How Doctors Think (34 page)

"Rachel came back to me, and I felt awful for her," Tepler said. "And she said that she knew I must be angry, that I had told her not to do this. Yes, I said, I had disagreed, but I also pointed out, honestly, that no one can predict what is going to happen in any particular instance." This fundamental truth is too rarely expressed by doctors, and it shows Tepler's humility. Although he has confidence in his own clinical judgment, he accepts that sometimes he may be wrong or that he cannot definitively predict an outcome. In this case, Tepler accepted that there were shades of gray: Mrs. Swanson could have had the tumor successfully removed without the postoperative complications and the explosion of growth in the other metastases; in fact, the resection would have proved to be prudent if in the coming months the tumor had perforated the bowel. The patient's choice, Tepler told me, was in keeping with her character. She wanted to be "proactive" with her cancer. "Understandably, people want the home run," he said. "But often in oncology what we achieve is less than that. And the risk is, by going for the home run, you can strike out."

When Tepler believes that any further chemotherapy is futile, he promises patients that he will be there for them until the end, and he further promises that they will be comfortable with the time they have left. When they press him for numbers—weeks or months—he gently invokes Stephen J. Gould's remark, "The median is not the message."

Many people seek out Memorial Sloan-Kettering based on its well-deserved reputation as a preeminent center for cancer treatment. But often the doctor matters more than the hospital. A friend of mine, an artist in her fifties with bladder cancer who went to Memorial, saw this for herself. She'd had surgery there, and she adored her surgeon. Even after her metastases appeared and there was no reason for further surgery, he visited her in the hospital. She was not a celebrity, not wealthy, so there appeared to be no ulterior motive on his part; she was warm, outgoing, a sparkling person, and the doctor was showing how much he cared, how much he enjoyed her company and the company of her husband, a novelist, by visiting them.

Her distress came from her interaction with her oncologist. He had treated her with what he said was the "best protocol" available, and when her cancer returned after a brief remission, his response to her queries about further treatment left her frightened and paralyzed. I spoke with the oncologist about her case. "She has a projected seven-month survival," he said. "There are no data that any other drugs have more than a ten to fifteen percent response rate—at best." I asked about several drugs in development. "There are phase-two studies," he said, referring to the second phase of evaluation, which is designed to assess the benefit in patients following the first phase, when toxicity is defined. I knew that several patients with bladder cancer had responded well to the drugs in these phase-two studies. "It's much too early to know whether those responses were meaningful," he said, "and no one knows the optimal duration of therapy or the optimal dose." What the oncologist said to me was precisely what he had told the artist and her husband—in his flat, direct way. "She should go home and live out her life. There are no data to support treating her at this point," he concluded.

"I am fifty-six years old," the artist told me. "I am not ready to go home and die in seven months. I have two sons and a husband I adore." She consulted an oncologist at another Manhattan hospital. He gave her one of the drugs that was still in phase-two testing. She had a dramatic response and lived well for more than a year. When the cancer returned, causing a bowel obstruction, she decided that she was ready to die, that there was no real likelihood of sustaining her quality of life. She passed away at home with her family at her side.

"Fundamentally, it's not about the hospital," said Karen Delgado, "although there are those with better support services, better nursing, and more expertise in certain diseases. All of that matters, but what matters most is the doctor. And, I tell people that a physician might be the right doctor for you but not the right doctor for another individual."

Delgado's words rang true. Before George Franklin met Stephen Nimer, he was treated by another specialist at Memorial Hospital. The two did not click; in fact, Franklin and his family took a deep dislike to the oncologist. But a journalist friend of mine who also had an aggressive lymphoma adored the specialist that Franklin could not abide. "There are times I feel like strangling him," the journalist said. "But that's part of why I like him so much. He is incredibly direct. He never pulls any punches. He tells me exactly what he is thinking and why. He can be infuriating, but he is a great doctor for me."

A physician is definitely not great, however, if he abandons any person or family when things go sour clinically. Another friend of mine, who worked in the intelligence services and was a heavy smoker, developed widespread lung cancer in his early sixties. He had retired from the spy business, but prided himself on being an acute observer of people. Yet when he fell ill, he became blind to reading certain doctors' personalities. He was hell-bent on being cared for at Memorial Sloan-Kettering, convinced that some magic there could reverse his dire disease. He finally secured an appointment with a young doctor on the staff, whom at first he found charming. But when the lung cancer grew after several cycles of intensive chemotherapy, the oncologist wouldn't return his calls. When he was admitted to the hospital with complications, the oncologist spent a few fleeting minutes at his bedside and then seemed to disappear entirely. His office said that he was traveling a lot. My friend was in the hospital for days without his oncologist visiting or even telephoning. He was emotionally devastated, afraid and alone. My friend ultimately sought an outstanding oncologist in his hometown in New Jersey who was attentive and made sure that his final days were as comfortable as possible.

My novelist friend theorized that the oncologist who treated his artist wife but would not consider treatment beyond statistics and protocols, and the oncologist who abandoned the intelligence operative, both suffered from fear of failure, and probably fear of death. "I know it sounds strange," he said, "supposing that an oncologist who sees so much death would flee from it. But I think posing as highly rational, acting only when all the numbers are in hand, is in fact an irrational way to care for people with cancer. You refuse to try anything creative, refuse to put yourself on the line. He must have known that we would leave him, that we would seek another doctor at this most difficult point in our lives, when we were facing death. It is a more subtle form of abandonment than what happened to your friend with lung cancer."

This is a fundamental schism in the field of oncology, between those who are driven almost entirely by data and those who are willing to treat patients outside of proven protocols. Sometimes veering too far from widely tested therapies can result in unnecessary toxicity and suffering. But I found a deep resonance in what the novelist said—that what appeared to be a rational form of thinking was actually irrational when applied to a patient's needs and goals, and might reflect the emotional state of the oncologist more than the clinical needs of the patient.

Nimer and Tepler try to understand a patient's character and factor this understanding into their clinical judgments. My novelist friend showed me how patients and their families can understand their oncologists' character and weave that understanding into their decisions. People with cancer and other serious diseases can face a dizzying array of choices. Which path they take pivots on clinical facts and the dimension of character—their own and their doctors'. This applies not only to oncology but to all of medicine, a mix of science and soul.

Epilogue: A Patient's Questions

L
ET'S IMAGINE
you are sitting in a doctor's office. For several weeks you've had a symptom that has not gone away—say, discomfort in the center of your chest, beneath the breastbone. The doctor has taken the history, performed a physical examination, and ordered some tests. He reviews with you the information he has gathered and believes that you have acid reflux, a common problem when the irritating juices from the stomach move up into the esophagus.

In most cases, a physician arrives at the correct diagnosis and offers appropriate treatments. But not always. If, after a while, you are not getting better, the discomfort persists or has worsened, then it is time to rethink the diagnosis. Recall that most misguided care results from a cascade of cognitive errors. Different doctors have different styles of practice, different approaches to problems. But all of us are susceptible to the same mistakes in thinking.

How to make the correct diagnosis? There is no single script that every doctor or patient should follow. But there are a series of touchstones that help correct errors in thinking. Doctor and patient will start again searching for clues to solve the problem. The first detour away from a correct diagnosis is often caused by miscommunication. So a thinking doctor returns to language. "Tell me the story again as if I'd never heard it—what you felt, how it happened, when it happened." If he doesn't ask you to do this, then you can offer to retell your story. Telling the story afresh can help you recall a vital bit of information that you forgot. Telling the story again may help the physician register some clue that was, in fact, said the first time but was overlooked or thought unimportant. This will prompt him to look in new directions for answers.

These days, when we are not getting better, most of us return to see the doctor with ideas about what might be wrong. Our notions sometimes come from knowing a friend or relative with a similar symptom, or ideas may have been sparked by looking on the Internet. Our thoughts about our unrelieved symptoms often focus on the worst-case scenario. Such self-diagnosis is a reality that neither patient nor physician should ignore. Since the doctor may not address it, you should. "I'm most worried that what seemed like acid reflux could be the first sign of cancer," one patient might say. Or another might recount to the doctor how her friend was told she had indigestion but it was actually a brewing heart attack. For some, articulating such fears is exceedingly difficult to do because of magical thinking—the notion that saying it might make it real. I recall one middle-aged woman with discomfort in the chest whose face was a mask of worry when we were searching for a diagnosis. "Tell him what is really frightening you," her husband said with loving firmness. A relative had died of a pulmonary embolus, and she was terrified that this was the cause of her chest pain. After she told me, she admitted that she'd been scared to say it, since doing so might make it true.

A thoughtful doctor listens closely to these worries. Alerted to your deepest concerns, he may be prompted to ask more probing questions, to have you describe your symptoms in greater detail. This expands the breadth of your dialogue with him and removes inhibitions that could hide clues.

But the answer may not be revealed quickly by a fresh dialogue. The doctor may need to repeat your physical examination, focusing more intensively on one or another part of your body. Or he may begin to doubt the value of a particular laboratory test, or the reading of your x-ray. As we've seen throughout this book, physicians tend to go with their first impressions. The initial biases in a physician's thinking are often reinforced by his selective survey of diagnostic data. We all are inclined to seize on an apparently positive finding and ignore what may be negative and contradictory.

Sometimes he may need to repeat laboratory tests and sophisticated scans. This can be costly. In the current environment of medical practice, repeating tests is strongly discouraged as not being cost effective. The imperative from hospital and managed care administrators is to be economical. And arriving at the correct diagnosis may not require actually repeating tests, only doubting them. As we saw, there can be significant differences in how different radiologists read the same image, how different pathologists assess the same biopsy. Revisiting the diagnosis means the doctor returns with a sharp and discerning eye to inspect all the results to date—blood tests and x-rays and pathology reports.

Yet there are times when repeating a test is essential. There are instances when the first CT scan was not correctly calibrated, as Dr. Herb Kressel recounted in the case of the woman with chest pain believed to have a pulmonary embolus but actually suffering from a tear in her aorta. There are times when the first biopsy misses the lesion. In my field of hematology, more than one bone marrow examination may be needed to find a malignancy like a lymphoma, because tumors are not uniformly present in the bones, and I may have placed the biopsy needle in an area of the marrow that did not contain the tumor. After review or repetition, the tests still may not give the answer.

"What else could it be?" is now the question you or your loved one should ask the physician. The cognitive mistakes that account for most misdiagnoses are not recognized by physicians; they largely reside below the level of conscious thinking. When you or your loved one asks simply, "What else could it be?" you help bring closer to the surface the reality of uncertainty in medicine. "What else could it be?" is a key safeguard against these errors in thinking: premature closure, framing effect, availability from recent experience, the bias that the hoofbeats are horses and not zebras. Each cognitive error constrains the pursuit of answers, and correcting the error helps the doctor think of a test or procedure that he didn't previously consider and can make the diagnosis.

"Is there anything that doesn't fit?" may be your next question. This follow-up should further prompt the physician to pause and let his mind roam more broadly. He will begin to survey more of the clinical territory, aided by a vision that comes from doubt. "Is there anything that doesn't fit?" was the question underpinning Rachel Stein's insistence that her infant daughter Shira's atypical case might not be atypical at all, but something altogether different.