Red meat, processed meat, and cancer – how strong is the evidence?
Grant Schofield, George Henderson
On the 26th October the IARC Monograph Working Group, who work under the World Health Organisation, declared that processed meat is a carcinogen and that red meat is a possible carcinogen.
Of course the media ran away with this, comparing bacon with smoking, to the despair of anti-smoking campaigners, but what sort of evidence is the claim based on?
“A majority of the Working Group concluded that there is sufficient evidence in human beings for the carcinogenicity of the consumption of processed meat. Chance, bias, and confounding could not be ruled out with the same degree of confidence for the data on red meat consumption, since no clear association was seen in several of the high quality studies and residual confounding from other diet and lifestyle risk is difficult to exclude. The Working Group concluded that there is limited evidence in human beings for the carcinogenicity of the consumption of red meat.
There is inadequate evidence in experimental animals for the carcinogenicity of consumption of red meat and of processed meat. In rats treated with colon cancer initiators and promoted with low calcium diets containing either red meat or processed meat, an increase in the occurrence of colonic preneoplastic lesions was reported in three and four studies, respectively.”
In other words, there is no experimental evidence that meat or processed meat is carcinogenic in animals (unless you feed the animals carcinogens, which kind of defeats the point, plus restrict calcium, and add extra heme iron to the diet). In some experiments feeding bacon to rats treated with carcinogens actually decreased the incidence of colon cancer. As the usual process for screening compounds for cancer involves feeding them to animals to see what happens, this leaves a gap in the evidence.
This means that the main evidence for meat causing cancer is epidemiological. That is, people who eat the most processed meat (by their own choice – not as part of a study that tells them to or not) in large, long-term diet studies tend to have a higher incidence of some cancers than people who eat the least.
The European EPIC study from 2013, which combined data from several smaller studies (collected in multiple ways, some more reliable than others) involving 448,568 men and women, is a good example of this type of evidence and gave rise to similar headlines when it was published. Evidence from this study population is undoubtedly included in the IARC report. In EPIC the highest intake of processed meat was associated with an increase in mortality.
We can look at this study in detail to see how conclusions are reached.
“As of June 2009, 26,344 deaths were observed. After multivariate adjustment, a high consumption of red meat was related to higher all-cause mortality (hazard ratio (HR) = 1.14, 95% confidence interval (CI) 1.01 to 1.28, 160+ versus 10 to 19.9 g/day), and the association was stronger for processed meat (HR = 1.44, 95% CI 1.24 to 1.66, 160+ versus 10 to 19.9 g/day). After correction for measurement error, higher all-cause mortality remained significant only for processed meat (HR = 1.18, 95% CI 1.11 to 1.25, per 50 g/d). We estimated that 3.3% (95% CI 1.5% to 5.0%) of deaths could be prevented if all participants had a processed meat consumption of less than 20 g/day. Significant associations with processed meat intake were observed for cardiovascular diseases, cancer, and ‘other causes of death’. The consumption of poultry was not related to all-cause mortality.” 
The paper contains a breakdown of mortality by cause. High intake of processed meat (160+ g/day) is associated with increased cancer and cardiovascular mortality. It looks to us as if only the cardiovascular mortality is statistically significant with odds of 1.72 (1.29, 2.30). However, looking at these tables for more insights it struck us that the number of deaths in the high intake of red meat and processed meat categories is very low. There are only 79 cancer deaths in the highest processed meat category compared to 2223 in the lowest (figure 5). This is because the categories (there are six of them) are based on arbitrary cut-offs, instead of dividing the people in the study into groups of the same size, which is the usual practice. Colorectal cancer accounts for about 15% of cancer mortality in middle aged New Zealanders, so even if processed meat increases the risk we are probably looking at a very small number of cases in this huge study among people with the highest processed meat intake.
How many of the 448,568 men and women in the EPIC study were eating 160+ grams of processed meat a day? The paper doesn’t give the numbers for each category, but we were able to add together the numbers of current smokers, former smokers, and never smokers in the baseline data to get the total (figure 1). 2158 men and 622 women = 2780. For red meat the figures are a bit higher, but they are still less than 1% of the study population, and while there are still associations in lower categories for processed meat consumption the risk for red meat really only relates to this upper 1%. To be in this category you would need to eat a large serving of red meat every day of the week, and never have a chicken or fish dinner.
Let’s take a step back now and imagine a study on the associations between sugar and cancer, or rice and cancer, or anything and cancer, that isolated the highest 1% of the population in terms of consumption. I’m pretty sure we’d see some striking correlations. However, I’ve yet to see a method this selective applied to anything but meat. I’m pretty sure that if we took this approach to soft drink consumption we’d be laughed at and told that the results are irrelevant to 99% of the population. As they would be.
Should I eat LCHF?
We’ve had a few queries from LCHF eaters over the last day as this meat and cancer story broke. Even with this scare evident – just say we are not promoting a high meat diet anyway.
The LCHF diet is not a high red meat diet, because protein intake should be moderate (LCHF is a high fat diet) and there are so many other protein foods to eat (eggs, chicken, fish, dairy, nuts to name the most common).
We don’t recommend consuming processed meats and think you should stay away from processed foods in general, but we also know that some preserved meats are more processed than others, and that there ancient ways of preserving meats that are less likely to be harmful than grinding meat finely, soaking it in a chemical bath, and adding processed grain and soy fillers. The processed meat epidemiology doesn’t try to distinguish between prosciutto and luncheon sausage, but good luck getting a slice of prosciutto for the price of a slice of luncheon meat. It doesn’t distinguish between biltong and saveloys. It doesn’t explain why some minced venison in a sausage skin should be different from the same mince out of the skin. In short, it’s a very blunt instrument.
If we look at colon cancer epidemiology we find high risk ratios for hyperinsulinaemia, and a protective association with high HDL cholesterol, and high serum levels of odd-chain fatty acids (found in dairy fat), DHA and EPA (found in fish and lamb), and PLP (the active form of vitamin B6, which is found in good amounts in all whole protein foods but in low amounts in processed meats).[4, 5, 6, 7]
These are all very strong correlations and serum biomarkers are measured with much more accuracy than dietary intakes of foods. The LCHF diet is intended to improve these biomarkers, and if it is permissible to speculate from this biomarker epidemiology, this improvement will significantly reduce colorectal cancer risk. This is of course equally as speculative as saying that a reduction in meat intake by any individual will reduce their cancer risk, except that the speculation rests on more reliable data and the predicted benefit is greater.
There is an interesting finding from one of the studies included in the EPIC data, EPIC-Oxford. It was a study of “health conscious” individuals, and in this population, which had a low mortality rate overall, vegetarians had a significantly higher rate of colorectal cancer than meat eaters.
“The incidence rate ratio for colorectal cancer in vegetarians compared with meat eaters was 1.39 (95% CI: 1.01, 1.91).“ 
We’d like to think that the readers of this blog are health conscious individuals.
Also – short note from a co-colaborator Dr Simon Thornley, a specialist public health epidemiologist
That population attributable risk (PAR) calculation of 3% is telling, along with the relatively modest effect sizes… 20 to 40% increase, comparing ~8 times increase in intake. The PAR means that on the assumption that this exposure is causal, removing the exposure to lowest levels will only reduce the incidence of disease by a negligible 3%….
 Carcinogenicity of consumption of red and processed meat. Bouvard M, Loomis D, Guyton KZ, et al. IARC Monograph Working Group. The Lancet Oncology October 26, 2015 http://dx.doi.org/10.1016/S1470-2045(15)00444-1
 Effect of meat (beef, chicken, and bacon) on rat colon carcinogenesis. Parnaud G, Peiffer G, Taché S, Corpet DE. Nutr Cancer. 1998;32(3):165-73. [link]
 Meat consumption and mortality – results from the European Prospective Investigation into Cancer and Nutrition. Rohrmann S, Overvad K, Bueno-de-Mesquita HB et al. BMC Medicine 2013, 11:63 doi:10.1186/1741-7015-11-63. [link]
 Metabolic syndrome, hyperinsulinemia, and colon cancer: a review. Giovannucci E. Am J Clin Nutr September 2007 vol. 86 no. 3 836S-842S
 Blood lipid and lipoprotein concentrations and colorectal cancer risk in the European Prospective Investigation into Cancer and Nutrition. van Duijnhoven FJB, Bueno-De-Mesquita HB, Calligaro M et al. Gut 2011;60:1094e1102. doi:10.1136/gut.2010.225011
 Erythrocyte membrane phospholipid fatty acid concentrations and risk of colorectal adenomas : a case-control nested in the French E3N-EPIC cohort study. Cottet V, Collin M, Gross AS et al. Cancer Epidemiol Biomarkers Prev. 2013 Aug;22(8):1417-27. doi: 10.1158/1055-9965.EPI-13-0168
 Vitamin B6 and Risk of Colorectal Cancer. A Meta-analysis of Prospective Studies. Larsson SC, Orsini N, Wolk A. JAMA. 2010;303(11):1077-1083. doi:10.1001/jama.2010.263
 Cancer incidence in vegetarians: results from the European Prospective Investigation into Cancer and Nutrition (EPIC-Oxford). Key TJ, Appleby PN, Spencer EA et al. Am J Clin Nutr. 2009 May;89(5):1620S-1626S. doi: 10.3945/ajcn.2009.26736M. [link]