Monday, December 10, 2012

Researchers ask "Is everything we eat linked with cancer?"

The answer's "No," but here's why it can seem that way


In the early 80's, the singer/songwriter Joe Jackson captured in the refrain of his song "Cancer" a frustrated sentiment many people were feeling back then - and not surprisingly still do now - that: "Everything gives you cancer."

Even in the 80's - at the cusp of the information revolution - the amount of health news filling newspaper pages and airwaves was overwhelming.  On top of this, stories could oftentimes seem conflicting, and at times, just plain arbitrary.  It's a trend that's only gotten worse as we make our way toward a world of universal information access, little context, and not nearly enough studied curators.

What got us here at CNiC thinking about these things was the recent publication of an intriguing and entertaining paper seemingly inspired by Mr. Jackson's song, called Is everything we eat associated with cancer? A systematic cookbook review.

The authors, Jonathan Schoenfeld of Harvard Medical School and John Ioannidis of the Stanford Prevention Research Center, picked 50 food ingredients from random recipes in a basic cookbook and reviewed the scientific literature to see which of the items had been studied and what the prevailing results of each were.  Food items included in the study were things like, veal, salt, carrots, mushrooms, milk, cheese, cinnamon, tea, and raisins.  Most were items that could easily be found in kitchen cupboards and refrigerators.

What the study found was that of the individual studies of these foods, over 70 percent found that the food either increased or decreased the risk of cancer.  Only 23 percent of studies showed no effects.  This means that the vast majority of studies looking at common foods found some link with cancer.

On the positive side, Schoenfeld and Ioannidis also looked at the results of meta-analyses of the same items.  Meta-analyses collect and analyze the results from many different studies, and they provided a more measured view on the food list's link with cancer.  Whereas 72 percent of individual studies found links with cancer, only 36 percent of meta-analyses did.  The larger size of a meta-analysis and its usual attention to design takes into account much of the noise or marginal results from the individual studies, providing something that is likely more solid.

Of course, it can be tall order for most people to delineate the niceties of such results, which places a greater burden on reporters and scientists to spend extra time to place results into context, to not overreach with conclusions, to spend time thinking about how the messages will be received (for more on this, see a recent editorial of ours).

Until then, though, there are a few simple tips readers can use to help bring some clarity to the cluttered world of diet and health news.  When trying to figure out what the latest health findings in the news mean for you, ask these simple questions:
Was the story about a study done in humans or animals?  Animal/lab studies have less importance than human studies when it comes to health choices. 
Was the story about the results of a single study?  The results of single studies are usually not important enough to base health choices on.  
Was the story about new recommendations or guidelines based on many studies? Recommendations or guidelines developed by the federal government or other reputable organizations are usually best for guiding health choices.  They are made by experts in the field using the entire weight of scientific evidence on a topic.

For more tips on understanding health news and information, visit these sites:
How to Evaluate Health Information on the InternetOffice Dietary Supplements, National Institutes of Health 
MedlinePlus Guide to Healthy Web SurfingNational Library of Medicine, National Institutes of Health

No comments:

Post a Comment