Is Butter Ultimately Good for Health?
Butter would not pose a health risk and would even be favorable, according to a recent study. Other work is more nuanced. We have deciphered the scientific data for you.
Great return of butter?
Several studies have reported that butter was associated with an increased risk of cardiovascular disease. Despite everything, We have never ostracized this food or recommended to avoid it, in its nutritional advice gathered in the book The Best Way to Eat, simply encouraging moderation.
A meta-analysis published in June 2016 (1) concluded that the consumption of butter was not associated with mortality or cardiovascular disease and was low and inversely related to the risk of diabetes. Immediately, several Anglo-Saxon magazines assured that “the butter is back”. We could safely use it widely in cooking.
This was a meta-analysis combining the results of 9 observational studies with a total of 636,000 participants. However, the meta-analyzes of cross-sectional studies (without time tracking) have major weaknesses.
First, observational studies do not reveal causal relationships, except in specific cases (smoking). Second, these studies tend to mix everything which tends to diminish the strength of associations. Thirdly, in this type of meta-analysis, there is no specific comparison (eg butter versus olive oil) but a general comparison (butter against the rest of the diet). This means that the butter is compared to a wide mix of refined carbohydrates, sodas, pastries or cold cuts…
More appropriate studies such as intervention studies find different results. A meta-analysis of 15 intervention studies found that replacing saturated fats (such as butter) with unsaturated fats (such as vegetable oils) greatly reduced the risk of heart attack and cardiovascular disease.
What do the other studies say?
Another meta-analysis published on July 5, 2016 (2) provides more accurate results. This meta-analysis combined data from 2 large longitudinal studies (with time-tracking), for a total of 120,000 men and women who were healthy at the start of the study: 43,000 men The Health Professionals Follow-up Study was followed between 1986 and 2012 and 83,000 women in the Nurses’ Health Study between 1980 and 2012. Participants were asked to respond to a food questionnaire every 2 to 4 years.
Results. After taking into account other mortality factors, the researchers found that total fat intake was inversely related to mortality (-26%) compared with carbohydrates: people who ate the most fat would live longer.
This result seems to contradict the recommendation to eat less fat. According to the researchers, fat seems to have a protective effect because it is compared to carbohydrates, which are often refined (low in fiber, vitamins and minerals) and rich in sugars.
More specifically, by comparing the extreme quintiles, that is, comparing the 20% of people consuming the most (from a given nutrient) to the 20% of people consuming the least, saturated fats (animal fats, Butter) and trans fats are associated with an increase in mortality (+ 8% and + 13%, respectively), while polyunsaturated and monounsaturated fatty acids are associated with a decrease in mortality (-19% and -11% respectively).
Within the polyunsaturated fatty acids, ω-6 fatty acids (sunflower oil, walnut oil) were associated with a decrease in mortality of -15%. Surprisingly, omega-3 fatty acids, often considered more beneficial in ω-6 fatty acids, were associated with a decrease in mortality of only 4%… It should be recalled that this study being an observational study, One can not conclude on a causal link. Especially since the saturated and trans fatty acids are markers of a diet rich in processed products (industrial cakes, pastries, sausage, cheese, etc.), not very interesting nutritionally.
According to Frank Hu, one of the study leaders and professor of medicine and nutrition at Harvard, ” Our study shows the importance of eliminating trans fatty acids and replacing saturated fatty acids with unsaturated fats composed of Omega-6 and omega-3 fatty acids. This means replacing animal fat with a variety of vegetable oils. “
Taken together, these studies do not modify our recommendations, namely:
– No reason to impose a drastic limitation of fatty substances: unlike the National Health Nutrition Program, we don’t advise ” eat less fat “; The fatty substances can represent 30 to 40% of your calories.
– Avoid industrial fatty acids, the most harmful fats, obtained by partial hydrogenation of liquid fatty substances. These trans fatty acids are systematically associated with higher mortality, even in small doses. Fortunately, there are fewer and fewer diets. You have to read the labels. It is not yet known whether trans fats in fatty dairy products (cheese, whole milk) and ruminant meat (beef, sheep) are also deleterious or less.
– Limit the amount of saturated fats (red meat, charcuterie, cheese, fresh cream).Saturated fatty acids do not appear to be harmful at moderate levels. The recommendation of The Best Way to Eat is 10 to 12% of energy intake. Concretely, for an adult consuming 2000 kcal per day, this corresponds to 26 g. A 100 g portion of soft cheese provides about 20 g, a 200 g rib steak brings 16 g, while one tablespoon of olive oil (14 g) brings only 2 g.
– Prefer mono-unsaturated fatty acids (olive oil, avocado) and poly-unsaturated (rapeseed oil). Monounsaturated fats could represent 14 to 20% of your total calories, polyunsaturated 4.5 to 6.5% of calories, making sure that the omega-6 / omega-3 ratio is less than 4.
(1) Laura Pimpin, Jason HY Wu, Hila Haskelberg, Liana Del Gobbo, Dariush Mozaffarian. “Is Butter Back? A Systematic Review and Meta-Analysis of Butter Consumption and Risk of Cardiovascular Disease, Diabetes, and Total Mortality.June 29, 2016 http://dx.doi.org/10.1371/journal.pone.0158118
(2) Dong D. Wang, Yanping Li, Stephanie E. Chiuve, Meir J. Stampfer, JoAnn E. Manson, Eric B. Rimm, Walter C. Willett and Frank B. Hu. Specific Dietary Fats in Relation to Total and Cause-Specific Mortality. JAMA Internal Medicine, July 5, 2016 DOI: 10.1001 / jamainternmed.2016.2417