STATS ARTICLES 2007
Harvard’s Inconvenient Data About Diet and Cancer
Harvard’s massive pooling project is challenging conventional wisdom about diet and health – and a delay in publishing the most controversial finding has politicians demanding answers.
For years Harvard’s School of Public Health has recommended keeping red meat to a minimum in one’s diet – Dr Walter Willett, Fredrick John Stare Professor of Epidemiology and Nutrition published a book in 2001, “Eat Drink and be Healthy, ” which recommended eating red meat “sparingly” and favoring grains and vegetables. The new food pyramid he constructed took a much tougher line on red meat and diet than the one favored by the Department of Agriculture.
Dr. Willett is a giant in the field of nutrition. As Harvard colleague Dr. Frank Speizer told U.S. News and World Report in 2004, Willett deserves the credit for actually delineating the relationships between diet and chronic disease. He led the campaign against trans fats and, more controversially, believes that the glycemic index can be used to determine which foods are propelling the growth in obesity
As his profile at the School of Public Health notes, he has been involved in work “a positive association between animal fat and red meat consumption and risk of colon cancer.” This association is significant because colorectal cancer accounts for some 56,000 deaths each year in the U.S. (other risk factors include obesity, smoking, a diet low in vegetables and fruit, a diet high in animal fat, heavy alcohol consumption, and low physical activity). But it is an association that has been contentious and inconclusive, depending on which study you looked at.
Then, in 2005, a study published in JAMA, Meat Consumption and Risk of Colorectal Cancer Chao et al, seemed to nail the link, at least according to media accounts: among the 148,610 men and women participating in the study, those who consumed large quantities of red and processed meat over a ten-year period were 30 percent more likely to get colon cancer and 40 percent more likely to get rectal cancer. The study was accompanied by an editorial by Willett, then chair of the School of Public Health’s Department of Nutrition, who concluded that:
“The relation between red meat consumption and colorectal cancer may not be conclusive, but prudence would suggest that red meat, and processed meats in particular, should be eaten sparingly to minimize risk. When combined with other healthful diet and lifestyle factors, it appears that approximately 70% of colon cancer can potentially be avoided. Replacing red meat with a combination of fish, nuts, poultry, and legumes will also reduce risk of coronary heart disease, in part, because some of these foods have positive benefits. This substitution is an important part of the Mediterranean dietary pattern, which improves blood lipids and other metabolic parameters and has been related to lower rates of total mortality. Thus, keeping red meat consumption low is best viewed, not as an isolated goal, but as part of an overall dietary and lifestyle strategy to optimize health and well-being. Fortunately, substituting pistachio-encrusted salmon and gingered brown basmati pilaf for roast beef with mashed potatoes and gravy is not a culinary sacrifice.”
On closer inspection, STATS found that the 30% increase in risk for colon cancer claimed by Chao et al, was not statistically significant – nor did it apply to women, where there was no correlation between meat consumption and colon cancer. The same study did find, however, a statistically significant increase in rectal cancer (71 percent) among those who had a high intake of red meat (but not processed meat)
And now it turns out that another study, the largest ever to examine the link between colon and rectal cancer and red meat consumption with 725,258 subjects, found no association between higher red meat consumption (including processed meat) and a higher colorectal cancer risk.
The study, “Meat and fat intake and colorectal cancer risk: A pooled analysis of 14 prospective studies,” by Eunyoung Cho, and Stephanie A. Smith-Warner for the Pooling Project of Prospect Studies of Diet and Cancer Investigators, was abstracted in 2004. But it has never been published. And now two senators from a state where beef is a major industry are wondering why.
On September 24, Senators Pat Roberts (R--Kansas) and Sam Brownback (R- Kansas) wrote to Mike Leavitt, Secretary of Health and Human Services:
“We are concerned about the timeliness of a National Institutes of Health (NIH), taxpayer funded research project concerning diet and cancer. This important, coordinated effort has been underway for more than seven years. As part of a NIH funded initiative, nineteen studies on dietary factors have been published since 2000. However, there is one study that was completed in 2004 and presented in an abstract but remains unpublished.”
The reason the senators are concerned is that on Nov 1, the World Cancer Research Fund (WCRF) will publish its 2007 report on “Food, Nutrition, Physical Activity and the Prevention of Cancer – an update to its 1997 report, which will evaluate some 5,000 to 10,000 new studies on cancer completed over the past 10 years. Professor Willett is a member of the WCRF’s 2007 expert panel. As Senators Roberts and Brownback note,
“This report will make numerous nutrition recommendations and stands to have a significant impact on public policy worldwide. We find it concerning that a landmark study of this size and scope and summarized for presentation in early 2004 was not published as a full manuscript in time for consideration by WCRF’s panel.”
The letter concludes by noting that taxpayers have funded this research and asking for the status of the study and when it would be published. STATS contacted the authors of the study directly with the same questions and got a more prompt reply.
“After we conducted the initial preliminary analyses examining the association between meat and colorectal cancer risk, we became aware of a few more studies that met our inclusion criteria and were eligible to join the Pooling Project,” wrote Stephanie Smith-Warner, Assistant Professor of Nutritional Epidemiology at the Department of Nutrition at Harvard’s School of Public Health. “We decided to include the studies in our analyses and are planning to submit the final results for publication within a year.”
Perhaps these extra studies will sufficiently alter the original study to show a positive association between increased red meat consumption and colorectal cancer, or perhaps not. Either way, there will always be new studies.
What’s odd is that the Cho and Smith-Warner’s study was, by far, the biggest study ever conducted on the association between red meat and cancer, and many of the other pooling studies were published with fewer subjects. The statistical results look robust, if only from the perspective of the abstract, and they conform to earlier studies that found no association or only a statistically insignificant associations. Moreover, STATS specifically asked both researchers if they had, upon publishing the abstract, found an error in their methodology that might have invalidated the results – and the question was not answered. Given that a standardized methodology was applied to pooling the data across all the areas of dietary research, a significant error in design or analysis could have implications for other findings.
To add to the appearance that awkward data is being delayed because it runs counter to long established and deeply held nutritional positions, the pooling project has just rained on another dietary parade (albeit one already damp from accumulating data).
On September 25, a study of 756,217 men and women found that eating fruits and vegetables was not associated with a reduced risk of colon cancer. The online publication of this study also appears to come too late for inclusion in the WCRF report, although, it should be noted that some earlier studies have not found positive associations with high fruit and vegetable intake and lower cancer risks.
It remains to be seen how what the WCRF recommends when it comes to red meat and the risk of cancer. One can only hope, given the way the press has played up the risks from red meat in the past, that the WCRF considers the possibility that Cho and Smith-Warner’s conclusions might not be changed by adding more studies. Flip flops on dietary recommendations have a way of undermining the public’s assurance that science can reliably inform health decisions.
View the Technorati Link Cosmos for this entry