This week a scary report was issued linking colon cancer to the consumption of red meat. According to Steven Milloy , however, the study this report was based on wouldn't even win first prize in a junior high school science fair:
Researchers from the National Cancer Institute and the American Cancer Society concluded in this week's Journal of the American Medical Association (Jan. 12) that "prolonged high consumption of red and processed meat may increase the risk of cancer in the distal portion of the large intestine."
Their "conclusion" is based on a study of 148,610 adults aged 50 to 74 years who provided information on meat consumption in 1982 and again in 1992/1993 as part of their enrollment in the Cancer Prevention Study II . Through Aug. 31, 2001, a total of 1,667 cases of colorectal cancers were reported among the study subjects.
I suspect that the researchers actually had no conclusion worth reporting after they did an initial analysis of their data. They reported, in fact, no association between red meat consumption and overall colon cancer risk after considering the study subjects' exposures to other colon cancer risk factors.
They parsed the results as follows, "High intake of red meat reported in 1992/1993 was associated with higher risk of colon cancer after adjusting for age and energy intake but not after further adjustment for body mass index, cigarette smoking and other [risk factors]," they stated in the study. Facing the prospect of no result, I think the researchers then engaged in some slicing-and-dicing of their data in hopes of discovering some statistical correlation they could point to as a "risk."
Since there was no correlation between red meat consumption and overall cancer risk, the researchers examined their data looking to see whether there was an association between red meat consumption and cancer of the proximal colon, distal colon and of the rectosigmoid and rectum. The more analyses performed, after all, the greater the likelihood that some newsworthy result will be found, albeit, due to chance perhaps.
These subsequent analyses produced three correlations on which the scary headlines are based: a 50 percent increase in distal cancer risk among high consumers of processed meats; a 53 percent increase in distal cancer risk among those with the highest ratio of meat to chicken and fish consumption; and a 71 percent increase in rectal cancer risk among high consumers of red meat.
With respect to the claims concerning distal cancer risk, both results are of unimpressive size - risks smaller than 100 percent don't have much credibility. They are of borderline statistical significance - meaning that there is a good possibility that the results are due to chance. This statistical weakness is in large part due to the fact that they are based on analyses involving only 79 and 92 cases of distal cancer, respectively. For these analyses to start to be taken seriously, they should involve hundreds, not dozens, of cancer cases.
As to the reported 71 percent increase in rectal cancer risk among red meat eaters, I can only conclude that this result was cherry-picked for sensationalistic purposes. The 71-percent claim is based on an analysis involving only the 1992/1993 data. When the analysis includes the 1982 data, the result drops to 43 percent and becomes statistically insignificant.
These weak statistics are just the surface of the problem. Likely nullifying the entire study is the unreliability of the researchers' data.
The data for the study was initially collected by 77,000 untrained volunteers who interviewed family and friends about their lifestyle habits. None of this lifestyle data was verified or validated.
Exactly what and how much the study subjects ate, smoked, drank, and how much they exercised is really anybody's guess. And forget about reliable information on genetic predisposition to colon cancer, which is thought to be a major risk factor. The researchers acknowledged in their write-up that they didn't even have any information on family history of colon cancer for the analysis of 1992/1993 data.
The results of previous studies on meat consumption and colon cancer have produced similar inconsistent, contradictory, weak and even nonsensical results. There really is no persuasive evidence that meat consumption is in any way related to colon cancer risk.
So what's up with this study?
Aside from the usual hijinks of researchers looking for media attention and their next grant, I noticed that one of this study's authors has somewhat of a track record trying to link meat consumption with cancer.
The National Cancer Institute's Rashmi Sinha has a long history of trying to use weak statistics to convict meat of causing cancer. I first brought her antics to the attention of my FoxNews.com readers in a November 2000 column amid her crusade to link well-done meat with cancer.
It appears that Dr. Sinha remains bent on using her position at the National Cancer Institute to scare us away from eating meat. She's been at it since at least 1994, but with little to show except a stack of scary, but unsupported headlines - which in itself is somewhat revealing.
If after all the time and effort Sinha has put into trying to link meat consumption with cancer, she still can't do it, isn't it time that the NCI reassign her to more productive work?
Indeed. And please make ours medium-rare.
