"Red Meat Consumption and Mortality" from the Harvard School of Public Health, published online March 12, 2012 in the Archives of Internal Medicine, ignited something of a media firestorm. Headlines proclaimed, "Scientists warn red meat can be lethal" and "Red meat is blamed for one in 10 early deaths," based mostly on the study's conclusion that each additional serving of red meat consumed daily was associated with a 13 to 20 percent increased risk of "all-cause mortality." The researchers suggested that almost one in 10 of the deaths of the study participants could have been prevented if they had kept their red meat intake under a half-serving daily.
It's an interesting study, but there are some important facts to consider:
- This was an observational study based on data gathered from the Health Professionals Follow-up Study (37,698 men) and the Nurses' Health Study (83,644 women). Observational studies can only establish correlation, not causation. Ideally, an observational study generates a hypothesis that can then be tested in a controlled manner in order to determine what's cause, and what's effect.
- The study was based on answers to food-frequency questionnaires, a notoriously unreliable way to track actual consumption.
- Frequent red meat eaters tend to have less healthy lifestyles overall than others. As the authors noted, "Men and women with higher intake of red meat were less likely to be physically active and were more likely to be current smokers, to drink alcohol, and to have a higher body mass index. In addition, a higher red meat intake was associated with a higher intake of total energy but lower intakes of whole grains, fruits, and vegetables." In other words, Americans who eat lots of red meat are also more likely to be sedentary, smokers, alcohol drinkers, overweight, and eat more calories and fewer plant foods. The researchers adjusted their data to try to account for these behaviors - known technically as "confounders" - but this is a challenging task in such studies, and it's difficult to know how appropriate or successful they were.
- While some effort was made to distinguish processed from unprocessed meats (the former, unsurprisingly, were found to be worse for mortality), this study made no effort to differentiate between grass-fed, grass-finished organic beef or pork; and grain-fed, antibiotic-and-hormone dosed, massively stressed factory-farmed beef or pork.
My take: I don't believe anyone needs to eat red meat to be healthy. We can get the protein and essential fatty acids we need from other sources, such as wild-caught, cold-water fish; free-range, omega-3 rich eggs; and tofu, beans and nuts.
If you do eat red meat, less is better than more, and grass-fed, grass-finished beef offers a far better omega-3 to omega-6 fatty acid profile, along with fewer contaminants and less fat.
At the same time, the superficial modern-media takeaway and "crisis headline"- red meat is lethal - is highly suspect. Human beings have a long evolutionary history of consuming meat from healthy, foraging four-legged animals, so it is unlikely that this practice is extraordinarily hazardous to health.
Conversely, we have a very short evolutionary history of eating highly processed and unnatural foods, especially industrial beef and pork, refined sugars and polyunsaturated vegetable oils. Being sedentary, smoking, and avoiding fruits and vegetables are also rather recent widespread human behaviors for which we are poorly adapted. I believe that these lifestyle habits - rather than eating red meat per se - are likely to be the true causes of increased mortality.
A far more useful study would be one that compared the health of active, nonsmoking, normal-weight, fruit-and-vegetable eating people, divided into groups depending on whether they get most of their protein from:
- vegetarian sources such as beans, tofu and nuts
- grass-fed, grass-finished beef and free-foraging pork
I suspect that members of all three groups would show significantly lower "all-cause mortality" than that of most Americans and that the fish-eating group would have the lowest risk by a narrow margin. But we won't know if that's true until such a study is actually conducted.
Andrew Weil, M.D.