eyes are closed. Their muscles are completely relaxed. They don’t open their eyes. They don’t talk. They don’t get up to use the toilet. They’re maintained by their families wearing nappies and fed through tubes.

LEVITT: This is for years in some cases.

O’SULLIVAN: Yeah. So, I went to Sweden because I wanted to understand this disorder better. And I met some young people who had been in this literally for years. There were young people who were going from the kind of early teen years, through puberty, into the later teen years in this condition. And they call it resignation syndrome. It wasn’t very common in people from African countries. It was much more from ex-Russian republics. So it was a kind of niche group of asylum seekers specifically in Sweden. Some of these children in the early days were in hospitals. Some of them stayed in hospital for a prolonged period, some of them stayed in intensive care unit. And then because it’s judged something that medicine couldn’t treat, they were sent home and therefore they were being cared for by their families who did makeshift physiotherapy for them and fed them with syringes and tried to keep them healthy that way at home.

LEVITT: So how do you make sense of this?

O’SULLIVAN: Well, this is probably what we call a sort of culture bound syndrome. It’s a set of symptoms that develop out of beliefs that occur within a specific community. It’s just the same as the psychosomatic conditions that I was discussing before with, say, Matthew: your body behaves in the way that you expect it to behave because your brain is a sort of prediction machine. So, you perhaps have one person who starts this phenomenon off completely unconsciously. So you would have one child when faced with an enormous amount of stress, becomes withdrawn, takes to their bed and becomes unable to communicate. And that’s a psychological problem that is entirely of their own. But then it becomes known within a community that people who are in this particular situation, facing being sent back to their country of origin, then they will get an expectation of how a child might get sick in that circumstance. And the child will not be immune to that expectation. You know, a child who feels irritable in this way sometimes gets withdrawn, and that sometimes means that they take to their bed, and people who are in bed can’t move, so now I can’t move. And I emphasize this is all happening unconsciously, but it’s a sort of pattern of illness that is culturally accepted as a way of expressing distress within a particular community.

LEVITT: And you document many examples across different communities in South America, in the United States, it reminded me of old economic research that was done by David Cutler and Ed Glaeser and Karen Norberg. And they were looking at patterns of teen suicide in the United States. And one of the things that was incredibly striking, I’ll never forget, this was written probably 25 years ago, is that in the U.S. only 1 in 400 teen suicide attempts is successful in the sense that the person who attempts suicide actually dies. And one possibility is that teens are totally incompetent and they can’t do things they want to accomplish. But the economic interpretation was that these teens have little direct familial or economic power, and a suicide attempt is a way to improve their bargaining power. And now that’s very economic language, but in the language you use, it maybe has some parallels to this resignation syndrome and that these teens are trying to express their pain in a way that people will listen. And in the U.S. that’s a suicide attempt, and for Swedish refugees, that’s complete disappearance into solitude.

O’SULLIVAN: I think we express our distress in a way that is intelligible to the community in which we live and in a way that is most likely to elicit the help that we are asking for. And very often, in the case of young people, and I would say young women in particular, just expressing your distress in a verbal way doesn’t necessarily get a lot of attention or doesn’t get results. There was another example of a culture-bound syndrome that I wrote about, which is called grisi siknis, which affects the Miskitu people of Nicaragua. This grisi siknis literally translates to crazy sickness. It affects young women in particular. Young women in their teen years, they’re faced with unwanted male attention, and they’re in a society that asks them to be modest. But they’re also being sexualized by men and it’s a very difficult situation for them and they don’t know how to express the need for help or to express how it distresses them. And in grisi siknis, it’s like a cultural way of expressing that distress to get the help you want. And grisi siknis manifests as seizures along with hallucinations and other symptoms. And it’s understood by the community when grisi siknis occurs, you need to band around this young woman as a community and help her. And they use a lot of traditional treatments, but I would suggest the support of the community is probably the thing that really helps them. That speaks to how you ask for help in a way that works and you know the way that works from the society in which you live. And I think that for these young people with resignation syndrome, this was a very powerful expression of distress. And without it, they could have complained and no one would have listened. 

LEVITT: It’s interesting that in Nicaragua, the response to the psychosomatic hallucinations and whatnot is for the community to band together and support the young girls, whereas in Western culture, our response to psychosomatic illness is typically the opposite, right?

O’SULLIVAN: I was so impressed by that because basically if you get grisi siknis and you get these seizures, you have an expectation of full recovery, actually. People do band together and support them and there’s nothing stigmatized about grisi siknis, people just have a different understanding of illness than we do. It’s not divided into psychological, which is slightly less important than physical. They don’t consider grisi siknis to be a psychological problem. They understand what they have to do to help it and therefore it gets better. Whereas we lock people away and make them feel ashamed of what’s happening to them and they don’t get better.

LEVITT: Your books have been extremely well received, both liked by readers and by book critics, and you’ve got a new book called The Age of Diagnosis, which I think people will like as well, but I think your book is also going to make a lot of people angry. Are you prepared for the firestorm of controversy that might come with your new book?

O’SULLIVAN: I cannot tell you how much I don’t like controversy. I could be a lot more provocative with some of the subjects that I write about, but I go out of my way not to be provocative. I appreciate that the Age of Diagnosis is going to be a difficult read for some people. But I also think that there are things that occur within medicine that are controversial within medicine, between professionals, between scientists. We’re aware of all of these controversies, but we don’t always share them openly enough with the general public. And I think there are also conversations about the issues in this book that deal with overdiagnosis that lots of us are having in our living rooms, but nobody is having loudly enough in public forums. And I think that something needs to be said about these issues.

LEVITT: Yeah, so the main thesis of your book is that we are living in a world of overdiagnosis. So what do you mean by overdiagnosis?

O’SULLIVAN: I think a lot of people think that overdiagnosis is occurring within the field of mental health, and that’s the first thing people think about when you talk about overdiagnosis. So that’s what I would refer to as kind of overmedicalization, where you are referring to difficulties or struggles that we would once not have considered to be as medical and directing them towards medical attention. So I’m talking about you know, the number of people who’ve been diagnosed with autism, A.D.H.D., depression, and so forth, that, you know, 20, 30 years ago, perhaps would not have attracted those diagnoses. But I’m not only talking about those mental health conditions. I’m talking about the overdiagnosis and over medicalization of physical parameters too. We have gradually changed the concept of how depressed you have to be to be called “medically depressed” but we’ve also gradually changed the concept of what it means to have high blood pressure or diabetes, and numerous other diseases like that. Scientists and doctors have been gradually changing the parameters of what counts as disease, and they do this for a really good reason. They do it because they want to miss as few patients that could benefit from help as possible. So it’s all done in a well intentioned way. But unfortunately, that trend of gradually loosening diagnostic criteria for both physical health conditions and mental health conditions has led to a sort of diagnosis explosion in numerous sectors of medicine. 

LEVITT: I was surprised that in the statistics you report that asthma, of all things, has increased by almost 50 percent when you might have thought it should be the opposite because many of the air quality problems that would have triggered it have been improved by public policy.

O’SULLIVAN: I think that when people think about diagnosis, they think about it as a more definite thing than it actually is. I’m not an expert in asthma, so I don’t necessarily want to speak to that in great detail. But, people think, well, if someone says that you have high blood pressure, or someone says you have asthma, or someone says you have diabetes or autism, that’s a definite fact based on clear scientific evidence. But first of all, diagnostic criteria are always changing for how much of something you need to have the disease. And then we also have the difficulty of medical tests and how we diagnose things as being highly subjective things. To a certain degree, you can bend tests and you can bend medical criteria to diagnose as many people as you want to. And we are encouraged to do so because underdiagnosing terrifies doctors. They don’t want to miss things. Terrifies patients because they don’t want to be underdiagnosed. But there’s always a grey area in every diagnosis. There’s always a grey area where you’re not completely sure. And it’s very easy to draw a lot of people in the grey area into the diagnostic category.

LEVITT: The most interesting chapter to me was focused on Lyme disease. Could you talk about how Lyme disease is diagnosed?

O’SULLIVAN: This is a really good example because you would think that Lyme disease is a fairly straightforward diagnosis. So obviously people will be aware that Lyme disease is caused by a bacteria and that you catch Lyme disease by a tick bite. And one might assume, therefore, that if you have the bacteria, you must have Lyme disease. And if you don’t have the bacteria, you don’t have Lyme disease. But Lyme disease is a beautiful example of how subjective diagnosis is both in the clinical symptoms and in the tests. The bacteria that cause Lyme disease are very hard to detect in the blood because there’s not loads of them, so you might take a blood test and you’re not necessarily going to see a bacteria. So instead what you look for is the immune reaction to the bacteria. So you immediately got a problem because now you’re looking at something that is indirect evidence for the disease. And there’s lots of things that will affect that indirect evidence. It could be that you have an immune reaction, but it’s a really old reaction. You’ve been living in a Lyme disease area your whole life. You’ve been exposed to this bacteria lots of times. Or it could be that you’re very sick for some other reason. You’re producing lots of antibodies and those antibodies cause false positives. So the blood tests that you do for Lyme disease are confounded by multiple different variables. Did you test for the right strain of the bacteria? All of these things can be changed according to how labs configure their tests and how doctors interpret the results of the tests. So the C.D.C. sets out standards for a Lyme disease diagnosis. And I think it was in about 2023, it recognized about 63,000 people had Lyme disease according to the C.D.C. standards. But nearly half a million people were being treated for Lyme disease according to their health records. So it’s likely that all of those extra people had a diagnosis that would not be necessarily considered correct according to C.D.C. standards. And that’s because medicine is an art, clinically medicine is an art. But also when you look at lab tests, they are subjective and they are open to interpretation.

LEVITT: And in this case, it’s really probably about incentives, right? The doctors have a patient, they’re sick, they worry about what will happen if they don’t treat them for Lyme disease. So they have no idea if they have Lyme disease, but they treat them anyway and hope for the best. In Lyme disease, that’s probably why there are so many more people who are told they have Lyme disease than the C.D.C. thinks are true.

O’SULLIVAN: I think it’s probably a bit more than that. A difficulty with Lyme disease is that some of the symptoms it produces can be quite vague and they could be symptoms of almost anything; tiredness or a skin rash, for example. So you can imagine how common those symptoms are if you’re a family doctor, you’re seeing them all the time. And then also there are lots of areas in the U.S. where Lyme disease is quite prevalent. It’s perfectly reasonable, therefore, if someone sees a patient in the first instance, they’ve got vague symptoms, they’re in a Lyme disease area to treat them for Lyme disease, even though you’re not sure. That’s the way doctors practice all the time. However, usually what we’ll do as doctors in that circumstance is we’ll try the treatment, see how it works. If it’s not working, then we’ll stop the treatment. That’s not what happens with Lyme disease. Unfortunately, a condition called chronic Lyme disease has taken hold of the Lyme disease community and is being, I would say, exploited. There is a very standard treatment for Lyme disease as laid out by the C.D.C. But there are a lot of people who are attending private doctors who get antibiotics for many years, even though the signs are not felt to fit very well with Lyme disease.

LEVITT: And one of the reasons why overdiagnosis would be problematic would be if there’s a psychosomatic component that kicks in when you’re led to believe that you have Lyme disease. Do you think there is a big psychosomatic component of chronic Lyme disease?

O’SULLIVAN: I would be quite sure that a percentage of people — I spoke to this very nice lady from Wales in the U.K. who suffered a kind of flu-like illness while she was on a holiday. She did not live in a Lyme disease area and she was on holiday in a city. There was no reason for her to contract Lyme disease. And she developed flu-like symptoms that didn’t go away. And then within a couple of weeks she had 90 to 100 different symptoms, and this progressed to the point that she was too tired to leave her house. Now, if you’ve got 90 to 100 symptoms, you’re bound to have a couple of symptoms in there that fit with Lyme disease. And after she saw a television program about Lyme disease, she developed the conviction that this was her diagnosis. I would say that people in her situation, I would certainly say there’s a large psychosomatic element. The belief in the diagnosis of chronic Lyme disease is very problematic because people get dependent on long-term antibiotics, and the belief that they won’t get better. And if you believe you won’t get better, then it’s very difficult to get better.

LEVITT: So, I started this conversation by suggesting that I thought that this book would make people mad. Now I’m going to ask you about long Covid and see how many people with long Covid you can offend because you are a real skeptic with respect to long Covid. Is that true?

O’SULLIVAN: First thing I want to say is that there should be nothing offensive in my view at all here, because my position always with these disorders is that people are suffering. Symptoms are real, but it’s how people conceptualize what’s causing those symptoms that I think is a problem. And it’s also the perception. So if I’m about to say that I believe long Covid for a lot of people is psychosomatic, that should not be seen as in any way lessening their suffering or saying that they are not suffering or saying that they do not need help. I’m merely saying that the mechanism of illness according to the evidence and according to my view, fits better with psychosomatic illness. But with regard to long Covid, I mean, it came about in a very unconventional way. The term arose on the 20th of May, 2020, when it was tweeted by an Italian person who was suffering with long term symptoms after contracting Covid. It very quickly took off in social media. There were people who were not hospitalized, who had a more mild form of Covid, who didn’t seem to be getting better quickly enough, and they banded together on social media under the hashtag long Covid, and found support in each other that way. The difficulty was that long Covid then spread very quickly to the mainstream media into the mouths of government officials who were telling us every day not only could we die and we might end up in hospital, but even if we got a minor illness, we could get this thing called long Covid. But there was no definition for long Covid. There was no specific symptoms that told you what it was. There was no test that proved what it was. So it was a name that could explain any type of suffering during the pandemic in those early stages. It quickly spread to medical journals. And that science just moved too quickly to be really accurate or meaningful. And for some people, it was being referred to as long Covid, and for others, it was probably given different names. There’s a few pieces of evidence that really support the psychosomatic theory. More than one study has shown that loneliness was a predictor for long Covid. There was a study in Germany where they followed healthcare workers, and people who had an expectation of symptom severity were more likely to get long Covid. So people who expected to get it got it. There was another study in which they showed that self diagnosis was more likely to lead to long Covid than an official diagnosis of Covid infection. Obviously people were suffering in the pandemic. They were suffering psychologically. Their diets changed. They stopped exercising. Everything about their body changed. And here was a diagnostic explanation for it. So I really feel that long Covid drew in lots of people who were suffering in lots of different ways under this umbrella. But I really don’t think people should feel that means the suffering isn’t real. Again, it’s an intelligible way of asking for help at a time that you need help. And perhaps what we should learn from it is that we need better support systems for people that don’t rely upon a medical diagnosis.

LEVITT: Yeah, it’s really true, right? In modern society, if you want to get help, you need a medical diagnosis. You can’t get access to a lot of the support you might want if a doctor hasn’t signed off and says you need it. It seems like that’s just a fundamental problem with the way we’ve organized society. Would you agree?

O’SULLIVAN: A hundred percent. Particularly for some of the mental health disorders or learning disorders, problems like autism, A.D.H.D., depression, et cetera. First of all, in schools, your child is struggling in school. You want to pay attention to that struggling and try and get them extra support. The only way to get that very often is through a diagnosis, or the only way for the school to justify extra staff is through a diagnosis. And similarly, outside of schools for older people, illness gives you a pathway to ask for help and it also makes you part of a community of other people who are suffering. It makes a lot of sense for distress to be conceptualized under a medical label. But my concern really is what happens after that? When you label people, you change how they see themselves, and you change how they are seen by others, and that can affect their future, either as a child or as an adult. And that’s the thing I want people to realize when they’re thinking, “Well, you know, my child, I don’t know, they’re on the borderline. Perhaps I notice social difficulties, perhaps a diagnosis of autism will get them extra help in school that will propel them forward.” But what else comes with that? Will that child feel that they have a brain abnormality they cannot overcome and change their future in that way? Will other people think that there’s something wrong with them? So I think we need to talk more about diagnosis is not inert. We don’t talk enough about the negative implications of a diagnosis. 

*      *      *

One of the most popular podcast episodes I ever did was with physician Peter Attia, who espouses something he calls medicine 3.0. It’s a more proactive approach to medicine that focuses on preventing diseases like diabetes and heart disease and cancer. But some of the points in Suzanne O’Sullivan’s latest book, The Age of Diagnosis, seem to challenge aspects of Peter’s thinking. I asked her opinion on medicine 3.0.

O’SULLIVAN: The minute you said medicine 3.0, I thought I probably was not going to agree with this doctor. I think prevention, absolutely. This is where you put your attention. If I use cancer as an example, there’s a lot of things we can do to prevent cancer; lose weight, stop smoking, eat a healthier diet, exercise, et cetera. These are the things we should pay attention to. But we have this assumption in medicine and in society, and this may be the sort of medicine 3.0 angle, which is that let’s just search for those cancers and get them at the earliest possible stage. And if we do that, we will be able to treat people more effectively and we will save lives. And that makes so much sense in one’s head. And that’s the basis for screening that we’ve all been doing for many years. But actually, time and time again, with screening programs for health, be it screening for cancer, be it screening for high blood pressure, we find that the maths isn’t adding up. So actually, most prostate cancer screening programs are no longer mandatory in most countries because what they discovered is that when you screen people for cancer, you find lots of early cancer cells and you treat those people, but just as many people are dying of prostate cancer. Our tests are so good now that if you look hard enough for a disease, you will find all these little early cancer cells and we don’t know yet how to distinguish the person who will get the advanced stage cancer, who will die of it, and the person in whom the cancer will never advance. So this sort of thirst for doing more tests and finding it earlier is proving to probably treat a lot of people who don’t need to be treated.

LEVITT: I was talking recently to a doctor who treats patients with prostate cancer. He said in his view the best treatment strategy for most prostate cancers is just to watch and wait. And that’s the advice that he gives to most of his patients. But he says they almost never follow his advice in this regard. Because once they know they have cancer, even without any symptoms, they want it gone. And so they opt for immediate radiation or surgery, and that’s an interesting case where I’ve assumed as we’ve been talking about overdiagnosis and over-medicalization that it’s maybe the fault of the medical profession, but here’s an example I think where it’s actually society and the way people perceive the word cancer that leads the patients to be driving treatment rather than the doctors.

O’SULLIVAN: There is an argument to say that some of these early cancer cells that we’re finding on very sophisticated tests should probably just be given a different name. For precisely the reason you were saying. There are watchful waiting programs. This has been a huge problem for breast cancer also. Most women really would be too terrified to enter a watchful waiting program. But the truth is that that treatment that they get can foreshorten their lives in other ways. So again, it’s not a matter of you get the treatment, “Oh thank God it’s gone.That’s the end of that.” You’ve had treatment that can affect your health negatively. And the outcome of that is if you look at a screening program and you look at how many people on that screening program died from cancer, it looks really successful because if you diagnose lots of cancers that never needed to be diagnosed and then you treat them and everyone survives, you’ll get wonderful figures in which everyone seems to be doing really well. If you look at all-cause mortality, so if you look at people dying for any reason, you’ll find that the cancer screening program has not resulted in any improvement in deaths. And that’s because you are saving one life from dying of cancer, but you’re probably negatively affecting another life with the treatment for cancer. And I think people need to understand these controversies are very open within the medical world.

LEVITT: What’s your vision? How should society and medicine interact?

O’SULLIVAN: So I really think we need to place more value and more investment in the clinical side of medicine. You get the most accurate diagnosis if you have a few different quality consultations with your doctor who lays hands on you and therefore can really understand the problem that you’re bringing to them, and then interpret the tests in light of that understanding. So I think at the moment we’re really heading more in the direction of saying, maybe this A.I. tool will diagnose millions of cancers much faster than a radiologist. But actually it’ll be nothing without that consultation that makes sense of what that scan is finding for the clinical context of the patient. It’s all about the patient and doctor or nurse interaction. And I think we need to start prioritizing that over technology.

LEVITT: So in the U.S., at least, I think every trend goes in the opposite direction, right? Doctors spend so little time with patients, they have such high caseloads that they don’t have time to sit down and to hear the patient because there’s seven other patients waiting in line.

O’SULLIVAN: Yeah, it’s a huge problem in the U.K. Very often general practitioners only have five minutes with their patients. How can you really get to the kind of nub of the problem? And that’s why doing tests and putting a medical label on something is often the easiest consultation for a doctor. Someone comes to me with a headache and they’re worried they have a brain tumor. Doing a scan to assure them they don’t have a brain tumor is the easiest thing I can do. It takes a few minutes. Explaining why doing the scan, A. it isn’t necessary, B. it could be harmful because it’ll pick up all sorts of incidental findings that will just add to the anxiety — that’s a 20-minute consultation. You know, if we go back to the example of chronic Lyme disease and I was talking about a lady called Shan, who had chronic Lyme disease, she saw lots of private doctors and I would say that there was some real charlatans amongst that group who charge her large amounts of money for very unconventional treatment that most doctors wouldn’t agree with. She loved those doctors. And I would say what was at the heart of that was that what she was really paying for was to be heard. These doctors spent time with her. They examined her. They listened to her and they wanted to meet her again. That’s what was sustaining her. And I think we in more conventional side of medicine do need to learn from that. We need to understand that you make more people better with slow medicine, and giving the doctor time.

LEVITT: I’m wondering about the many people who have friends or loved ones who are suffering from psychosomatic illness, and what advice do you have for those bystanders? How can we interact in ways that will be most helpful to those patients?

O’SULLIVAN: I think it’s extremely difficult for an individual to express this kind of concern about psychosomatic symptoms to their own loved one because it’s such a stigmatized disorder still, which is terrible given how common it is. I usually caution people to be reasonably careful about diagnosing friends and loved ones because it often isn’t taken terribly well. There are doctors who specialize in this now, so there are people who will be able to give a sound diagnosis. So it’s very important that people recognize that psychosomatic symptoms are real. And that they are completely unconsciously generated and that we start having the same attitude to them that we would have to any illness. If your loved one was in a wheelchair because they’d had a terrible accident and they’d injured their spine, you’d have an immense amount of understanding for that situation. A psychosomatic disorder is equally disabling. So we need to start giving parity to these illnesses so that we respect the suffering of the person and their experience and that we understand that what’s happening to them is outside of their control. But there’s also a balance there with also understanding that if you support them correctly, they can get better. and that is the difference, basically. we want to validate what they’re going through is real, but also support them to get better. People have to believe in recovery to get better.

I have to say, I found myself a little bit shocked by what Suzanne O’Sullivan had to say today. Of course I’ve heard of psychosomatic illness, but I always imagined it was both extremely rare and that there were real limits on the types of symptoms the brain could induce in the body. Hearing Suzanne’s stories, I feel like I need to dramatically update my worldview. The human brain is even more powerful than I imagined. I do think, though, that there is a really important, practical lesson to be gleaned from Suzanne’s work. Psychosomatic illnesses start small, and it’s only through a destructive feedback loop that they can eventually become debilitating. If one is aware of the sneaky way in which psychosomatic illnesses operate, one should be able to recognize the initial signs, the tricks the brain is playing, and short circuit the whole process. I’m going to keep a watchful eye out for this sort of feedback loop, and I’d suggest you do the same. If you’d like to dive deeper into Suzanne O’Sullivan’s ideas, I heartily recommend all four of her books, especially her most recent book entitled The Age of Diagnosis: How Our Obsession with Medical Labels Is Making Us Sicker. It is a scathing indictment of many aspects of modern medicine.

LEVITT: Now it’s a point in the show where I welcome on my producer, Morgan, to tackle a listener question.

LEVEY: Hi, Steve. So I actually don’t have a listener question for you today. Instead I want to talk about a documentary I just saw. It was released in 2024, and it’s called Counted Out. And you make an appearance! The film’s about the state of math education in the United States and it focuses on some innovative math programs that are happening in schools right now. Do you remember being interviewed for this film?

LEVITT: I do. It was a long time ago. It was right when I was first starting my center. And I remember that a documentary filmmaker reached out to me and she said, “Do you want to be in a documentary about math?” My knee jerk reaction is I never want to be in any documentary about anything, and I probably turned down a hundred of them. But I just started my new center at the U of C and I think we were just talking with Stephen Dubner about how we would do a Freakonomics episode on math and data, and so against all of my better instincts, I said yes, and I agreed to do that interview.

LEVEY: Right. So you guest hosted an episode of Freakonomics Radio in 2019 called “America’s Math Curriculum Doesn’t Add Up.” But before that, earlier that year, you were interviewed by Vicky Abeles, this film’s director and producer. Were you familiar with any of Vicky’s previous documentaries?

LEVITT: I hadn’t seen her movies, but when she reached out to me, before I agreed to say yes, I just looked and tried to see what she had done. She seemed like a really high quality, thoughtful documentary filmmaker and I loved the spirit of what she had done in the past. She had a film called Race to Nowhere, which is about this cycle of anxiety that kids have because of the tremendous pressure to achieve within the current high school system.

LEVEY: So you have not seen Counted Out, I take it?

LEVITT: I have not seen the movie, and I’m actually a little bit afraid because at the time that I talked to her I hadn’t thought at all about math and how we should change it. I bet I was not a very good interview, I’ll say that for sure.

LEVEY: They use one line from you.

LEVITT: Awesome. What was my one line?

LEVEY: This is what you say:

COUNTED OUT FILM: Parts of life that used to be driven by words and rhetoric — journalism, politics, law. Almost every occupation is now embedded in data. And it is transforming our ability to understand the world around us.

LEVITT: Ah, good.

LEVEY: Feel good about that line?

LEVITT: Yes. Hey, how do I look? Do I look younger?

LEVEY: Steve, you have to remember that I’ve never met you in person, A. 

LEVITT: That’s true.

LEVEY: B, the headshot that Freakonomics Radio Network has of you is from so long ago. C, you never let us record these listener questions with our cameras on. So to be honest, I have no idea what you look like today. But you look good.

LEVITT: I think they put a lot of makeup on me to cover up my imperfections. I probably got a haircut just for it too.

LEVEY: But I really enjoyed the film. You know it covered a lot of the ways that data science is the underpinning of so many areas of society, it covered things like gerrymandering, which has political implications; it covered things like recidivism; it covered things like Covid and the public’s understanding of how disease can spread so quickly through a community. And then a big part of the film was about math education. You know the beginning does paint a pretty bleak picture. There is a large number of adult Americans who just do not have a basic grasp on computation. And that’s a problem, because there’s a lot of professions today that require an understanding of data science, and basic math. But the documentary also highlighted these select math programs in different schools, and these select teachers who are providing additional support to students. It did end on sort of this uplifting note, there’s a lot of interesting and smart people that are thinking about math education in this country. 

LEVITT: Yeah, there sure are. And they’re needed. So within a year of doing this interview, we had launched Data Science for Everyone, this consortium that’s brought together a really wide variety of people who all care about how we teach math and think we should have more data in it. And that organization continues to be really effective. And honestly, we’re making bits of progress. I feel pretty good about that. Do any of the guests we’ve had on this show make an appearance? 

LEVEY: One does! Talithia Williams who is a professor of mathematics at Harvey Mudd.

LEVITT: And then Jo Boaler is in there as well I heard, right?

LEVEY: Jo Boaler is in there. She is in your episode with Freakonomics Radio.

LEVITT: Correct. She’s an educator at Stanford University.

LEVEY: You shouldn’t feel too bad because you are not the only academic that makes a very brief appearance. There are a lot of people in this film—

LEVITT: What makes you think I feel bad? Well, a normal person might feel bad, but I really had nothing to say then. I think if she came and talked to me now, I would have something to say. But back then I didn’t really have anything to say. In fact, I think talking to her helped me really get on the path of starting to have something to say. So in that sense, even though I have a very small role, it was helpful to me, and starting me in launching this school. 

LEVEY: That’s great. You’ll have to watch the film and see your 10 seconds of fame. Listeners, if you have a question for Steve Levitt, please send it to our email, PIMA@Freakonomics.com. That’s P-I-M-A@Freakonomics.com. If you have a question for Ken Goldberg, you can send that to us and we will get it to him and might answer it in a future listener question segment. If you have thoughts or questions about the show, send us an email. We do read every email that’s sent and we look forward to reading yours.

And in two weeks, we’ve got a brand new episode featuring U.C. Berkeley roboticist Ken Goldberg.

GOLDBERG: There is a huge wave, the biggest wave I’ve ever seen in my whole life, of interest in robots, and it’s specifically around the humanoids. And the big proponents of that, namely Elon Musk and Jensen Huang from NVIDIA, are saying that we’re on the verge of achieving this dream, finally. Like Rosie from The Jetsons coming in and clean up our house.

LEVITT: But she didn’t even look that much like a human.

GOLDBERG: Well, that’s true. In fact, she was always kind of breaking down and was always malfunctioning, which is actually the way real robots are.

As always, thanks for listening and we’ll see you back soon.

*      *      *

People I (Mostly) Admire is part of the Freakonomics Radio Network, which also includes Freakonomics Radio and The Economics of Everyday Things. All our shows are produced by Stitcher and Renbud Radio. This episode was produced by Morgan Levey, and mixed by Greg Rippin. We had research assistance from Daniel Moritz-Rabson. Our theme music was composed by Luis Guerra. We can be reached at pima@freakonomics.com, that’s P-I-M-A@freakonomics.com. Thanks for listening.

LEVITT: That’s a conversation my wife and I have all the time, she thinks because I’m such an incomplete person, something terrible must have happened to me as a kid, but I keep on fighting back and saying, “No, I’m just a really incomplete person, absent that kind of trauma.”

O’SULLIVAN: Well, the difficulty she has with doing that is you might just become more incomplete to prove your point, you know.