ROBERT SIEGEL, host:
From NPR News, it's ALL THINGS CONSIDERED. I'm Robert Siegel.
This week, two panels of medical experts recommended that women need fewer screening tests for breast and cervical cancer. Recently, men got similar advice about prostate cancer screening. These pronouncements were all based on medical evidence gathered through research.
As NPR's Joanne Silberner reports, it can be tricky applying that research, what's called evidence-based medicine, to real life.
JOANNE SILBERNER: The concept of evidence-based medicine is pretty simple, says Peter Bach; now a physician at Memorial Sloan-Kettering Cancer Center and once an advisor to the head of Medicare.
Dr. PETER BACH (Physician, Memorial Sloan-Kettering Cancer Center): The basic principle of evidence-based medicine is that clinical decisions that are made between doctors and patients should be driven by data that is accumulated through studies of groups of patients.
SILBERNER: But people don't always want to do what the data say to do, whether it'd be about screening or treatment.
As a family practitioner in North Carolina, Lori Heim sees that a lot, especially with sinus infections. She remembers one patient who demanded antibiotics for her viral sinusitis, even though multiple studies show that's the wrong thing to do.
Dr. LORI HEIM (Family Practitioner; President, American Academy of Family Physicians): The previous doctors gave her antibiotics and she got better. Well, she probably got better because the virus is a self-limiting illness.
SILBERNER: It was hard for Heim to convince her patient that antibiotics weren't the right thing. But she's sympathetic.
Dr. HEIM: They want medicine and science to be absolute and to be perfect. Unfortunately, it's not.
SILBERNER: And because there's always new information and new ways to analyze it, treatment regiments and guidelines are likely to keep changing, she says.
Insurers already use evidence-based medicine to some degree. The current health overhaul bills encourage it without any strict requirements. Economist Uwe Reinhardt says the country needs more of it.
Professor UWE REINHARDT (Economics, Princeton University): You cannot have a health system where every doctor and every patient is completely free to use their imagination and hunches to make clinical decisions.
SILBERNER: Evidence-based researchers estimate that more than one half of current medical practice is based on concepts that have not yet been proven, and there's a history of procedures turning out to be useless. Years ago, many doctors periodically x-rayed smokers' lungs, thinking catching lung cancer early would help. Research showed it made no difference in the course of the cancer.
So people insist on mammograms in younger, lower-risk women, or lung cancer screening for smokers, or other procedures of questionable value?
Prof. REINHARDT: If you want all this health care, pay for it and stop grouching about costs.
SILBERNER: The mammogram recommendations and the prostate and Pap smear recommendations for that matter didn't even consider cost. They only considered the medical risks and benefits.
Getting people to accept guidelines that can change over time is going to take some work, says Alan Garber head of the Center for Health Policy at Stanford University.
Dr. ALAN GARBER (Policy Director, Center for Health Policy, Stanford University): It's not enough for the public to know the results of the studies. When studies conflict, they need to have a better understanding of how we reach the conclusions we did about whether a test or a treatment works.
SILBERNER: Which suggests some long study sessions and talking with your doctor.
Joanne Silberner, NPR News.