Fearing the wurst: the association of red and processed meats with chronic disease risk

From the Dr. Johanna Lampe, Dr. Marian Neuhouser, and Dr. Ross Prentice Labs, Cancer Epidemiology, Prevention and Control and Biostatistics and Computational Biology Programs, Cancer Consortium.

I want you to take a moment to brainstorm “everyday” cancer risks. You know: the actions or interactions that are commonly accepted as heightening your risk of developing cancer. I’m a redhead, so sunburns and their association with skin cancer risk are constantly on my mind. I would wager a guess that a few of you thought of excessive smoking and the risk of lung cancer. But how many people immediately think of red meat as carcinogenic, or cancer-causing? In fact, whether red and processed meat consumption contribute to chronic disease risk is a major controversy in the field of Nutritional Epidemiology – just ask Dr. Cheng Zheng, of the University of Nebraska Medical Center Department of Biostatistics, and Drs. Ross Prentice and Marian Neuhouser, of the Fred Hutch Division of Public Health Sciences. They recently published a study in the Journal of Nutrition and Nutritional Epidemiology seeking to clarify the role that red and processed meat play in chronic disease risk in the diets of post-menopausal women.


“In 2015, the International Agency for Research on Cancer (IARC) declared processed meat a carcinogen and red meat as a probable carcinogen,” explains Dr. Neuhouser. “This declaration arose from the totality of evidence to date including experimental data in animals and research in humans that examined meat intake and cancers.” However, Dr. Prentice adds, “an international research group recently published four papers in an issue of Annals of Internal Medicine arguing that…evidence of adverse health effects from the consumption of red meat, and especially processed meat, was weak and that any such adverse effects had to be small. [Based on this meta-analysis,] the authors recommended that people continue their usual meat consumption habits. Prominent nutritional epidemiologists were aghast at this advice, and they argued that evidence of the adverse health effects of meat consumption on major chronic diseases was strong and consistent”. When head-to-head controversies like these arise, researchers often take a step back and examine their methods of data collection. It becomes particularly critical to standardize the way you conduct each experiment, especially if you intend to perform meta-analyses on studies from multiple research groups. Dr. Prentice pointed out that “a difficulty with the body of nutritional epidemiology research on this…is that in most studies, the dietary consumption data is assessed through self-report, with associated measurement errors possibly dominating the results of disease association analyses”. Your study participants might not measure portions the same way, or they might accidentally forget to record foods they ate if they are too busy with their day-to-day lives. This unreliability of self-reporting has led Drs. Neuhouser, Prentice, and others to turn to metabolites from urine and serum samples to standardize their evaluations of participants’ diets. Metabolites, the byproducts of your body breaking down food or drugs, can be used a biomarkers, or indicators of the presence or absence of diseases. The Fred Hutch-based research team hoped that using biomarkers could help them calibrate and standardize their large datasets, to improve the quality of their data and conclusions.

Packed cuts of red meat in a grocery store deli case.
In their recent study, Fred Hutch Women’s Health Initiative researchers Dr. Ross Prentice, Dr. Marian L Neuhouser, and Dr. Johanna W Lampe seek to define whether or not diets high in red meat and processed meats contribute to heightened risk of chronic disease. Figure provided by Dr. Marian Neuhouser

Fred Hutchinson Cancer Center is lucky in that, as part of the Cancer Consortium, it is home to the Nutrition and Physical Activity Assessment Study, a research group that includes “an interdisciplinary group of colleagues at [the University of Washington] and throughout the US via the Women’s Health Initiative (WHI),” explained Dr. Prentice. This talented research group began to comb through their own data for any evidence that would support or undermine the idea of red and processed meats as carcinogens. Specifically, they dug into prior data from studies spanning 1993 to 2014 that included data from serum and urine samples, to generate deeper and more reliable models calibrated with biomarkers, instead of relying on self-reported food diaries or questionnaires. “Our lightbulb moment came when we decided to use reports of metabolites associated with meat intake from other studies, including small scale feeding studies in Europe, to filter our metabolite profiles, much reducing their dimensionality,” said Dr. Prentice. Dimensionality is another way of describing the complexity of a dataset – when you reduce a dataset’s dimensionality, you are effectively eliminating the variables that might be responsible for the effect you are observing. “Biomarker identification, with metabolites restricted to this much smaller set, was successful for both red meat and for red plus processed meat combined. This put us in a position to use the newly identified biomarkers to correct self-reported meat intake data for measurement error, and to associate the resulting corrected, or calibrated, meat intakes to cardiovascular disease, cancer and diabetes in large WHI cohorts of over 80,000 US postmenopausal women”. This was a major win for the WHI team, as examining data from 80,000 individuals greatly increased their statistical power.


So, what did they find in the end? Well, as with many things, it’s complicated! “The analyses of calibrated red meat intake, as well as combined red plus processed meat intake, with chronic disease risk in the follow-up of WHI cohorts showed positive associations with coronary heart disease and other cardiovascular outcomes, with breast and colorectal cancer and several other invasive cancers, and with diabetes incidence,” explained Dr. Prentice. “However, these associations essentially disappeared when controlling additionally for other dietary variables that tend to be a part of a high meat diet; namely, high intake of saturated fat, sodium, and total calories. Hence, these overall dietary patterns, rather than red or processed meat intake per se, appeared to be responsible for the observed adverse health effects of meat intake, at least in this population”. That is to say, the overall context in which a person was consuming red or red and processed meat was just as important, if not more important, than the actual meat consumption. If your diet is high in red meat, but is also high in sodium and saturated fats, it could be more likely to lead to future health issues than someone who occasionally eats red meat but has an otherwise very balanced diet. “Our results indicate that it’s a good idea for US populations, most directly for postmenopausal women, to avoid a dietary pattern that is high in meat content especially if that pattern…includes high intakes of saturated fat, total calories and sodium,” Dr. Prentice concluded.

C Zheng, M Pettinger, GAN Gowda, JW Lampe, D Raftery, LF Tinker, Y Huang, SL Navarro, DM O’Brien, L Snetselaar, S Liu, RB Wallace, ML Neuhouser, and RL Prentice. 2022. Biomarker-calibrated red and combined re and processed meat intakes with chronic disease risk in a cohort of postmenopausal women. The Journal of Nutrition: Nutritional Epidemiology. 152:1711–1720. DOI: https://doi.org/10.1093/jn/nxac067.

Cancer Consortium members Dr. Johanna W Lampe, Dr. Daniel Raftery, Dr. Ying Huang, Dr. Marian L Neuhouser, and Dr. Ross L Prentice contributed to this work.

This work was supported by the National Heart, Lung and Blood Institute, the National Institutes of Health, the US Department of Health and Human Services, and the National Cancer Institute.