Reliable Information on the Internet

Written by Maryann Napoli at the request of the New Zealand Guidelines Group in anticipation of her appearance at the National Consumer Summit in Auckland, October 31, 2005.

  Much as I’d like to respond to this assigned topic with a list of sure-fire ways to identify trustworthy Web sites, I’m sorry to say that I cannot. From this USA-based consumer advocate’s point of view, there isn’t much I can wholeheartedly recommend out there. I wouldn’t think of relying on the standard sources like the American Heart Association and the National Cholesterol Education Program given their considerable pharmaceutical industry funding.

  And the fact that a board of academic physicians oversees a Web site’s content is no guarantee of high-quality information. Case in point is the U.S. National Cancer Institute’s Physicians Data Query (PDQ) database www.cancer.gov  I still see it as the best source of evidence-based cancer treatment information, despite the fact that I have found numerous gross errors in this Web site over the years. Lobular carcinoma in situ, for example, was described as a breast cancer (leading women to think immediate aggressive treatment is critical); whereas LCIS is merely a marker for cancer. In another instance, prostatectomy was described (in 1989) as the most common first choice treatment for prostate cancer—a misleading statement given the fact that there is no evidence that proves it superior to no treatment at all. Worse, the no-treatment option appeared in the “health professionals” section of the database but not in the material directed at “patients.” My critiques of the PDQ have always been taken seriously and often led to changes. “Watchful waiting”, for example, now heads the list of treatment options for men with early-stage prostate cancer…in both the “patient” and the “health professional” sections of the database.

  The American Cancer Society, on the other hand, was unresponsive when I asked why it instructs doctors to tell their patients about the risks and benefits only where it concerns the PSA screening test for prostate cancer. After all, that’s not the only cancer screening test with risks as well as benefits.

 Though the explosion of information available on the Internet is gratifying in many ways, it is still hard to find certain information crucial to informed decision-making. Just take a decision many of us will have to make: Should I take a prescription drug for the rest of my life? Millions of healthy adults around the world have accepted the idea that risk factors like bone loss and high cholesterol should be treated with lifelong drug therapy. 

 The woman with osteoporosis told to take the bone drug Fosamax will find lots of information about how the drug improves bone density. But how good is it at reducing the rate of hip fracture, the most serious consequence of osteoporosis?  It’s not easy to ferret out the three-year trial which found 2% of women on the placebo experienced a hip fracture, compared to 1% of women on Fosamax. These results are enlightening, considering how heavily promoted this drug is. Only 1% of the women on Fosamax actually benefited from the drug; and only 2% had a hip fracture without the drug. You can find this information at the U.S. Food and Drug Administration’s Web site (www.fda.gov). but it takes some doing. First, you must find your way to the right section of the site and then slog through reams of fine print of the drug labeling information that the pharmaceutical company publishes with FDA oversight.

  Even then, you wouldn’t know that this three-year trial provides the best long-term effectiveness information on Fosamax—and it applies only to elderly women with osteoporosis and at least one fracture. That doesn’t represent the primary Fosamax users in the USA, thanks to misleading drug advertising aimed at physicians. When Fosamax first came on the market ten years ago, its maker Merck mounted an advertising campaign featuring white women in early middle-age. No mention of prior fractures, but plenty of encouragement for bone-density testing (screening creates customers). As an antidote to this market-driven poison, osteoporosis researcher Susan M. Ott, MD, University of Washington, maintains an advertising-free Web site that helps women and doctors determine when drug therapy is and is not appropriate (http://courses.washington.edu/bonephys).

  Anyone about to go on an open-ended drug regimen would want to weigh the benefit against the odds of having a serious adverse drug reaction. Three years ago, when writing about the popular cholesterol-lowering drugs called “statins” (Lipitor, Zocor, Pravachol, etc.), I was pursuing that question on behalf of my readers with high cholesterol but no heart disease.  Fortunately, I found my way to a trustworthy source of drug information, Therapeutics Initiative in British Columbia, Canada (www.ti.ubc.ca Letter 48). It was the only place I found the odds of benefit compared to the odds of having a serious adverse drugs reaction. Five major trials have compared statins with placebos in people without heart disease. Only two trials, however, have released their serious adverse events (SAE) data! Working with what they had, the Canadian researchers combined the results of the two trials and found only one modest benefit to statins. These much-touted drugs reduced the odds of a non-fatal heart attack by 1.8% (in men only). But this modest benefit was canceled by a 1.4% increase in the rate of SAEs among the statin users! And if someone wants to check whether a particular statin—or any other drug—is so risky it should be taken off the market, there’s always the Washington, DC-based Health Research Group (www.citizen.org/hrg).

  Consumer advocates willing to take the time can find some gems of hypocrisy at the most mainstream of Web sites. Every November the federal government in the form of the U.S. Centers for Disease Control and Prevention scares us (via the media) with influenza death statistics. This has the desired effect of making people think they’re going to die if they don’t get a flu shot. (The death stats, by the way, are deeply suspect.) One consumer advocacy organization, the National Vaccine Information Center (www.909shot.com), led the media to transcripts of the FDA’s Vaccines Advisory Committee meetings freely available at the FDA’s Web site (though, once again, difficult to find). There for all to read were the flu experts saying in early 2003 that they had screwed up. One key viral stain had been omitted from the flu shot that was promoted vigorously as the 2003-04 flu season approached. Federal health officials just couldn’t bring themselves to be honest with the American public until the time to get a flu shot had passed. After all, we might not line up for our shots, and then the vaccine companies would have to discard their unsold products. Go to flu vaccine reviews at the Web site of the Cochrane Collaboration (www.cochrane.org), and you’ll find that, worldwide, flu vaccines haven’t worked very well in the last 35 years.

  I know people don’t want to hear this, but you practically have to become an investigative reporter to find what you want…or don’t yet know you want. One has to sift through a considerable amount of information on the Internet to learn how to determine what’s valid and what isn’t. My parting advice: No matter what the topic, start with a Google search (www.google.com). It produces a wealth of information—both good and bad—and often free access to entire articles.

Maryann Napoli, Center for Medical Consumers ©, November 2005