America’s iconic “Got Milk?” campaign was pulled this year after a successful run of over 20 years. Graced by the likes of Bill Clinton, Naomi Campbell, Elton John, David Beckham, and Angelina Jolie sporting a milk moustache, the campaign garnered wide recognition. Yet milk was losing favour against a growing variety of breakfast and drink options. The dairy industry therefore opted for a different positioning. Under the tagline “Milk Life,” the ads now tout the richness of milk’s protein content.

The evidence, though, on the health benefits of milk consumption is scant and divided.

C Mary Schooling, professor at the City University of New York School of Public Health and Hunter College, writes that dietary guidelines and recommendations are often based on findings from observational studies, which describe the eating habits of people who live longer or in better health. There are few randomised controlled trials that systematically measure the effect of specific foods on health outcomes. Sadly, opinions and assumptions sometimes drive dietary recommendations, and these may be blind to inherent differences in individuals and populations.

Milk has long been promoted as a source of protein and calcium essential for bone health. In the latest research published in The BMJ, Michaëlsson and colleagues study the association of milk consumption with mortality and fractures in two large Swedish cohorts. Using food frequency questionnaires, they collated data on consumption of milk, fermented milk, yogurt, and cheese, and linked this with information on deaths and fracture events from national registries.

Higher milk intake was observed to be associated with higher mortality in both men and women, and with an increased risk of fractures in women. Conversely, greater consumption of cheese and fermented milk products was associated with positive outcomes. They hypothesise that high D-galactose content in milk may be mediating this effect, as it has been shown to cause premature aging through oxidative stress and inflammation in animal studies.

These findings provide a new twist to the debate on milk but must be read with caution, say the authors. As an observational study, the effects of confounding and reverse causation may not be completely ruled out. The potential to extrapolate these findings to other contexts may be limited too.

As a consumer of milk and health information, I remain perplexed by what type of milk and how much of it everyday is good for me. As a doctor expected to tailor nutritional advice to individual needs and circumstances, milk continues to be a controversial dietary recommendation. There are not just health risks and benefits to be considered, but often also the availability and feasibility of consuming milk against deriving necessary nutritional value from other foods. With growing public interest in a vegan diet, the morality of consuming dairy products presents a new angle too.

Alas, the answers are not as vivid and compelling as having milk propellers, parachutes, and wings powering your everyday life, as depicted in the Milk Life ads.

Anita Jain is India editor, The BMJ.