点击显示 收起
Feb. 11, 2008 -- Breast cancer information on web sites is mostly reliable, with just one in 20 sites presenting inaccurate information, according to a new study. But researchers found complementary and alternative medicine sites more than 15 times as likely to contain false or misleading information.
"If I had to guess, I would think breast cancer information would be more accurate than some other [health conditions]," says Elmer V. Bernstam, MD, author of the study and an associate professor of health information sciences and internal medicine at the University of Texas Health Science Center, Houston. Bernstam has studied the accuracy of online health information for about eight years.
The study was conducted in collaboration with researchers from the M.D. Anderson Cancer Center in Houston and published online today and in the March 15 issue of the journal Cancer.
Bernstam declines to name names of web sites, even those with inaccuracies. "The web changes constantly," he says. "The fact [that web sites] are accurate today doesn't mean they are accurate tomorrow."
Bernstam and his colleagues used five popular search engines such as Google and Yahoo to identify 343 unique web pages using 15 breast cancer-related queries (such as breast cancer, breast cancer risk, mammogram, and mastectomy) from June 1 to July 30, 2004.
"We picked terms that people were likely to use," he says.
Next, two reviewers assessed the information. One reviewer applied 15 traditional quality criteria used to evaluate web pages, such as display of the author, date the page was created, and when the information was last modified.
The researchers found 41 false or misleading statements on 18 of the 343 sites -- about 5%. "That's not too bad," Bernstam says. "That's the good news."
When the researchers zeroed in on complementary and alternative medicine sites, they found they were more than 15 times as likely to have false or misleading information.
"We don't have the data to tell you why," says Bernstam. And, he adds, the misleading information, while found on a site discussing complementary or alternative medicine, "may or may not have been about the complementary or alternative medicine." For instance, there could have been an inaccuracy in a definition but not in the effectiveness of an alternative medicine treatment.
How bad were some of the inaccuracies? Bernstam recalls one site that posted this erroneous information: "The risk of developing breast cancer by wearing a bra more than 12 hours a day is 21 times higher than for women who remove their bra before that 12 hours."
From numerous quality criteria, Bernstam's team picked 15 criteria to evaluate the pages. Among them: is authorship identified? Does the user know when the information was last updated or modified?? Are the doctors' credentials identified? Are author affiliations disclosed?