September 7, 2016

expert reaction to study investigating the association between physical activity, alcohol drinking, and mortality

The extent to which physical activity can mitigate the harmful effects of alcohol on health is examined in a paper published in the British Journal of Sports Medicine in which the authors report that the association between alcohol intake and mortality risk was reduced in individuals who met physical activity recommendations.

Prof. Matt Field, Professor of Psychology, University of Liverpool, (part of the UK Centre for Tobacco and Alcohol Studies (UKCTAS)), said:

“This is a rigorous piece of research with some clear conclusions: the relationship between drinking alcohol to excess and increased risk of death is significantly weaker in people who are physically active (i.e. those who do at least 7.5 ‘metabolic equivalent task’ hours of exercise per week, which is in line with current guidelines for physical activity). Therefore, it appears that physical activity may partially offset some of the harmful effects of drinking, particularly alcohol-attributable cancers.

“However, it is important to bear in mind some limitations of the study. Firstly, this is a cross-sectional study so it is important to consider alternative explanations for the findings. For example, people who are already ill may be less active than those who are healthy. The authors report some additional analyses which suggest that these alternative explanations are unlikely, but further research is required to clarify this. Secondly, the researchers quantified alcohol consumption in terms of the number of units consumed each week, but they were unable to look at the role of different types of drinking (e.g. occasional binge drinking versus drinking a little alcohol on several days). This is important because different patterns of drinking may be differentially associated with mortality risk. Finally, the researchers only looked at data from people who were at least 40 years old, probably because premature death (all cause, from cancer, and from cardiovascular disease) is much less common in younger adults. Nonetheless, this means that we do not know how these findings might generalise to younger adults, particularly if they primarily binge drink rather than spread their alcohol units over the week.”

Prof. Kevin McConway, Emeritus Professor of Statistics, The Open University, said:

“This is a competent study, using a great deal of data. But we’ve got to be careful in interpreting what it means.

“In broad terms, the new study confirmed what many previous studies have found. Drinking alcohol, beyond very low levels at any rate, increases the risk that one will die in a given period of time, from cancer or from all causes of death taken together. Broadly, the more that is drunk, the greater the additional risk. But what’s new is that this study looked at whether this increase in risk with the amount a person drinks is affected by the amount of physical exercise they take. The researchers found that, for people who exercised according to the recommended guidelines, the risk increased more slowly than in those who exercised less. That pattern applied when all causes of death were looked at together, and also for deaths from cancer (where the effect of exercise was more marked).

“Does this mean that I don’t have to worry about the effect of drinking on my health, as long as I get enough exercise? Well, no it doesn’t, for a long list of reasons. As the press release makes clear, “this is an observational study so no firm conclusions can be drawn about cause and effect”. In other words, it’s possible that the changes in risk patterns in people who exercised were not because of the exercise, but because of something else that just happened to be different, in the people studied, between those who exercised a lot and those who didn’t. The report points out that the researchers could not take into account drinking patters such as binge drinking, or dietary factors, because they had no data on them. Maybe the people who exercised a lot tended to have different diets, or different drinking patterns, and maybe that is what changed the risk pattern and not the exercise at all. It’s also important to note that the results are all based on how the people involved reported their drinking and their exercise when they entered the surveys involved. There are always potential inaccuracies in such self-reported data – for instance, people who drink a lot might say they drank less than they actually did – and in any case, both drinking and the amount of exercise could change in many ways between the time the data were collected and the time the researchers stopped following them up (which was, on average, about 10 years later). And, if you’re young, you have to bear in mind that all the data used here are for people aged at least 40 when they entered the study. The results say nothing direct about younger people.

“Even if the exercise does on its own change the risk pattern, it’s important to remember that risk of death in a given time did still increase with alcohol intake, and that was particularly true for those who drank at a level the researchers labelled as ‘harmful’ (More than 35 units a week for women, more than 49 a week for men.) So acting on health advice on exercise certainly doesn’t mean you can ignore health advice on alcohol.

“None of this is meant to be a criticism of what these researchers did – I’m just pointing out the difficulties of drawing conclusions on how an individual should act, on the basis of studies of this kind. In fact, this study itself points out some of the problems that arise in dealing with observational studies. The new study gives a pretty clear indication that exercise should be taken into account when looking at the relationship between alcohol and health. Some previous studies have indeed taken exercise into account in their statistical analysis, but far from all. This new study also emphasises a point that has been made before – it doesn’t really make sense to compare the mortality of people who drink different amounts of alcohol, using all people who do not drink as a basis for comparison. Non-drinkers include people who never drank, but also people who have given up, and they may have given up for reasons related to illness, and the illness may affect their risk of death. Some previous studies didn’t take account of this. Some others did, but could include rather few people who had never drunk alcohol, which makes for statistical difficulties. Because this study included large numbers of people, it included enough who never drank to make a decent basis for comparison. ”

‘Does physical activity moderate the association between alcohol drinking and all-cause, cancer and cardiovascular diseases mortality? A pooled analysis of eight British population cohorts’ by Perreault et al. published in British Journal of Sports Medicine on Wednesday 7th September. 

Declared interests

Prof. Matt Field: I have received funding from the Medical Research Council, Wellcome Trust, Economic and Social Research Council, and Alcohol Research UK. I have no other conflicts of interest.

Prof. Kevin McConway: No conflicts of interest to declare