Long-life milk?

26 july 2009--”A pint of milk a day cuts chances of heart disease and stroke by up to a fifth,” The Daily Telegraph has said. The nation’s favourite dairy drink is also said to cut risk of developing diabetes and colon cancer. The findings may challenge the view held by some that too much dairy is bad for you.

These findings come from a systematic review that has combined the results of several observational studies, which found that consuming higher amounts of milk or dairy products is associated with reduced risk of heart disease, stroke and diabetes.

However, there are some limitations of the study that have to be considered when drawing conclusions from these results, in particular that the studies reviewed used variable methods for assessing milk consumption and it is possible participants misreported the milk they drank. Several other factors not measured in this study could be playing a role in disease risk, such as other dietary patterns, physical activity and lifestyle habits. Furthermore, variable and inconclusive results were also obtained for fat content of milk, meaning the study could not compare whole milk with low-fat milk.

Where did the story come from?

Peter Elwood and colleagues of Cardiff University, the University of Reading and the University of Bristol carried out this research. No sources of funding were reported. The study was published in the peer-reviewed Journal of the American College of Nutrition.

What kind of scientific study was this?

This is a systematic review and meta-analysis in which the researchers used cohort and case-control studies to investigate whether milk and dairy consumption affected the outcomes of vascular disease and diabetes.

The authors searched the Medline medical database for relevant studies using the phrases milk, milk protein, dairy, dairy calcium, heart disease, coronary artery disease, myocardial infarction, ischaemic heart disease, stroke, diabetes or metabolic syndrome.

The authors included studies that collected data on milk consumption at the start of the study and then followed people up over a period, examining a range of medical outcomes.

Among the 324 studies identified by the search there were 11 suitable studies on milk products and heart disease, seven on milk and stroke and four on milk and diabetes/metabolic syndrome. The researchers pooled the results of these relevant studies to determine the risk of their respective outcomes in relation to levels of milk consumption.

Within these individual studies there had been attempts to make statistical adjustments to account for the influence of confounders, though the exact methods of adjustment varied between studies. The authors then obtained extra data from studies that gave disease risks in relation to the type of milk consumed, e.g. whole or low-fat.

Finally, the authors summarised conclusions from the recent report by the World Cancer Research Fund and American Institute for Cancer Research, looking at observations between cancer development and milk consumption.

What were the results of the study?

The authors pooled the results of 15 cohort studies examining risk of heart disease and stroke, featuring over 600,000 participants and extensive follow-up times, in the range of 8 to 25 years. They found that the risk of heart disease in subjects with the highest milk or dairy consumption was reduced by 16% compared to those with the lowest consumption (RR 0.84, 95% CI 0.76 to 0.93). When looking solely at the seven studies examining stroke events, the researchers found that stroke risk was reduced by about 21% (RR 0.79, 95% CI 0.75 to 0.82).

The combined results of the four studies examining the development of diabetes depending on milk consumption found that the risk was reduced by 8% among those with the highest milk intake (RR 0.92, 95% CI 0.86 to 0.97).

The studies examined featured adjustments for various confounders including age, sex, BMI, smoking, physical activity, social class, cholesterol and blood pressure.

When the authors looked across all of the studies for any that had given separate results for whole and low-fat milk, risk results were highly variable and were generally not significant.

The researchers also report on other studies carried out that had made similar observations to their own results. Four case-control studies observed that high milk consumption reduced the risk of metabolic syndrome, which is a combination of risk factors occurring together of elevated blood glucose, high cholesterol, being overweight or obese, and high blood pressure (RR 0.74, 95% CI 0.64 to 0.84).

Additionally, four case-control studies asking women who had had a heart attack about their prior consumption of milk, found 17% reduction in risk from drinking the highest quantity of milk (RR 0.83, 95% CI 0.66 to 0.99).

The Report of the World Cancer Research Fund was examined to look at data on the relationships between various cancers and dairy consumption. The report’s results were based on a variety of cohorts and case control studies. Variable associations were found for studies examining prostate, colon and bladder cancer, and no associations were found for other cancers.

What interpretations did the researchers draw from these results?

The authors conclude that the results of their review provide evidence of an overall survival advantage from consuming milk and dairy products, highlighting the high proportion of UK deaths currently attributable to vascular disease, cancer and diabetes.

What does the NHS Knowledge Service make of this study?

This review, which has combined the results of various observational studies, found that consuming higher amounts of milk or dairy products is associated with reduced risk of heart disease, stroke and diabetes.

However, the systematic review pooled the results of studies of varying quality, study length, inclusion criteria, disease outcomes and methods of assessing milk or dairy consumption. These individual studies may have had several biases. There are also other aspects of this study that must be considered when interpreting the results of the study:

• The researchers looked at case-control studies in which a person has already experienced the disease outcome, e.g. heart attack or stroke, and is then asked to recall their past consumption of milk. This may involve recall bias, where a person who has the disease recalls differently compared to those who have not, as a way to try to find a possible explanation.
• Additionally, the review pooled data from cohort studies, which have a design that may be more reliable for assessing causation as the person hasn’t yet developed the disease. However, these individual studies had considerable variability in their methods.
• One important variation was that milk consumption had been variably assessed through food questionnaires or 24-hour food recall, and such estimates are likely to involve some inaccuracy. Also, the studies used variable exposure categories. For example, some studies compared people who drank milk to people who did not. Others looked at number of days of the week milk was drunk on, others at the number of pints or glasses drunk per day or per week. As such, it is very difficult to gain any indication of the optimal quantity of milk to consume. Furthermore, it is unclear whether other dairy sources such as cheese, yoghurt or cream had been assessed.
• There were differences in results between some of the cohort studies. As the authors say when pooling the studies evaluating risk of heart disease they excluded the results of one study, in which reduced risk was observed from consumption of low-fat milk, but increased risk with consumption of whole milk. This finding differed from the other studies pooled.
• The individual studies had attempted to adjust for different confounders but there was inconsistency between studies in the factors that had been considered. Particularly, important lifestyle ones such as other dietary habits or physical activity, smoking or alcohol consumption may be confounding results.
• Randomised controlled trials would be the best way to investigate the health benefits of drinking milk but the researchers think these are unlikely to be practical to perform.
• There were no consistent results associating milk with risk of any type of cancer. Also, variable and inconclusive results were obtained in assessment of whole compared to low-fat milk.
• Conclusions that drinking more milk reduces risk of death are only indirect ones, following on from the fact that heart disease, stroke and diabetes are a significant cause of morbidity and mortality in the UK. The studies in this review have not actually examined death rate, survival or quality of life in those who did or did not develop disease.

Links to the headlines

A pint of milk a day cuts chances of heart disease and stroke. The Daily Telegraph, July 23 2009

Drinking milk cuts ‘risk of dying from heart disease and stroke by one fifth’. Daily Mail, July 23 2009

Links to the science

Elwood PC, Givens DI, Beswick AD et al. The Survival Advantage of Milk and Dairy Consumption: an Overview of Evidence from Cohort Studies of Vascular Diseases, Diabetes and Cancer. JACN, 2008