Harms of alcohol in different age groups
BMJ 2022; 378 doi: https://doi.org/10.1136/bmj.o1979 (Published 10 August 2022)Cite this as: BMJ 2022;378:o1979
August 10, 2022
Concern should be focused on older drinkers
“Our message is simple: young people should not drink, but older people may benefit from drinking small amounts,”1 was the eye catching conclusion of a recent study from the Global Burden of Disease project.2These results seem to have important implications for clinical practice: should we be advocating abstention in the under 40s while encouraging older teetotallers to take up drinking for health reasons?
The study estimated how the risks of harm, measured in disability adjusted life years, vary with alcohol consumption, age, sex, and country and estimated the level of alcohol consumption at which drinkers face the same risks as non-drinkers. The authors find that, in many parts of the world, including western Europe, younger drinkers face increased risks of harm with any alcohol consumption. In contrast, older drinkers see their risks of harm fall at low levels of drinking before rising at higher levels of consumption.
These differences arise because of the different profiles of risk faced by each age group. A much greater proportion of overall harm among under 40s comes from injuries, and any alcohol consumption will exacerbate this risk. Adults over 40, however, have a much higher prevalence of cardiovascular disease, and low levels of alcohol may reduce this risk.
On the surface, these findings seem to support a greater focus on the drinking habits of younger people, but important additional factors should be considered before we change clinical practice, advice, or guidance.
Firstly, this study compares risk only within age groups–that is, whether 30 year olds who drink face greater risks than 30 year olds who abstain. The authors do not compare risks between age groups. Nor do they look at absolute levels of risk. Without knowing the absolute magnitude of alcohol related risks in different age groups we cannot compare them with the myriad other risks we all face and accept from everyday behaviours such as driving a car or boiling a kettle. Recent reviews of drinking guidelines in the UK3 and Australia4 have focused on an “acceptable risk” approach, setting guidelines that maintain absolute risk at levels comparable with those from other everyday activities.
On this basis, the risks faced by younger people are substantially lower than those faced by older people. The Global Burden of Disease project’s own analysis shows that globally, it is people in their 60s who face the highest rates of alcohol attributable harm, far higher than the rates faced by those under 40.5 It therefore seems excessively paternalistic to suggest that people under 40 should not drink at all on health grounds, given the low absolute risks faced by young people who drink moderately.
Secondly, the conclusions of the paper do not acknowledge the substantial remaining uncertainty around the existence, or otherwise, of a protective effect of low levels of alcohol consumption on cardiovascular risk. Many observational studies report protective effects, and plausible biological mechanisms might explain them, but this research is limited by consistent methodological problems.67 Several studies using alternative approaches have found no evidence to support protective effects.89 The suggestion that we might encourage older people to consume small amounts of alcohol for the benefit of their health seems premature, potentially even unethical, since the apparent benefits may be illusory.10
Thirdly, the study’s authors find little difference in risk between men and women at similar levels of alcohol consumption and conclude there is little justification for separate guidelines for men and women. This may be true, but it is important to acknowledge the potential risks associated with drinking in pregnancy (such as fetal alcohol spectrum disorders) and provide appropriate information and guidance.
Finally, any linking of alcohol consumption with potential health benefits for older adults should be done with particular caution in settings where older drinkers are heavier drinkers. In the UK, for example, adults aged 55-74 are more likely than other adults to drink alcohol, and those who do, drink more alcohol on average than any other age group.11
Overall, this new study provides some interesting new insights into the risks associated with drinking but finds little to warrant a change in practice. There may be good reasons to tackle problematic drinking among younger people, but if we want to reduce the substantial burden of societal harm caused by alcohol, our attention may be better directed at older drinkers.
Competing interests: The BMJ has judged that there are no disqualifying financial ties to commercial companies. The author declares no other interests. Further details of The BMJ policy on financial interests are here:
- Provenance and peer review: Commissioned; not externally peer reviewed.