Natural Foods Merchandiser

JAMA study considers conflicting research results

When new research seems to contradict older findings regarding the same nutrient, what information do you pass on to customers? A study published in the December 2007 Journal of the American Medical Association implies that observational studies may be misleading to the public, especially when their findings are still widely accepted despite being contradicted by randomized clinical trials. But according to the Council for Responsible Nutrition in Washington, D.C., this black-and-white conclusion may not be entirely accurate.

"There's an inherent bias on the part of the authors that randomized clinical trials automatically trump observational studies," said Andrew Shao, Ph.D., vice president of scientific and regulatory affairs at CRN. "But that can't be applied across the board because it depends on whether the studies were attempting to answer the same question or not."

The researchers featured in JAMA first looked at two popular epidemiological studies on vitamin E from 1993 that associated the vitamin with reduced risk of cardiovascular disease. By 2006, each of the studies had received more than 1,000 citations, and according to the report, "based on these articles, vitamin E was considered cardioprotective for many years."

The researchers then pointed out that the most-cited clinical trial on vitamin E, the Heart Outcomes Prevention Evaluation—or HOPE—trial in 2000, contradicted those findings, and a 2004 meta-analysis found that high doses of vitamin E may increase the risk of death.

The researchers also looked at articles on beta-carotene for cancer prevention and estrogen for dementia prevention, finding similar conflicting results in all.

But according to Shao, differing conclusions do not necessarily mean that one of the studies is irrelevant. The critical consideration, Shao said, is to look at the question the study is asking. To retailers, he suggested looking at what the researchers set out to find, and trying to bring that clarity to customers.

Dr. John Ioannidis, lead researcher of the JAMA study, agreed that the focus of a study can confuse the conclusion.

"The observational studies have given the wrong answer or, to be more exact, the answer to the wrong question," Ioannidis said. "What the observational studies asked is, ?Do people who take vitamin E happen to also have better health?'? If the answer is yes, then better health may have nothing to do with vitamin E, but perhaps the study subjects who took vitamin E were also wealthier and more health-conscious, or some other aspect of what they did influenced their health.?The randomized clinical trials ask the direct question, ?Will I improve my health if I take vitamin E?' And the answer we found was ?no.' "

But Shao stressed that scientists, depending on their association, will see the evidence differently. A nutritional scientist, for example, will see the role of vitamin E as an antioxidant, recognizing that nutrients interact with each other, and consider other data in addition to randomized trials. The clinical scientist, on the other hand, is more apt to look solely at the trial findings and treat the nutrient and the results as one would a drug trial.

This divergence in the scientific community puts some of the burden of educating consumers on retailers and manufacturers. So how to address consumers when it comes to conflicting results?

"I would put more weight on the results of randomized trials," Ioannidis said, "especially if they are large and well-conducted and without apparent conflicts of interest."

Shao suggested not getting caught up in the latest headlines or hype. He said because no filtering mechanism in the scientific publishing community exists, the latest studies distributed quickly to the public can draw attention to important points, but might not be taking in the whole picture.

"We all need to avoid making sweeping conclusions based on the latest study," Shao said. "And associates at retail stores need to understand and communicate to customers that for most nutrients and most products there is a body of evidence that exists, and it points us in a direction in terms of the effects of a nutrient, so one new study doesn't necessarily trump all past studies."

Chris O'Brien is a Boulder, Colo.-based freelance writer.

Natural Foods Merchandiser volume XXIX/number 1/p. 15

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.