Cover Story

Coping with clinical uncertainty

Ever fret over decisions you made in the inexact science called medicine? You're not alone.

By Robert Lowes
Senior Editor

Generals talk about making decisions in the "fog of war." Physicians practice in the fog of medicine, where the only certainty is uncertainty. Often they must settle for a probable diagnosis. What's more, each treatment option comes with its own probability of success, side effects, and fatality.

Medical education aims to prepare future doctors to walk on the thin ice of tenuous conclusions. "It's important for medical students to learn about the limits of science, as well as their own limits," says pediatric gastroenterologist Robert Rothbaum, a professor at the Washington University School of Medicine in St. Louis. "If there's no doubt, there's no self-examination, and you may end up overestimating what you know, or what you can do."

For some physicians, though, educators make uncertainty a dirty word. "In medical school, it was implied that you should always know the answers," says internist Valencia Clay in Chattanooga, TN. "You got the impression that saying 'I don't know' was a shameful thing."

The doctor-on-a-pedestal model of medicine also makes it hard for doctors to voice their doubts. Patients naturally prefer comfort and reassurance to any talk about percentages. Many physicians assume that anything less than supreme certitude on their part will discourage them and set back their progress.

In the end, doctors have no choice but to live with unease. Some, such as ED physician Tony Dajer in New York City, say that doubt makes them better doctors, because they're more prone to look at a test result a second time or consult a colleague. "The most dangerous doctors," says Dajer, "are the ones who aren't uncertain."

Here are the stories of five physicians and how they find their way in the fog of medicine. If nothing else, their experiences offer reassurance that you're not the only one losing sleep.

Fifty doctors, an autopsy, and still no answer

Internist Frank Sonnenberg of New Brunswick, NJ, has grappled with the issue of clinical uncertainty more than most doctors: He's editor of the journal Medical Decision Making. But for all the scholarly papers he's written or edited, one of the biggest insights he's gained about the limits of human knowledge came from his first year of residency at an academic medical center on the West Coast.

"I admitted a middle-aged gentleman with abdominal pain," says Sonnenberg, a professor of internal medicine at the Robert Wood Johnson Medical School. "Turns out he had GI hemorrhaging due to ulcers. We got the hemorrhaging under control, but for some unknown reason, he started to deteriorate rapidly. Every organ system in his body shut down, and he went into a coma.

"We called in cardiologists, rheumatologists, infectious disease specialists. There must have been 50 doctors who worked on his case at one time or another. But nobody could sort out what was wrong. It was a collective puzzlement.

"The man finally died. We did a detailed autopsy, but we still couldn't make a diagnosis.

"The case was disappointing, but it taught me a valuable lesson. Even the best doctors don't always figure out what's going on."

An educated shot in the hospital dark

Valencia Clay, too, was rudely introduced to medical uncertainty during residency training nine years ago. But in her case, pluck—and perhaps luck—led to a happy outcome.

"One of the patients on my floor coded, and by the time I got to her room, she was already intubated," says Clay, a former professor at Morehouse School of Medicine who now works for a disability insurer. "But, the nurses were still struggling to get a good connection on the heart monitor leads. The fact that she was obese might have made it difficult to get a reading. I couldn't make a good connection, either.

"The trouble was, I was taught to order a set of medications for a coded patient depending on the heart rhythm I saw on the monitor. But I couldn't see hers. So what was I going to order?

"It finally occurred to me that the first medicine in every code was epinephrine. So that's what I ordered. I had to do something; the woman wasn't breathing. This wasn't a med school problem where I could gather up all the information and think about it for two hours.

"Fortunately, a cardiologist intervened and attached the leads to the back of the woman's shoulders. We got a rhythm, and it seemed to respond to the epinephrine—I can't be sure. But we saved the woman's life.

"Later on, the cardiologist told me, 'You did what you had to do with the best information available.' I still felt guilty about not knowing where to put the leads."

It wasn't until several years into private practice that Clay realized she couldn't possibly know everything in medicine. "My way of coping is to get as many brains as possible involved in curbside consults," says Clay. "When a specialist tells me, 'I don't know,' I feel a little better about my own ignorance."

Clay says she's honest about the medical uncertainty of a situation when she talks with patients. "When I tell them about treatment options, I don't cover up the risks," says Clay. "If I said, 'You'll do great, you won't have a problem,' I'd lull them into a false sense of security. I'd give them the impression that doctors are perfect, the body is a machine, and everyone responds to treatment the same way."

This FP got beyond the obvious diagnosis

FP Peter McGough in Bellevue, WA, is the first to admit that he doesn't know it all.

"A family physician's knowledge of medicine is broad, but on any one issue, a specialist's knowledge is deeper," says McGough, a former president of the Washington State Medical Association. "I'll always remember what one of my mentors said in residency training: 'Know your limits. Know when to get help.' If I haven't come to a clear diagnosis after three visits, that's often when I'll bring in a consultant."

Another lesson McGough learned early on was the value of clinical uncertainty—it can keep a doctor from jumping to conclusions. That lesson was underlined for him two decades ago when a bright yellow, obviously jaundiced woman came to the walk-in clinic where he practiced.

"We ran a urine test and found elevated levels of bilirubin," says McGough. "I told her that I thought it was hepatitis A, recommended rest and dietary changes, and scheduled her for a follow-up visit a week later. We'd have some blood work results by then.

"Instead, she came back in two days, this time with her husband. She was still jaundiced, but she now had low blood pressure and shortness of breath. That didn't fit hepatitis A.

"So I went through her medical history again and asked if anything unusual was going on when she noticed her skin turning yellow. Her husband piped up, 'Tell him about the chest pain you had last week.'

"The woman revealed that she had been overcome with chest pain during a baseball game and had to sit down for an hour before she could walk again. We did an ECG and discovered she'd had a massive heart attack, which apparently damaged her liver. We admitted her to the hospital immediately. Unfortunately, she died.

"Sometimes the diagnosis that's staring you in the face isn't the right one."

The woman, adds McGough, never mentioned her chest pain during the first visit, even though she was quizzed about heart problems. Her silence on that subject points to another lesson about clinical uncertainty: A doctor's diagnosis is often only as good as the information that the patient provides.

Perfect surgical execution, imperfect patient anatomy

General surgeon Fernando Ugarte in Marysville, KS, earned his medical degree in 1965. But, after all these years, he still has to remind himself that good care doesn't necessarily lead to good outcomes.

In the process, he's had to unlearn some of what his teachers taught him during residency. "We had a weekly morbidity and mortality conference, and every complication and death was reviewed," says Ugarte. "If a patient had an appendicitis and the wound became infected, everyone assumed that it was your fault. The truth is, wounds can become infected regardless of what you do.

"I wish somebody had said that some outcomes have nothing to do with your actions. That would have kept me from beating myself up."

Ugarte was tempted to do just that after he performed a laparoscopic cholecystectomy on a young woman recently. Two days after the operation, the patient complained of abdominal pain. The problem was a bile leak

"I wondered, 'What did I do wrong? Did I cut a bile duct?' I put drains in her abdomen for the bile leak and pondered my options. Part of me wanted to open her up, find the leak, and fix it. But I practice in a rural area, and I didn't have the surgical backup for this kind of procedure.

"I was very bothered, but I didn't tell the patient how I felt. I told her that outcomes are unpredictable, no matter what we do.

"I referred her to another surgeon for exploratory surgery. He gave me permission to watch, so I drove three hours on a Sunday to Omaha.

"We were in for a big surprise. The bile duct had indeed been injured during surgery, but only because it was in an abnormal location, between the gallbladder and liver. Nobody would have been able to see it with the laparoscope. It was almost microscopic. The other surgeon told me, 'This is a very rare bird.' He fixed the leak. Now the lady's okay.

"I was very happy. I'd done nothing wrong after all. I had just encountered something unexpected."

Too much worry is not productive

Oklahoma City internist Mary Ann Bauman woke up five times during the night a few months back. She had just given her first Botox injection earlier that day to a woman in her 50s. And now Bauman had a bad case of the rookie jitters.

"The injection was between her eyes, and I was afraid of a complication, like one of the eyebrows drooping," says Bauman. "It wouldn't have been permanent, but I still worried about it."

In the end, the Botox patient had the rejuvenated face she wanted, and Bauman was able to sleep easier. Medical insomnia generally isn't a problem for Bauman, because she has trained herself to accept the unpredictable nature of medicine. She walks the fine line between a little worry, which is constructive, and excessive worry, which is paralyzing.

"I need to be conscientious, to keep things from falling through the cracks. If I'm not sure about a new diagnosis I made, I'll call the patient the next day and ask how he's feeling. Or I'll schedule an office visit for the following week. And I'll ask my colleagues, 'Would you do this test?'

"However, too much worry is not productive. If I really thought long and hard about each decision I make in the course of treating 15 to 20 patients a day, I'd be frightened.

"It's so easy to be down on yourself. When a patient has a bad outcome, I undergo this huge self-scrutiny. I call it putting myself under the retroscope. 'I should have done this. If I had just acted more quickly.' However, this second-guessing isn't always accurate. It's possible I could have done nothing to prevent the bad outcome.

"I question myself a lot when a patient dies. I find it interesting, though, that while I'm battling guilt, the family members thank me for taking care of their loved one. The truth is, they recognize more than we do that nothing is guaranteed in medicine, and that doctors are doing their best."

Of course, other patients do expect perfection, and file suit over anything less. The fear of litigation as well as an understandable unease with medical uncertainty prompts many doctors to err on the side of caution, says Bauman.

"We run extra tests or hospitalize a patient to monitor him. We feel better as a result, but maybe we're not being cost-effective. We're treating our anxiety, but not necessarily the patient. The high cost of health care is a national crisis, and sometimes we're part of the problem."

For further reading

 

Robert Lowes. Coping with clinical uncertainty. Medical Economics Oct. 24, 2003;80:53.