Medical misdiagnoses happen, so patients need to be proactive

By Kyung M. Song
Seattle Times staff reporter

	E-mail this article Print this article Search archive

Research in California and Michigan showed cardiologists miss a third of the evidence for significant heart disease.

And in a 1998 experiment led by the University of Chicago, radiologists at a breast-cancer conference on average failed to detect or gave false positive results for 30 percent of the mammograms — including one radiologist who scored just 8 percent correctly.

Diagnosis: Doctors don't always know best.

In fact, misdiagnoses occur much more frequently than patients may realize. Diagnostic errors have triggered more malpractice suits in the past two decades than any cause except improper medical treatments.

Consider a second opinion if: • The diagnosis isn't clear: For instance, your test results are inconclusive. Another doctor may be able to zoom in on a problem that your first doctor missed. • Surgery is recommended: Another doctor may give you other treatment options. • The course of treatment isn't clear: Find out what other experts have to say before jumping into a difficult or debatable course of treatment. • Your treatment doesn't seem to be working: You're frustrated with the lack of progress. Ask your doctor to suggest another physician or find one yourself. You may need to see a specialist. • You want to evaluate your treatment: You may want confirmation from another doctor that the treatment strategy being followed is effective or you may want suggestions about how to enhance your treatment. • Your confidence is shaken: You're uncomfortable with your doctor or don't have confidence in his or her decision-making ability. A second opinion may confirm what the doctor has said or confirm your misgivings. Source: Dr. Pamela F. Gallin, author of "How to Survive Your Doctor's Care: Get the Right Diagnosis, the Right Treatment, and the Right Experts for You" (LifeLine Press 2003)

The true scope of incorrect diagnoses is difficult to measure; some mistakes get buried with patients. But a 1983 study of 100 randomly selected autopsies that ran in the New England Journal of Medicine discovered that physicians had missed a major diagnosis in 22 percent of the dead.

Medicine is too complex for doctors to avoid mistakes 100 percent of the time, says Dr. Doug Paauw, associate professor of medicine at the University of Washington.

"That standard is unachievable," Paauw said. "Even a well-trained and excellent doctor makes errors."

That means passive patients, those who never question, could be jeopardizing their health — perhaps their lives.

To get the most from their medical care, patients should — in addition to describing symptoms clearly, asking questions and following all treatment orders — take action if something doesn't feel right, even if it means second-guessing their doctors.

Paauw estimates that a typical doctor makes as many as 1,000 decisions daily, from authorizing a prescription refill to weighing whether to refer a patient to a specialist. Most of the decisions must be made quickly.

Diagnosis for the most part is a simple task. But some symptoms can leave doctors stumped — or lead them astray.

A patient may have a rare condition the doctor has never seen. Or symptoms may mask the underlying malady.

For instance, a depressed patient may complain of lack of energy or headaches, which could be caused by any number of conditions, but the patient may not describe feeling persistent sadness. Someone with a blood clot in the lungs may describe shortness of breath, which also occurs with asthma and other respiratory ills.

Sometimes, doctors rely on subjective judgments rather than empirical data in making diagnoses. For example, in 1999, medical records of 599 male patients from Palo Alto, Calif., and Birmingham, Mich., were sent to 142 physicians to look for signs for heart disease. Fifty-eight percent of the men had significant heart damage detected by an angiogram, an X-ray exam that involves inserting a catheter into the blood vessel to check for blood flow.

The doctors were given results from patients' treadmill tests to gauge heart function as well as clinical data — but not the angiogram — and asked to estimate the probability of coronary-artery disease. Of the "expert" cardiologists, 69 percent scored accurately, compared with 65 percent for the other cardiologists and 66 percent for the internists. Nearly a third of the estimates were wrong.

Diagnostic roadblocks

Sometimes the trickiest diagnoses can happen when symptoms don't follow the common pattern for a condition. Appendicitis typically causes pain in the lower right side of the belly, where the pouchlike appendix is connected to the intestine. But some patients hurt on the left side, or in the back.

What's more, other ailments can mimic appendicitis. An inflamed lymph node in men or an ectopic pregnancy in women can produce similar symptoms. The complex female anatomy — unlike men, a woman's abdominal cavity contains ovaries, fallopian tubes and the uterus — can make diagnosis even chancier.

In this state, thousands of appendixes are removed unnecessarily each year. Dr. David Flum, a UW surgeon, and three other researchers analyzed nearly 64,000 appendectomies performed in Washington hospitals between 1987 and 1998. They concluded that 9,880 patients, 15.5 percent, did not have appendicitis, meaning the surgeries were unnecessary. Among female patients, 22.8 percent were misdiagnosed, Flum said.

The findings were published in the October 2001 edition of the Journal of the American Medical Association.

Flum said surgeons have traditionally accepted that 15 percent to 20 percent of all appendectomies will turn out to have been unneeded. His research showed that the accuracy rate did not improve during more than a decade despite the introduction of new diagnostic tests such as computed tomography — CAT scans — and ultrasounds.

"We're doing more and more tests, but we're getting the same amount of diagnostic confusion," Flum said.

Flum estimates unnecessary appendectomies cost Americans as much as $1 billion a year, not to mention post-surgery complications and the mental toll of being misdiagnosed. And appendicitis is just one of a multitude of illnesses that are misdiagnosed.

"People often confuse clinical medicine with hard science. But it's not. People interpret data differently, or misinterpret data," Flum said.

Avoiding pitfalls

Paauw, who teaches third-year medical students at the UW, says doctors commonly make mistakes through "premature closure."

Tips for preventing inaccurate diagnoses • Don't wait to see a doctor if you don't feel well. Doctors can't make a diagnosis — right or wrong — unless they examine you. • Be accurate and thorough in describing your symptoms. Keep a written record, if necessary, to remember the details. Getting an accurate medical history is especially critical if the patient is an infant, doesn't speak English or suffers from depression or mental impairment. • Ask your doctor for the medical term for your condition. This will help you research your disease. • Call back for your test results. Don't assume no news is good news. • Ask for a copy of your medical records and check for errors. • Check out the hospital before scheduling your surgery. Hospitals that handle a high volume of particular procedures generally, but not always, are better. Ask the hospital or the doctor about track records. • Don't follow a doctor's orders blindly if your symptoms persist or worsen. If necessary, return to your doctor instead of waiting until the next scheduled appointment. — Kyung M. Song

"Someone sees a pattern and says, 'I know what that is' and jumps right into a diagnosis," said Paauw, who co-wrote a 1996 article for Resident & Staff Physician on ways to reduce diagnostic errors.

Paauw said doctors who don't listen to patients closely and don't get good medical histories set themselves up for errors. It's the same when doctors depend too much on charts and other written data, which can be incomplete, inaccurate or misinterpreted.

Diagnostic tests, too, can be faulty. Dr. Daniel Stryer is a director with the Agency for Healthcare Research and Quality, the U.S. government's lead health-services research arm. Stryer says sometimes simple human error is to blame, as when a pathologist misreads a biopsy result.

But more often, the diagnostic tests themselves are unreliable. For instance, an electrocardiogram, or EKG, used as the first test to detect heart attack misses 43 percent of actual heart-attack cases, Stryer said. A treadmill test can yield more definitive results but isn't wholly accurate, either.

Only cardiac catheterization, which involves inserting a thin tube into an artery or a vein, qualifies as the "gold standard" test that can accurately measure the blood flow, Stryer said.

Many lab tests, including PAP smears and HIV tests, can give false results, positive or negative.

"With every diagnosis, there is an element of uncertainty," Streyer said. "It may sound like a horse, smell like a horse and look like a horse. But in fact, it's a zebra."

When errors are made, Stryer said, doctors more commonly diagnose a condition the patient doesn't actually have than mistakenly give a clean bill of health. That's because "most times, people leave (the doctor) with a diagnosis" of some kind, Stryer said. "We have to put a label on it."

Misdiagnoses are just one type of medical error. Others include mistakes with medications (giving the wrong drug or dosage or in wrong combinations) or errors in treatment (operating on the wrong limb, improper transfusion) or failure to anticipate complications such as infections.

Medication errors in particular are a growing problem. A 1999 report by the Institute of Medicine called "To Err is Human" estimated that 7,400 Americans died in 1993 from drug errors. That's one out of every 131 people who died outside of hospitals and one out of every 854 inpatient deaths. The Institute of Medicine is part of the National Academy of Sciences.

Protecting yourself

Stryer said patients must have trust in their doctors — otherwise, they should look for other physicians — but there is nothing wrong with questioning doctors about the care they give.

The most helpful thing you can do for your doctor is to give him or her a thorough description of your symptoms. Patients with anemia, or a low number of red blood cells, may complain of fatigue or weakness, but leave out the fact that they've been bleeding. Tell the doctor everything.

Use your judgment instead of relying entirely on your doctor's advice. Even if you're not scheduled for a follow-up exam for four weeks, go back sooner if you're worse or your symptoms change.

"Some patients do exactly what the doctor says, and they'll suffer for that month," said UW's Paauw.

If you have any doubt about your diagnosis, get a second opinion. Or a third.

In 1977, Massachusetts began requiring low-income people covered by Medicaid to get a second opinion before elective surgery. Of the nearly 1,600 patients who sought a second opinion, 11.3 percent were advised against surgery. Eighty-two of them sought a third opinion, and 70 percent of the second opinions were reversed.

Many health-insurance plans, even those that require prior authorization for visiting a doctor, will pay for second opinions. Seattle-based Regence BlueShield, for instance, will cover third opinions if the first two conflict. (Check benefit details to be certain.)

UW's Paauw advises physicians themselves to seek second opinions if they're ever unsure about a diagnosis. If your doctor hesitates to refer you to a specialist, call your insurer or switch doctors.

"If the same symptoms persist, find someone you can trust," Paauw said.

Kyung Song: 206-464-2423 or ksong@seattletimes.com

Copyright © 2004 The Seattle Times Company