The neurologist Jean-Martin Charcot holding the arm of a female patient suffering from locomotor ataxia, Hôpital Salpêtrière, Paris, circa 1885

Adoc-photos/Art Resource

The neurologist Jean-Martin Charcot holding the arm of a female patient suffering from locomotor ataxia, Hôpital Salpêtrière, Paris, circa 1885

Several years ago, my right wrist became swollen and inflamed. My primary care physician ordered blood tests and X-rays, but the cause remained obscure. Empirical treatment with a splint and anti-inflammatory medication did not improve my symptoms, and so I was referred to a hand surgeon. He ordered further tests, including a bone scan, which evaluates not only the wrist but all of the bones in the body. That night, the surgeon called me at home.

“I just saw the results of the bone scan,” he said. “The wrist is not your major concern. It looks like there are multiple metastases in your ribs. You’re an oncologist. You need to speak with one of your colleagues about what to do next.”

I hung up the phone in shock, and within minutes my ribs felt as if they had been hit by a hammer. I lay down and took deep breaths, but the pain did not abate.

My wife, also a physician, was away on a ski trip. After several hours of phoning, I finally reached her. She tried to be reassuring, saying what I already knew, that bone scans can produce artifacts, suggesting disease where none exists. First thing in the morning, she said, I should have X-rays of my ribs; if there really were multiple cancer deposits, they would be obvious.

I was unable to sleep. Although I realized that I had had no discomfort before the call from the surgeon, I couldn’t shake the sense that the accelerating pain was confirming what the bone scan had found. And as a cancer specialist, I knew the implications were dire. Few tumors that have metastasized to bones can be cured.

Early the next morning, I arrived at my hospital’s radiology suite. X-rays of the ribs showed nothing abnormal. Slowly, the pain subsided.

I didn’t expect my reaction to the surgeon’s telephone call. During training, medical students often become hypochondriacal, developing symptoms of a disorder like Hodgkin’s disease after learning about it. I didn’t. And in one of our classes, a psychiatrist demonstrated techniques of hypnosis, selecting me as a subject. I proved not to be “suggestible,” a person who can be easily hypnotized. But after the incident with the deceptive bone scan, I experienced how powerful the mind can be in generating bodily symptoms.

Suzanne O’Sullivan is a neurologist specializing in epilepsy who practices in London. Many of her patients suffer from so-called conversion disorders: somatic symptoms caused by psychological distress that defy ready diagnosis by medical tests or physical examination. “They are medical disorders like no others,” O’Sullivan writes. “They obey no rules. They can affect any part of the body…. Almost any symptom we can imagine can become real when we are in distress.”

Physicians who practice family medicine, pediatrics, or internal medicine learn that a substantial proportion of people seeking care have inexplicable complaints. Some surveys indicate that at least a quarter of such patients report symptoms that appear to have no physical basis, and that one in ten continues to believe that he has a terminal disease even after the doctor has found him to be healthy.1

Understandably, because the symptoms obscure the psychological genesis, patients seek a physical disorder to explain their condition, and turn to doctors like O’Sullivan to provide a diagnosis. Her findings are striking:

My first consultant post…saw me running a service whose main purpose was to investigate people with epilepsy who were not getting better with standard treatment. It transpired that approximately 70 percent of the people referred to me with poorly controlled seizures were not responding to epilepsy treatment because they did not have epilepsy. Their seizures were occurring for purely psychological reasons.

While not a psychiatrist, O’Sullivan proposes that their collapse and convulsions “happen for a reason. When words are not available our bodies sometimes speak for us—and we have to listen.”

That listening is no longer valued in today’s medicine. The patient’s “history” was once the centerpiece of his medical record, his story written in narrative form. With current electronic templates, information is fragmented into chunks designed to meet so-called quality metrics and maximize revenue from insurers. The patient’s story has been reduced to telegraphed key words that trigger prefigured algorithms, which generate pop-ups on the computer screen for further testing or generic therapies. O’Sullivan bemoans similar changes in the British medical system.

Under the time pressure of factory-like care, the physician uses a checklist rather than talking with patients in an open-ended way; in fact, some doctors often skip taking a fresh medical history and simply cut and paste the initial evaluation from the electronic record.2 This shift in the physician–patient interaction limits the kind of deep inquiry vital to diagnosing psychosomatic illness. “There are always two realities,” O’Sullivan writes: “the one which exists in the notes and the one which lives in the patient’s memory. I needed to know both and I knew that neither version could be wholly relied upon.”


Patients with imaginary illnesses are denigrated by many doctors. During my internship at the Massachusetts General Hospital, I was introduced to a new vocabulary beyond technical medical terms. Among the words were “crock” and “turkey.” These were patients who offered an array of complaints likely to be psychosomatic in origin. Pulled in every direction to urgently attend to others with heart attack and sepsis and pulmonary embolism, doctors had scant patience to deal with such individuals. Although I resisted using such language, I still found myself, desperate for time, deeply resenting these patients’ requests for attention. “Psychiatric disorders manifesting as physical disease are at the very bottom of that pile,” O’Sullivan notes. “They are the charlatans of illnesses. We laugh at them.”

O’Sullivan has a wide range of references to explain the plight of those with imaginary illness. Cassandra was the daughter of the king of Troy, both blessed and cursed. “Her blessing was one of prophecy, Cassandra could foresee the future. Her curse was that she was not believed. That is how people with psychosomatic disorders feel. Their suffering is real but they do not feel believed.”

That belief goes beyond the doctor and includes the patient:

It can be very difficult for a patient to accept that they suffer from a conversion disorder (a medically unexplained neurological symptom) when that assumption is based almost entirely on what is missing. It requires great trust between patient and doctor. Every week I tell somebody that their disability has a psychological cause. When they ask me how I have come to that conclusion, all I can provide is a list of normal test results, evidence for ruling out diseases. When a person is paralyzed or blind or suffering from convulsions, it is not difficult to see why they find that a very unsatisfactory explanation.

O’Sullivan illuminates one of medicine’s most fraught moments, when a physician reaches the conclusion that there is no physical (or “organic”) disease. Matthew is a desperate patient convinced that he suffers from neuropathy. How definitive should O’Sullivan be in challenging this false notion? “Was I as sure as I said I was that Matthew did not have a disease? Should I simply have agreed with him that nothing is ever unequivocally certain?” She decides to be definitive in her assessment:

Any shred of possibility that a physical disease had been missed offered him hope that his illness was not psychological and he might cling to that. If I allowed him any glimpse of my doubt I could be sending Matthew on a quest for a disease that might easily take up a lifetime.

Matthew rejects his general practitioner’s diagnosis of psychosomatic illness and searches the Internet, which offers up a host of alternative causes: diabetes damages nerves and could result in the pins and needles he feels; trapped nerves might cause his symptoms; poor circulation might result in his discomfort. He consults a chiropractor, who wonders if Matthew might have a slipped disk in his neck. Soon, the areas of numbness spread from Matthew’s limbs to his trunk; paralysis follows blurred vision:

As I listened I tried to spot an anatomical pattern that would explain everything, but all I could see was that what Matthew was describing was impossible…. But at the same time, I wondered if he was simply elaborating on a simpler story. Maybe he had an organic neurological problem and it was being magnified and contorted somehow by the depth of his concern. So I kept listening.

This is precisely what competent clinicians learn to do: keep listening and reexamine initial assumptions. Ultimately, O’Sullivan is confident in her assessment of psychosomatic illness. But again, she wonders if she should leave room for doubt. Medicine, after all, is an inexact science:

The mistake of offering a patient an organic diagnosis just in case has led to many people suffering lifelong seizures with no abatement. It happens for a number of reasons. Doctors are frightened to face the almost inevitable anger that will occur when a psychosomatic illness is mentioned. But protecting the patient from that upset is not in their interest in the long term if they are being denied a diagnosis. Also, doctors worry about calling a symptom psychological and discovering later that there was an organic cause after all. Calling an organic problem functional [psychological] is a mistake that is guaranteed to engender anger in a patient and their relatives, and can lead to a lawsuit.

Some years ago, a colleague introduced me to a woman I’ll call Anne Dodge.3 She had lost count of the number of doctors she had consulted for her abdominal pain and weight loss. Twice she had been hospitalized in a psychiatric unit for an eating disorder. Her primary care doctor refused her request for further opinions, convinced that her complaints were entirely a reflection of mental distress.


At the prompting of her boyfriend, Anne sought out my colleague, a specialist in gastroenterology. He listened carefully to her long history of abdominal pain, reviewed the numerous tests and procedures she had undergone, and, most importantly, kept an open mind. It was likely she suffered from a psychosomatic illness, he thought, but there also could be an underlying physical malady. Occam’s razor is a scholastic principle taught in medical school: try to settle on a single explanation for a multiplicity of symptoms. But this injunction occasionally is contradicted by the reality of human biology. My colleague discovered that she had celiac disease. Indeed, a large proportion of people who suffer with this autoimmune disorder linked to gluten are initially told that the problem is in their head, not their gastrointestinal tract. Further, he concluded that she was indeed depressed and had become phobic about food, but who wouldn’t be after years of consulting physicians and being told that it was all imaginary?

O’Sullivan refers to a similar experience when she concluded that a patient’s complaints of chronic pain and arm weakness were psychosomatic. A brain scan proved otherwise:

There, right in the middle of the image, superimposed on the gray of the brain, was a white circumscribed ball of tissue that most certainly should not have been there. Fatima had a brain tumor, and it sat in just the place that when compressed would lead to weakness of the arm.

I have thought of Fatima often since that day. I use the memory of her to remind myself that a clinical suspicion is only that, an unsubstantiated opinion. A doctor forms a medical diagnosis in part based on knowledge of disease, but much is also drawn from the qualitative nature of the story that a patient tells. Doctors struggle when a patient’s complaints or level of disability seem to outstrip what they can find on examination. We expect people to complain only in proportion to our idea of their illness.

Our deepest learning as doctors occurs when we are wrong, when we think in a skewed way and jump to conclusions. Diagnostic error, alas, is hardly a rare occurrence. It is estimated that some 5 to 15 percent of patients are never diagnosed correctly or that there is a significant delay in the diagnosis, resulting in harm.

Drawing by Edward Gorey

Edward Gorey Charitable Trust

Drawing by Edward Gorey

O’Sullivan provides a lucid summary of the history of hysteria, told through the work in the late nineteenth century of the famed neurologist Jean-Martin Charcot at Hôpital Salpêtrière in Paris. He stood on a platform beside women who then were hypnotized and had been prepared to convulse as he explicated the stages of the seizures:

He showed the audience the depth of the women’s detachment from their surroundings by asking them to partake in activities to which they would never agree in their fully conscious state. Women undressed or crawled around the room on hands and knees like a dog. He experimented with metallotherapy, using magnets to shift symptoms from one part of the body to another or even from one woman to the next. Convulsions could be transferred from the right arm to the left. Even more incredibly, an affliction could be taken from one woman and given to the one beside her.

Charcot’s clinical presentations attracted not only physicians, but “the fashionable and elite of Paris,” O’Sullivan writes. “There were artists and actors and even the occasional royal. When the phases of la grande hystérie were described, le tout Paris were present to hear them.” Fortunately, there are now legal and ethical constraints that prevent such voyeuristic displays.

Despite hysteria defying the anatomy of the nervous system, and its demonstrable social contagion, Charcot regarded the malady as a hereditary brain disease. Moreover, O’Sullivan notes, he asserted that “hysterical phenomena could be induced to appear or disappear instantaneously just through applying pressure to the ovary.” Charcot also diagnosed men with hysteria, and claimed that pressure on their testes produced the same beneficial effects as pressure on the ovary.

The publicity about Charcot’s presentations resulted in a sharp rise in cases of hysteria. O’Sullivan writes:

By bringing the scientific study of hysteria to the fore, Charcot created a plague of hysterical seizures that quickly spread to all of France and then throughout Europe. In a single year he alone saw more than three thousand patients, eight hundred of whom were diagnosed with hysteria. The late nineteenth century was the age of hysteria.

Charcot died in 1893; detractors then disparaged his views on the disorder, and the epidemic was stemmed. O’Sullivan notes that “sufferers disappeared into the shadows.”

In contrast to Charcot, Sigmund Freud posited that psychological trauma was converted to physical symptoms. In his studies with Josef Breuer, hysteria was associated with mnemic symbolism. The case of Anna O. (Bertha Pappenheim) is an example: she suffered from severe cough, paralysis of her right arm and legs, and altered vision, hearing, and speech. This was attributed to her unexpressed resentment over the trauma of her father’s death. O’Sullivan embraces this approach: a person who experiences an insult as if it were a slap in the face might develop severe facial pain; a woman who “swallows unkind words, or a truth that she is not permitted to say,” might then become mute, or feel something is stuck in her throat.

O’Sullivan recounts a personal experience akin to mine after the telephone call from the hand surgeon. She visited an elderly friend, and as she entered his home, “a wave of odor hit me, a slap of wet dog directly in the face.” The living area was filled with old newspapers, the kitchen countertops held unwashed crockery. Two dogs sat on the sofa and, after they were shooed down, she took their place. She writes:

Even now I can feel vividly the discomfort I felt that afternoon. My imagination was attributing life to that sofa that it did not contain. My skin was so invaded by itching that when my friend left the room for a moment I had to stand and shake out my clothes looking for imaginary insects. Even when I went home I could not escape the feeling of a thousand flea bites. Only when I had washed my clothes and showered did I feel any relief.

She realized that there were no insects in the sofa, that nothing was crawling on her skin, nothing nipping her:

But the itch and tickle felt absolutely real. My mind had produced real physical sensations triggered only by an idea. Even with the evidence of my eyes that saw no fleas I simply couldn’t shake the imaginary feeling of being bitten. And even though it was years ago, and all in my mind, I am experiencing it all over again now just by remembering it.

O’Sullivan’s writing is at its best when she provides a vivid picture, like that of the unkempt house and dog-soiled sofa. But much of the book reads as if she were telling a colleague about a case or explaining to a patient her thoughts about psychosomatic disorders. Moreover, most of the patients we meet have convulsions caused by a conversion disorder. This is understandable, given her specialty, but the reader feels as if he or she has already covered this clinical territory as the pages unfold.

Freud advocated a “talking cure” for conversion disorders; others pursue a cognitive behavioral approach. O’Sullivan regularly refers her patients to psychiatrists:

There is no single solution to psychosomatic illness. To look for one is akin to looking for the cure for unhappiness. There is no single answer because there is no single cause. Sometimes you just have to figure out what purpose the illness serves, find what is missing and try to replace it.

This sounds somewhat Freudian, but no further detail is given on how to accomplish such findings and replacements.

For those who have never had a psychosomatic symptom, and doubt that emotion can dramatically alter physiology, O’Sullivan cleverly invokes a universal experience:

How easily we accept…different facets of laughter. It is a physical display of emotion, its mechanism is ill-understood, it is not always under our voluntary control, it affects our whole body, it stops our breathing and speeds up our heart, it serves a purpose, it releases tension and communicates feelings. Laughter is the ultimate psychosomatic symptom.

Our mind can forcefully speak through our body not only in distress, but in joy.