June 6, 2017

expert reaction to study looking at drinking alcohol and the structure and function of the brain

In a new study published in the BMJ researchers report brain changes associated with chronic alcohol consumption, even at moderate levels. A Before the Headlines analysis accompanied these comments.

Prof Sir David Spiegelhalter, Winton Professor for the Public Understanding of Risk, University of Cambridge, said:

“In spite of conducting repeated large batteries of tests of cognitive function, a possible relationship with alcohol was only found with one measure of language fluency, and this seems to have been due to a decline in the initially higher scores of drinkers.  So the changes in brain matter seem to have had little impact on brain function.”

Prof. Carl Heneghan, Director of the Centre for Evidence-Based Medicine, University of Oxford, said:

“Does moderate alcohol consumption lead to adverse brain outcomes and cognitive decline?  The BMJ cohort study that tries to address this question has significant problems that prevent firm conclusions being drawn – the sample size is small, there are significant methodological problems and the results are sometimes at odds with each other.

“For example, the risk of right-sided hippocampal atrophy was significantly greater at >14 alcohol units a week compared with abstinence, but for left-sided atrophy this was only the case at >30 units a week.  Yet, there is no reason given for this discrepancy.

“Also, over the 30 year period, weekly alcohol intake did not increase in the study participants.  This suggests there were significant problems with self-reported alcohol intake in the cohort, as it does change over time.  Analysis of life-course trajectories of alcohol consumption in the UK reports a rapid increase in ‘alcohol intake during adolescence leading to a peak in early adulthood, followed by a plateau in mid-life, and then a decline into older ages’, see: https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-015-0273-z

“I also want to make an important point for interpreting results.  The use of odds ratios in the paper are misleading – they suggest the effect is much greater than it actually is.  In the no alcohol group approximately half of the people had hippocampal atrophy (see table 2).  The results in the abstract suggest that if you are a heavy drinker then your risk is five times higher.  This is not the case – as relatively your risk as heavy drinker can be only twice as high if the baseline risk is 50% (it’s actually about 1.8 times higher – 24/31 (74%) versus 9/22 (41%)).

“These type of studies also cannot account for all the confounders, and therefore they cannot, and should not, conclude causation.

“Using all the available evidence provides a much more balanced approach for the public on deciding how much to drink.  As an example, a recent BMJ observational study pointed to non-drinking being associated with increased risk of heart attack, coronary death, heart failure and stroke when compared with moderate drinking, see: http://www.bmj.com/content/356/bmj.j909.”

Prof. Tom Dening, Professor of Dementia Research and Director of the Centre for Old Age and Dementia, University of Nottingham, said:

“Firstly, this is a most impressive study and I think it will cause us all to reconsider the advice that we give to patients about alcohol consumption. I have always been a bit sceptical about the alleged protective effect of light/moderate drinking as in other studies this finding could have arisen from other confounding factors.  The data here provide no comfort for the ‘protective effect’ view. Perhaps we should all drink a bit less.

“One big question with a study taking place over 30 years is how accurate (honest) the estimates of alcohol consumption actually are. Self-reporting of drinking is not very accurate and people tend to underestimate their actual consumption, partly to appear more respectable. If actual consumption was under-reported, then the apparent adverse effects of modest amounts of alcohol could have been magnified in this study.

“The findings from research on alcohol often generate strong emotional responses, depending on people’s own views, preferences and lifestyles. One difficulty that we have is that simple messages of ‘drink less’ are not always well followed.  We probably need to think in a more sophisticated way about how we frame messages to middle-aged and older people who like a drink but may not (yet) think about alcohol in terms of causing subtle brain damage.”

Dr Elizabeth Coulthard, Consultant Senior Lecturer in Dementia Neurology, University of Bristol, said:

“This robust study takes advantage of a large cohort to demonstrate that alcohol intake is associated with reduced brain volume and integrity. There was only a hint here that high alcohol intake results in poorer memory – an effect that could predispose to dementia.  An observational study cannot truly prove that alcohol causes dementia, but the findings are in keeping with my clinical experience. The toll of high alcohol consumption on cognitive health, often evident to those of us who run memory clinics, is not widely acknowledged publicly.  Hopefully this research will contribute to a greater understanding of true safe limits for alcohol consumption that ensure protection from future dementia. Until we have further studies, the good news is that low alcohol intake was not associated with brain or memory decline in this sample.”

Dr David Reynolds, Chief Scientific Officer, Alzheimer’s Research UK, said:

“The association highlighted in this well-conducted research suggests that even moderate levels of drinking may have negative consequences for the long-term health of our brains.  The risk of declining brain health increased with higher alcohol consumption so, while we can’t identify a definitively safe level of drinking from this study, the findings suggest that people should not exceed the levels set out in UK government guidelines. The study did not show any protective effect for light drinking over being teetotal, casting doubt on the perceived benefits of reaching for ‘that one glass of red wine’.

“Studies like this, which monitor lifestyle and health over many years, are extremely useful for unpicking factors that could influence the long-term functioning of the brain, but they cannot prove cause and effect. More research is vital if we are to more fully understand the effects drinking alcohol can have on the brain and how it might affect the risk of diseases like Alzheimer’s.

“As well as limiting alcohol consumption, existing evidence points to a number of other lifestyle changes that can help people maintain a healthy brain for longer. These include not smoking, eating a balanced diet, staying mentally and physically active and keeping blood pressure and cholesterol in check.”

Prof. Paul Matthews, Division of Brain Sciences, Imperial College London, said:

“This study concludes that there is a simple dose dependence of chronic alcohol consumption on the brain. Effects on the brain extend even into what are currently considered ‘moderate’ levels of consumption (7-14 units of alcohol per week).

“While this is a relatively small study, it employs sensitive brain imaging measures; the imaging substantially enhanced the power of the study to detect effects. It showed evidence for ‘hidden’ damage to the brain (changes in some brain structures) even in people who likely were not aware of impairment of their thinking. The meaningfulness of the imaging was generally supported by consistent findings in one of the tests of function – a cognitive test of word finding abilities.

“This alone is not definitive.  Further work is needed, as the study still had limitations of size, and not all relevant variables were matched across the compared groups – there are differences in blood pressure between groups, for example, and higher blood pressure has previously been associated with white matter and hippocampal changes. Also, this study was largely confined to men.

“However, the work highlights the value of advanced imaging measures of disease – just as we measure weight, height etc. in epidemiology we can now quantitatively assess organ structures directly. This study illustrates how sensitive these assessments may be for characterising the earliest changes associated with disease. The study also suggests we need to re-examine whether there is a truly ‘safe’ level of chronic alcohol consumption – the structural brain changes were seen with moderate alcohol consumption.”

Dr Jennifer Wild, Senior Research Fellow in Clinical Psychology, University of Oxford, said:

“Medical science is under pressure to find modifiable risk factors for cognitive decline since dementia will be a global epidemic by 2050. Here the authors have discovered a robust link between what most people would consider normal levels of alcohol consumption and later degeneration of core brain regions linked to memory function.

“The results are encouraging since they suggest that reducing alcohol consumption today could prevent or delay the onset of diseases linked to hippocampal atrophy, such as Alzheimer’s.  But the study needs to be replicated and importantly, with women. Most of the sample in this study were men.

Dr Doug Brown, Director of Research and Development, Alzheimer’s Society said:

“While we’ve known for some time that drinking excessive amounts of alcohol can cause damage to the brain which can increase your risk of developing dementia, this study goes further to suggest that even moderate alcohol consumption can affect your brain’s health.

“Although this research gives useful insight into the long-term effects that drinking alcohol may have on the brain, it does not show that moderate alcohol intake causes cognitive decline.  However, the findings do contradict a common belief that a glass of red wine or champagne a day can protect against damage to the brain.

“There’s no need to fret about whether to abstain from alcohol altogether as long as you stay within the new (2016) recommended guidelines. Further research is needed to better understand whether there is any relationship between light or moderate alcohol consumption, damage to the brain and a person’s risk of developing dementia. If you are worried about you or someone else, speak to your GP.”

* ‘Moderate alcohol consumption as risk factor for adverse brain outcomes and cognitive decline: longitudinal cohort study’ by Anya Topiwala et al. will be published in the BMJ on Tuesday 6 June 2017.

Declared interests

Prof Sir David Spiegelhalter: “No conflict of interest.”

Prof. Carl Heneghan: “CH is a moderate alcohol drinker.”

Prof. Tom Dening: “I have no conflicting interests other than I admit to consuming 6 units per week.”

Dr Elizabeth Coulthard: “I have no conflicts of interest.”

Prof. Paul Matthews: “PMM is on the Steering Committee of the UK Biobank and the Executive Group of Dementia Platforms UK.  His institution has received honoraria, educational or research funding on his behalf from industry (GlaxoSmithKline, Biogen, Novertise, Roche and Adelphi Communications). PMM holds stocks in GlaxoSmithKline.”

Dr Jennifer Wild: “I have nothing to declare.”

Dr Doug Brown: “No interests to declare.”

None others received.