When a study published this spring showed that Avandia, a drug that lowers diabetics' blood sugar, also increased their risk of heart complications, Dr. Mary M. Newman, a Lutherville internist, was in a quandary about what to advise her patients taking the medication.
"It was easy for patients to feel the drug was proven to be dangerous and everyone was concerned," Newman said. "But the thing is, there were benefits." In the end, she said, the decision on whether to continue the medication varied with each patient.
Last week, assessing Avandia became even more complicated for Newman and her patients when an FDA advisory committee voted 20-3 in finding that Avandia increases the risk of heart disease in Type 2 diabetes, but at the same time the panel voted 22-1 to leave it on the market.
At a time when more information than ever is broadly available about illnesses and the effects - positive and negative - of the drugs used to treat them, doctors and patients are facing a growing challenge as they attempt to sort out the benefits and risks of medications and therapies.
Every week, professional journals and other sources report the results of new medical studies - a rising tide of information that sometimes seems contradictory. Patients can read the studies on the Internet, or read about them in newspapers, and bombard their doctors with questions that frequently don't have simple answers.
"So many patients go on the Internet now and with what the press bombards them with, they have a lot of information. They need to be able to understand the literature, so they can take it to the doctor and ask informed questions," said Erik Rifkin, a retired environmental consultant and co-author of The Illusion of Certainty, a new book that argues for increased clarity in the reporting of medical research.
Rifkin and his co-author, Edward J. Bouwer, both of whom live in Baltimore, say researchers should state findings in terms that are easier to understand. Statistics used in studies - and repeated in news reports - are often misleading, they say.
William Mohler ran into information overload last year when he turned to the Internet for help in understanding a condition that predisposed him to cancer.
The retired plant engineer had developed Barrett's esophagus with high-grade dysplasia, a condition that experts say carries roughly a 30 percent risk of developing cancer of the esophagus.
Online, Mohler found detailed descriptions of the usual treatments: inserting a laser into his esophagus to burn away the pre-cancerous tissue, a procedure that would make him sensitive to sunlight; and surgery that would have been "the equivalent of a quadruple bypass," he said.
"I got so much information that I had to quit because with the Internet, there was so much out there," said Mohler, 69, of Frederick.
His doctor referred him to a specialist at the University of Maryland Medical Center, where as part of a clinical trial he had an endoscope inserted into his esophagus so the abnormal tissue could be sprayed with ultra-cold liquid nitrogen, destroying it and allowing healthy tissue to grow in its place. "It's been a godsend," Mohler said.
Increasingly, some physicians say, patients come to them with fixed opinions based on what they have heard or read about in the latest medical studies. "They actually tend to believe more what's in the newspaper and what's on the Internet than they believe me," said Dr. Joanna Saba, an Owings Mills internist.
Consumers also may reach incorrect conclusions about studies they read or hear about. Some studies, doctors say, make medications sound attractive to people who don't need them. Others scare away patients from potentially beneficial therapies.
Nearly five years ago, women were alarmed when researchers linked estrogen replacement therapy to an increased risk of heart attack and stroke, Saba noted.
"That study really freaked women out and everyone wanted to dump their hormones," she said, "but the numbers of women actually at risk in that study were actually very small."
Last month, another research team found that the only women at significant risk of heart attack and stroke from hormone therapy were those who delayed taking it for at least 10 years after menopause. One newspaper reported the results with the headline: "Doctors Change Course Again on Estrogen Therapy."
"So a lot of women dumped their hormones and became miserable post-menopausal women and didn't have to," Saba said.
Published research, especially in widely read, peer-reviewed journals, should provide sufficient information to help make informed decisions about courses of treatment, experts say.
"What a doctor needs to know is what the benefit is to patients overall, and to patients in different risk categories," said Dr. Peter C. Wyer, an associate professor of medicine at Columbia College of Physicians and Surgeons. "The results have to be really badly reported for you to not have those numbers."
But in their book, Rifkin and Bouwer, a professor of environmental engineering at the Johns Hopkins University, say study authors often mislead the public by limiting published numbers to figures showing how a treatment has affected a study group's "relative risk."
Relative risk is a comparison of one group outcome with another - in a study, those taking a medicine versus those given a placebo, for example. The authors say researchers also should include the absolute risk, a number that compares outcomes based on how often they occur in an overall population.
Here is the difference: If a study determines that 12 out of 10,000 people given a new drug have a side effect and that six out of 10,000 people given a placebo have that side effect, there a 100 percent increase in the relative risk of the side effects from the drug. But the absolute risk is based on the difference between the two, or six more people having the side effect, out of every 10,000 people who take the drug. Those who need the drug might consider it worth that increased risk.
Although Rifkin and Bouwer aren't medical doctors, some medical experts agree it's important for researchers to include both relative and absolute risk numbers in study results.
"It's not that relative risk is wrong, but it doesn't give you the absolute number of people who benefited," said Dr. Michael Lincoff, a cardiologist who is vice chairman for research at the Cleveland Clinic.
There have been other calls for clearer reporting in medical studies In 2004, Dr. Victor M. Montori, a researcher at the Mayo Clinic and a critic of the way many studies are conducted, published a "Users' Guide to Detecting Misleading Claims in Clinical Research Reports" in the British Medical Journal.
In an interview, Montori said there is increased awareness about the need for researchers to more clearly spell out their methods and findings.
He sees some moves in the right direction: Pharmaceutical firms are beginning to post study results on their Web sites; federally funded clinical trials must now be listed on a government Web site (www.clinicaltrials.gov); and reformers are calling for researchers to use more uniform study methods so results can be more readily compared.
Montori said another problem is that researchers often have an incentive to exaggerate study results.
"The study authors want to attract editors of top journals, so they highlight the four or five outcomes that are most dramatic," said Montori, who is an endocrinologist and clinical epidemiologist.
But he said that can lead to conflicting results when there's more than one study on a medication or treatment.
A study in 2004 concluded that patients with Type 2 diabetes benefited from taking statins, a class of medications that lower cholesterol and have become one of the most widely prescribed medications in history. Then last year, another study found statins had a negligible effect on diabetes patients, Montori said.
The first study, emphasizing benefits to diabetics, was more widely reported, Montori said.
"Doctors reading journals, what's getting to them is the positive stuff, the overestimation of the benefits from treatment," he said. "Practicing clinicians do have not time to go looking for this stuff."
dennis.obrien@baltsun.com
Sun reporter Sindya Bhanoo contributed to this article.