In my youth the government encouraged people to eat more eggs and butter and drink more milk for the sake of their health. Perhaps it was the right advice after a prolonged period of war-induced shortage, but no one would offer, or take, the same advice today. Nutritional advice is like the weather and public opinion, which is to say highly changeable.
How quickly things go from being the elixir of life to deadly poison! A recent paper from Sweden in the British Medical Journal suggests that, at least for people aged between 49 and 75, milk now falls into the latter category, especially for women.
Milk was once thought to protect against osteoporosis, the demineralization of bone that often results in fractures. It stood (partially) to reason that it should, for milk contains many of the nutrients necessary for bone growth.
On the other hand, it also stood (partially) to reason that it should do more harm than good, for consumption of milk increases the level of galactose in the blood and galactose has been found to promote ageing in many animals, up to and including mice. If you want an old mouse quickly, inject a young one with galactose.
In other words, there is reason to believe both that the consumption of milk does good and that it does harm. Which is it? This is the question that the Swedish researchers set out to answer.
They followed up tens of thousands of men and women aged 49 to 75 who had filled out questionnaires about their diet two or three decades earlier and, thanks to the impeccable (but slightly unnerving) medical record-keeping of the Scandinavians, were able to estimate rates of bone fracture and death among those who drank less than one glass, one to two glasses and three or more glasses of milk per day.
Women who drank three or more glasses of milk per day were nearly twice as likely to die as women who drank one or less glass. They were also 60 percent more likely to have suffered fractures. For men, the results were less dramatic: drinking milk did not increase their rate of fractures but did increase slightly their risk of death.
As everyone never tires of saying, but always forgets, correlation is not causation. However, there are reasons for thinking that the relationship is causative: the study complies with at least some of Austin Bradford-Hill’s postulates that ought to be fulfilled for an association between a risk factor and a disease to be accepted as causative, for example a dose-response curve (the bigger the dose, the stronger the association), and the existence of a biologically plausible mechanism by which the association is brought about.
Fermented milk products such as cheese and yogurt, far from having the same effect as milk, exerted a slightly protective effect and promoted longevity. I was delighted by this because I don’t like milk but love cheese and yogurt. It is nice to know that (for now, at any rate, until future scientific refutation) one’s tastes are healthy.
The authors warn that their research needs to be confirmed by other studies and that it might not apply to other age groups, most obviously children. Moreover, the population of Scandinavia had relatively few people who are lactose intolerant; in other populations, the results might be different.
I noticed that the consumption of milk was inversely related to the consumption of alcohol: the more milk the less alcohol and vice versa. The authors did not remark on this. Could it be that, for women aged between 49 and 75, alcohol rather than milk is the elixir of life, the sovereign protector against fracture? As we doctors say, further research is needed.