A series of articles were published in the November 2019 issue of the Annals of Internal Medicine that came to a very controversial conclusion. A team of researchers led by Dr. Bradley Johnston, an epidemiologist at Dalhousie University in Canada, re-examined much of the available research on the impact of red meat consumption and the risk of cardiovascular disease (CVD) and cancer. Dr. Johnston and his team concluded that the available evidence does not support the long-standing recommendations from the American Heart Association, American Cancer Society, and the most recent 2015-2020 US Dietary Guidelines. These and many other dietary guidelines have been recommending that Americans and people in other countries ought to limit their intake of red meats and especially processed meats in order to reduce the risk of cancer and CVD.
The blowback against Dr. Johnston’s articles began even before they were published. It is very unusual to see such criticism of research at this point. A January 15, 2020 editorial article by Rita Rubin describes in detail this almost unprecedented face-off between some of the leading public health researchers and Dr. Johnston and Dr. Christine Laine, the editor-in-chief of the Annals of Internal Medicine. Dr. Laine was caught in the middle and decided to publish Dr. Johnston’s articles in the November 2019 issue of the Annals of Internal Medicine. I would suggest everyone read Rita Rubin’s article titled “Backlash over meat dietary recommendation raises questions about corporate ties to nutrition scientists” here before reading the rest of this reviewer’s take on Dr. Johnston’s controversial research and his recommendations that include no more epidemiological research be done: https://jamanetwork.com/journals/jama/fullarticle/2759201.
My Take on Dr. Johnston’s Research
One of Dr. Johnston’s articles was a meta-analysis that reviewed 61 earlier articles of 55 different cohort studies with more than 4 million participants. Unlike the authors of those studies, who largely found a significant correlation between meat intake and CVD, Dr. Johnston’s meta-analysis concluded there is only a low certainty of evidence that a reduction in unprocessed red meat intake to no more 3 servings per week was associated with only a very small reduction in risk for CVD mortality, stroke, myocardial infarction, and type 2 diabetes. Likewise, Dr. Johnston found a low certainty evidence that a reduction in processed meat intake to no more 3 servings per week was associated with only a very small decrease in risk for all-cause mortality, CVD mortality, stroke, MI, and type 2 diabetes (1).
Another systematic review by Dr. Bradley C. Johnston’s team concluded “The possible absolute effects of red and processed meat consumption on cancer mortality and incidence are very small, and the certainty of evidence is low to very low.” (2).
It is not surprising that Dr. Johnston’s conclusions have been met with quite a bit skepticism, especially from many (but not all) public health researchers. One of the more outraged researchers was Dr. David Katz, the founding director (1998-2019) of Yale University’s Griffin Prevention Research Center, Past-President of the American College of Lifestyle Medicine, and the Founder/President of the True Health Initiative (THI). Indeed, Dr. Katz took the unusual step of writing to the editor of the Annals of Internal Medicine demanding that they re-consider publishing Dr. Johnston’s series of studies in November. He even published a column in his October 6, 2019 edition of the THI newsletter that equated Dr. Johnston’s soon to be published articles as “information terrorism”. THI bills itself as “Enlisting leading experts and commissioning a collection of research to fight fake facts and fix a broken system” (3).
Nor was this group of recent articles the first time Dr. Johnston has published an article that has shaken the nutrition establishment. An earlier article by Dr. Johnston’s group on sugar intake concluded: “Guidelines on dietary sugar do not meet criteria for trustworthy recommendations and are based on low-quality evidence. Public health officials (when promulgating these recommendations) and their public audience (when considering dietary behavior) should be aware of these limitations” (4). Dr. Johnston is not the only researcher pushing back against dietary advice that seems to be ever-changing to the American public. The latest US Dietary Guidelines (2015-2020) dropped dietary cholesterol as a nutrient of concern. And we have seen several articles questioning the benefits and even suggesting there are serious health risks associated with limiting dietary salt or sodium intake to low levels. And we have seen the recommendation for dietary fats and carbohydrates be in what seems to be constant flux with various researchers suggesting different goals for decades. Why all this controversy and who should we believe?
Nutrition Research is Complicated
The truth is that the impact of diet on CVD, cancer, obesity, type 2 diabetes, and numerous other ills is complicated. For one thing, these diseases take decades before leading to clinically-significant morbidity and mortality. It simply is not feasible, with human subjects, to control what people eat over many decades. And any nutritionist or RDN can tell you that determining what people eat with any precision is more art than science. Most people cannot tell you with any precision what they ate yesterday, let alone what they have been eating over their prior lifetime. And people who are informed that they are at risk of a disease may alter the way they eat. If your father has a heart attack and your doctor tells you your cholesterol level is above average, might you alter your diet? This tendency for those at higher risk for a disease to alter their diet confounds attempts to correlate diet with disease. These changes create what statisticians call reverse causation. Obese people may have cut back on sugar-rich sodas and switched to diet sodas years ago. This can make it appear that diet soda intake correlates with obesity even though it was often the genetic propensity to gain weight that led people to change their dietary habits. While research showing that diet sodas are as bad or even more associated with obesity than regular sodas has been published, it seems likely such data are misleading due to reverse causation.
People with elevated blood pressure (BP) or serum cholesterol levels may be more likely than the overall population to cut back on salt or saturated fat and cholesterol-rich foods, respectively. This statistical confounding is nearly impossible to correct for in large epidemiological studies of homogeneous human populations, where most people’s diets are not all that different. And the well-known errors in measuring dietary intake accurately compound the ability to see real causal relationships between diet and disease. Given all this statistical “noise,” it is not surprising that such research studies can find far weaker association than likely exists.
Imagine if nearly everyone smoked between one and two packs of cigarettes, starting as teenagers. Would we see the dramatic differences in lung cancer and emphysema that we see between life-long smokers and never smokers? Obviously not. Smoking twice as much does not lead to twice as much lung cancer, and only about 20% of smokers die from lung cancer, so clearly smoking is not the only factor promoting lung cancer. However, because we have smokers and non-smokers who otherwise live similar lifestyles, it was possible for epidemiologists to find that smoking is in fact by far the greatest causal environmental factor promoting lung cancer and emphysema. But with diet, we rarely see people either eat frequently or never eat red meat over a lifetime. This makes the statistical “noisy” data on red meat intake and risk of disease far more difficult to evaluate within a more homogeneous population where almost everyone eats red meat.
One can compare different populations around the world where we know people have been eating very different diets over their entire lives. However, such studies also suffer from other confounding variables. For example, isolated populations that add little or no salt to their food see far less increase in their BP with age as do Americans. Of course, in these isolated human populations, one cannot look at large difference in the consumption of say salt, sugar, or red meats and see impacts on longevity we could compare with the American population where most eat red meat several times a week. Why? Because Americans live with far better pubic health measures and have far better access to quality medical care than people in less advanced modern populations. Therefore, these impact health and longevity more than elevated BP for excess added salt. Given all these research challenges, it should be no surprise the correlations between any dietary component and disease are difficult to determine with any precision.
Bottom Line: While Dr. Johnston’s research does show that the correlation between diet and diseases such as CVD, type 2 DM, and cancer seem rather weak, there remains more than enough research from controlled clinical trials that show unequivocal changes in metabolic risk factors known to promote these serious illnesses. There is little doubt that increasing intake of red and processed meats, salt, and refined sugars promotes cancer, CVD, type 2 DM, and other serious ills that significantly increase the diseases that cause a large percentage of the morbidity and mortality of Americans. Next month, we will take a look at a recent study conducted in the USA that examined the association between red meat intake and the risk of cancer and total mortality using a unique group of Americans who eat very little or no red meat at all over their entire lives.
By James J. Kenney, PhD, FACN
- Red and Processed Meat Consumption and Risk for All-Cause Mortality and Cardiometabolic Outcomes: A Systematic Review and Meta-analysis of Cohort Studies. Dena Zeraatkar, Mi Ah Han, Gordon H. Guyatt, et. al. Ann Intern Med. 2019;171:703-710. doi:10.7326/M19-065.
- Mi Ah Han, Dena Zeraatkar, Gordon H. Guyatt, et.al. Reduction of Red and Processed Meat Intake and Cancer Mortality and Incidence: A Systematic Review and Meta-analysis of Cohort Studies. Ann Intern Med. 2019;171:711-720. doi:10.7326/M19-0699.
- Erickson J, Sadeghirad B, Lytvyn L, Slavin J, & B. C. Johnston. The Scientific Basis of Guideline Recommendations on Sugar Intake: A Systematic Review. Ann Intern Med. 2017;166:257-267. doi:10.7326/M16-2020.
Stephanie Ronco has been editing for Food and Health Communications since 2011. She graduated from Colorado College magna cum laude with distinction in Comparative Literature. She was elected a member of Phi Beta Kappa in 2008.