Intended for healthcare professionals

CCBYNC Open access

Milk intake and risk of mortality and fractures in women and men: cohort studies

BMJ 2014; 349 doi: (Published 28 October 2014) Cite this as: BMJ 2014;349:g6015

The latest study on the dangers of milk does make sense, and here is why

Consumers get to read just about everything and the contrary regarding the benefits or harmful effects of milk on health. There is an urgent need for clarification.

Let us look at this from consumers’ vantage point. You get to hear and read just about everything and the contrary about milk. The simple fact that the same word can be used for something man-made that has never come out of any mammal’s udder (rice milk, almond milk, soy milk…) is baffling in itself, and adds to the confusion. This latest scientific article deals with hip fracture risk and mortality among milk drinkers and cheese eaters. Data does not provide any definite answer. Still, it is possible to draw a few common sense tips from it, based on evidence.

It was an observational study: the research team monitored two large Swedish cohorts, one with 61 433 women (39-74 years at baseline 1987-90) and one with 45 339 men (45-79 years at baseline 1997). They were submitted food frequency questionnaires. The women responded to a second food frequency questionnaire in 1997. Their health status was assessed using health insurance data, in order to know if they suffered from a hip fracture or if they died. This method makes the study very reliable in terms of following up on the patients.

Here are the results:

High milk intake (3 glasses of milk a day or more) was associated with higher cardiovascular and cancer mortality, as well as and with higher fracture incidence in one cohort of women. With male participants, we observed a higher cardiovascular mortality. The risk is limited among men but significant among women.

Why is a cautious interpretation of the results recommended? Well, here we see correlation and not causality. This is not an interventional study and we did not ask a group of people who did not consume milk to start consuming it. Given the observational study design, there is the inherent possibility of residual confounding and reverse causation phenomena.

Let us also note that fermented dairy products consumption was not linked to higher mortality or fracture incidence.

The research team offers several hypotheses to explain the correlation between higher consumption of milk and higher hip fracture and morality. What can we take out of it?

1. “Milk = Calcium = Bone health” is simplistic and most likely erroneous, if we do not account for other factors. This is not news, since milk does not help fighting osteoporosis nor heal fractures (Feskanich D, Willet WC, Stampfer MJ, Colditz GA. Milk, dietary calcium, and bone fractures in women: a 12-year prospective study. Am J Public Health 1997;87:992-7). Bone’s complex structure is more relevant to its health than calcium actually is. If the bone structure is deficient, calcium will not properly be integrated and the bone will break easily. The structure is made out of proteins (50% of bone volume, 1/3 of its mass) and it gets stronger with exercise and vitamin D production due to exposition to the sun. Our indoor, sedentary lifestyle is what weakens the bone. Therefore we need to use our muscles as they pull on the bone, which is what induces bone strengthening. We also need to go out in the sun in order to activate vitamin D production, or take food supplement in the winter to make up for the lack of sun. Those are the prime factors to consider before calcium.

2. Throughout the evolution of the human species, we swapped plants and vegetables for cereals. They contain but little calcium and can promote decalcification because of phytic acid. We ought to avoid calcium supplements, as they increase cardiovascular mortality, at least among men. Therefore milk is really useful for nutrient intake. It should not be the sole calcium intake though, and vegetables are both accessible and affordable, provided we select those in season. We ought to increase the quantity of fruits and vegetables at the expense of processed foods, mostly those derived from wheat (pastries, cake, industrial bread, pasta etc.) for a varied calcium intake. In vegetable leafs (salad, spinach, Swiss chard, parsley, algae) calcium is highly absorbable.

3. Which nutrients other than calcium are essential for the bone? Well, there are at least two.
First of all, proteins: dairy products are a prime source, vegetable proteins, meat and fish. The protein intake among seniors is insufficient.
Then long-chain omega-3 acids: they are synthesized by fatty fish for the most part. The amount in our body really depends on diet and is correlated with a decreased risk of hip fracture among women. Sardines, mackerel, tuna and anchovies are excellent sources of long-chain omega-3 acids.

4. According to this Swedish study and other research, it seems very likely that milk is less adapted to our adult needs than cheese, especially pressed cheese. Why? Several hypotheses are discussed by the authors. In particular, there is one about lactose, the sugar in milk being made of a molecule of glucose and one of galactose (sugar was discovered by Pasteur in 1856). But galactose induces a form of oxidative stress and glucose, hyperglycemia similar to what happens in the body after fruit juice absorption. Oxidative stress and hyperglycemia promote chronic inflammation, which leads to cardiovascular diseases and bone fragility. Hard, pressed, ripened cheese contains less glucose, since it is used by bacteria for fermentation, and a bit less galactose. Some cheeses contain just traces of them. This could explain such a big gap in effects: milk and non-fermented dairy products can have deleterious effects on adults, but are necessary to new-born and children, while other cheeses are neutral. Study authors showed chronic inflammation among milk drinkers, in proportion to their consumption.

5. In view of the stakes, the controversies on milk being “a poison” or “cancer inducing” are senseless. They promote food scares, which are a way to pursue business to some food brands. It is also noteworthy that we consume much less milk than people from Northern countries do. However, when it comes to food recommendations, “one size fits all” simply doesn’t apply. Current nutritional recommendations are just too similar. Some digest lactose very poorly (most notably Asians and Africans), other will digest sheep or goat milk much easier.
These natural adjustments are often made by the consumer himself, since enzymatic resources are linked to our own genomics. Some more difficult cases can be managed by an allergist.
But ultimately, pressed, ripened cheeses are well digested and preferred to milk consumption among adults. Beyond the milk / cheese duality, the dairy products industry has seen its offer evolve a lot these last few years and many more products derived from milk are now available. These industrial dairy products are often made out of little or not-fermented milk, or even skimmed milk. In that sense, they are included in the amount of milk consumed. Furthermore, they are almost always sweetened, which our metabolism can hardly handle anymore. It is best to avoid them since there are already so many sources of sugar (simple sugars, starch of all sorts).

Lastly, this study shows no evidence that fat-free is any better. Quite the opposite in fact: cheeses not correlated with increased mortality are, for the vast majority, made with whole milk. Nothing in this study could tell us to avoid fats from milk or cheese.

Competing interests: No competing interests

04 December 2014
Guy-André Pelouze
Thoracic and cardiovascular surgeon
St John's hospital, Perpignan France
20 Languedoc avenue 66046 Perpignan France