Heavy drinking is a well-established risk factor for poor health and premature death. However, the health effects of drinking in older age remain debated. Some observational studies suggest that moderate drinkers have a lower risk of death than non-drinkers, but these findings may be misleading because of differences in health and lifestyle between the groups. Our new paper in the European Journal of Epidemiology investigated this question among older British men.
What did we do?
We analysed data from 4,582 men in the Whitehall-1 study of London civil servants. Their alcohol consumption was self-reported by postal questionnaires in 1997-98, when they were aged about 76 on average. We excluded men with a history of cardiovascular disease or cancer at that time. The men were followed for an average of nearly 12 years (up to 23 years).
We examined how alcohol consumption related to deaths from all causes and from specific diseases. We compared different groups with current light drinkers (who drank fewer than 7 UK units a week at the time of assessment) and distinguished former drinkers from lifelong non-drinkers.
For context, one UK unit contains 8 g of pure alcohol, and the UK low-risk drinking guidelines advise both men and women not to drink more than 14 units a week on a regular basis.
What did we find?
Among current drinkers, higher alcohol consumption was associated with a higher risk of death. For every additional 10 units consumed per week, the risk of death was about 5% higher. Men drinking 42 or more units a week had approximately 40% higher risk of all-cause mortality than light drinkers. A similar association was seen for deaths from non-vascular causes, particularly alcohol-related cancers, whereas the findings for deaths from vascular causes were less clear.
At lower levels of alcohol consumption, we found no clear evidence of differences in overall and non-vascular mortality between lifelong non-drinkers, light drinkers, and those drinking 7-27 units a week. However, former drinkers had a 21% higher risk of death than light drinkers. One possible explanation is that some people stopped drinking because of poor health, making those who continued drinking appear healthier by comparison. This could partly explain why moderate drinking has appeared beneficial in some previous studies, an issue we and others have discussed before.
What do these findings mean?
Our findings suggest that heavier drinking is associated with higher mortality in older men, but they also need to be interpreted cautiously. The sample size was relatively small, so some of our estimates are uncertain, particularly for vascular deaths and other cause-specific deaths. Also, alcohol consumption was self-reported and measured only once, in later life. As an observational study, we could not fully account for other differences in health and lifestyle between drinking groups. The study also included only older British male civil servants, meaning the findings may not apply equally to women or other groups of older people.
Nevertheless, our study adds to the evidence on alcohol and health in older age, and fits with wider evidence on alcohol’s harms. It also highlights why apparent benefits of moderate drinking need to be interpreted carefully, particularly when former drinkers and lifelong non-drinkers are not distinguished.
These findings are particularly relevant as older people account for a growing share of heavy drinkers in England. According to a recent Nuffield Trust analysis, the proportion of heavy drinkers aged 65 and over increased from 12% in 2012 to 21% in 2022, highlighting the need for alcohol policies that better address older people’s needs. Our results add support to efforts to reduce alcohol consumption among older people, while offering no convincing evidence that moderate drinking helps people live longer.
Written by Dr Stephanie Ross, Teaching and Research Fellow at Nuffield Department of Population Health, University of Oxford and Dr Pek Kei Im, Epidemiologist and Wellcome Career Development Fellow, Nuffield Department of Population Health, University of Oxford.
All IAS Blogposts are published with the permission of the author. The views expressed are solely the author’s own and do not necessarily represent the views of the Institute of Alcohol Studies.
Reference: Yusuff S, Fohrmann A, Sherliker P, Leon DA, Kivimaki M, Emberson JR, Clarke R, Lewington S, Im PK, Ross S. Alcohol consumption and cause-specific mortality in older men: Whitehall-1 study of London civil servants. European Journal of Epidemiology 2026. doi:10.1007/s10654-026-01465-4