Your doctor has just ordered up yet another test. Is it really necessary, or is it another instance of "defensive medicine"?

According to a Harris Interactive poll released last week, nearly 80% of physicians say malpractice fears lead them to order more tests than are medically needed.

More than half said they suggest invasive procedures such as biopsies more often than they would based on their professional judgment alone.

But malpractice fear isn't the only factor driving unnecessary tests. Last year, health-care fraud accounted for more than half the $1.5 billion the Justice Department collected from civil fraud cases, including doctors and medical groups who billed Medicare for unnecessary lab tests from blood draws to heart probes.

While the debate over malpractice reform rages on, it's often the patients who must decide whether to undergo procedures that may be invasive, uncomfortable, risky, expensive or all of the above. An MRI, or magnetic resonance imaging, often isn't useful for diagnosing or treating back pain, but what if the doctor suggests one, just to rule out a tumor?

It isn't all the doctor's fault: Consumers tend to want whatever new technology is available, from whole-body scans advertised on billboards to disposable "gut cams" that patients swallow like a pill to painlessly survey the digestive system. And as anyone who has ever been diagnosed with a serious disease knows, defensive medicine isn't a bad thing if it catches illness early. If that suspicious mole turns out to be cancerous, thank heavens a slide of it went to the pathology lab.

But when retiree George Mueden returned from a cruise in the Baltic with viral symptoms, his doctor ordered up $400 of blood tests. Like most viruses, it ran its course in a couple of weeks. "I can't help but wonder if all that [testing] was necessary," Mr. Mueden says.

In many cases, there are simply no hard and fast rules about which tests are necessary. There isn't even agreement on who should get a mammogram or a PSA test for prostate cancer. "It's easy to define retroactively what was an unnecessary test, but to define it prospectively is much tougher," says Carolyn Clancy, the new director of the federal Agency for HealthCare Research and Quality.

Dr. Clancy advises patients to take an active role in determining whether a test is right for them or a family member. "Patients need to ask clear questions about whether a test is necessary, whether it is necessary now, or whether it is better to wait and see what happens." It helps to research the tests your doctors have ordered.

AHRQ's Preventive Services Task Force, which develops testing and screening guidelines based on the latest medical evidence, is updating guidelines for major screening tests and plans to publish them on its Web site (www.AHRQ.org³) as soon as each is completed. Last week, it released the latest recommendations for diabetes and cervical cancer. Among its new recommendations, for example, are that any sexually active woman be tested by Pap smear for cervical cancer within three years of onset of sexual activity or age 21, whichever comes first, with a follow-up every three years. But AHRQ says women who are over 65 or have had a hysterectomy shouldn't be routinely tested.

Lab Tests Online (labtestsonline.org⁴), a site for consumers developed by the nonprofit American Association for Clinical Chemistry and other professional lab groups, provides detailed reports on a wide range of medical tests. It also offers guidance on how tests are interpreted and how to evaluate new home medical tests.

Because many tests are minimally invasive and present few risks, patients often go along with a doctor's suggestion of extra bloodwork or a urinalysis to screen for, say, high cholesterol or metabolic disorders. But many routine tests haven't been subjected to rigorous research about their usefulness. "Even if tests are inexpensive and noninvasive, if the results are abnormal, you are looking at more testing and more invasive procedures," says Robert Shmerling, a rheumatologist at Beth Israel Deaconess Medical Center in Boston and associate professor at Harvard Medical School.

Tests can often be wrong, too. Stress tests, for example, often suggest there is cardiac disease when there is none, or turn up a normal result when someone does have heart disease, says Dr. Shmerling, who offers consumer advice on medical tests for the online health joint venture of Harvard and insurer Aetna (www.intelihealth.com⁵).

Medical sites such as MedlinePlus.gov⁶ are useful for researching diseases, as are academic medical centers and teaching hospitals that specialize in research and treatment. Johns Hopkins University, for example, has a Vasculitis Center whose site (vasculitis.med.jhu.edu⁷) contains details on all the tests required to diagnose the disease, which effects the blood vessels. Other disease-specific sites such as Spine-health.com⁸, a site started by a Wisconsin orthopedic and reconstructive spine surgeon, provide useful information on when to get that MRI for back pain.

Dr. Clancy of AHRQ also urges patients to keep records of previous tests they have undergone, particularly if they have changed doctors or health plans or moved to a new community. Tests are often most meaningful when they can be compared against an earlier or baseline test to determine if there have been any changes in a lab reading or diagnostic image. "The most clearly useless test is the one that's already been done and you didn't keep a record of," Dr. Clancy says.

• Send comments to Informedpatient@wsj.com⁹
 

Updated February 13, 2003