A new science study on the links between processed meat and chronic diseases indicates that studies that show a relationship between the two are of very low quality and suffer, as the authors say, “a serious risk of bias and imprecision.
This statement is not surprising, as it follows a series of recent analyses that have shaken the world of nutrition. Those earlier studies, published in the Annals of Internal Medicine, concluded that guidelines warning against eating less red and processed meat are based on evidence with very little certainty. The researchers who conducted those analyses said there is no way to determine, for a given individual, what the risks or benefits of eating meat might be.
In that sense, this new study is similar not only in its contradiction to the decades-old dietary guideline, but in its accusation of the science behind the nutritional guideline in general.
Unlike the Annals of Internal Medicine studies, the new study was not a systematic review or meta-analysis of available studies, but a review of such reviews. The eight authors, all from the University of Copenhagen, used two methods to assess the quality of all studies considered.
The first method, called AMSTAR, is a widely used tool specifically designed to determine the quality of a systematic review or meta-analysis. Using this tool, the researchers in this new study found that previous systematic reviews or meta-analyses linking processed meat to chronic disease were of poor quality. They also found that the better the study design, the less likely an association was found.
One of AMSTAR’s most important criteria is whether the researchers provided a public version of their research plan before data collection began. Such a plan limits researchers’ opportunities to “massage” protocols or models so that they can change the results obtained. Only one of the 22 reviews of this study did. For the other 21 reviews, the researchers could not tell if the original authors followed where the data went or made sure the data went where they wanted it to go. The fact that so few of these studies used any kind of predetermined research plan points to a critical flaw in the way nutrition research is currently conducted.
The other method used by researchers in the study of processed meat is the GRADE system. GRADE’s transparent and highly recognized framework is the most widely adopted tool for assessing the quality of evidence used to make recommendations. It was also the method used by the Annals of Internal Medicine studies that caused so much fuss. The GRADE system stands out for the fact that the results of observational studies (case-control and cohort studies) are considered by default to be of low quality, a perspective with which Diet Doctor agrees. The quality of observational studies can be improved if the studies are of the highest quality-no problems of confusion, with large effect sizes, and a consistent dose-response relationship-but this is rare in nutritional research.
In the analysis of processed meat, the researchers who evaluated the reviews on this topic indicated that none of the studies involved met these standards. In fact, as the researchers explained, “the certainty of the effect estimate was reduced to very low, due to a serious risk of bias and/or serious imprecision.
As with the studies in the Annals of Internal Medicine, this study does not say that processed meat is “good for you” or that you should eat more of it. It simply says that guidance to reduce consumption of processed meat is based on very low quality studies, and these conclusions are very uncertain.
At Diet Doctor, you can count on us to follow the science where it takes us – like our recent re-evaluation of soy – even if some of our readers find the conclusions disturbing. After all, it is unlikely that advice that is not based on solid scientific evidence will produce positive health outcomes, as we have discovered after 40 years of low-fat, high-carbohydrate dietary guidance. At Diet Doctor, we believe the public deserves better.