Intended for healthcare professionals


Healthy behaviours yield major benefits in ageing

BMJ 2013; 347 doi: (Published 20 August 2013) Cite this as: BMJ 2013;347:f5156

Re: Healthy behaviours yield major benefits in ageing

Professor Kenfield and colleagues highlight the null findings when examining an association between alcohol consumption and health outcomes (1), referring to 2 papers from the nurses’ health study that suggests some benefit to health in drinking between 5 and 30g of alcohol per day, which equates to approximately between 0.6 and 3.75 UK units of alcohol per day. For older women who drank up to one drink per day, the second study concluded that this amount ‘does not impair cognitive function and may actually decrease the risk of cognitive decline’. This figure is identical to the recommended limit stated in Our Invisible Addicts (2).

Over the past 30 years, there has been much speculation over the putative protective effects of alcohol on health, much of which has centred on a ‘J’ or ‘U’ shaped relationship between the amount of alcohol consumed and the presence of health problems or deaths associated with certain diseases. The most definitive conclusions regarding this relationship have been drawn from an analysis of 84 cohort studies examining the association between alcohol consumption and health outcomes (3). This meta-analysis concluded that the lowest risk of coronary heart disease mortality occurred with between 1.5 and 3 UK units of alcohol per day, but for stroke mortality lower risk was conferred by an intake of 1.5 or fewer units per day. In addition, the analysis found that there was a 14–25% reduction in the risk of all outcomes associated with drinking 1.5 or fewer UK unit of alcohol per day assessed compared with those abstaining from alcohol.

The association between moderate drinking and mortality from coronary heart disease replicates a previous study in men aged 40-59 which concluded that this association was see only with coronary heart disease, with an increased risk of death from accidents, cancer and stroke rising from drinking above approximately 1.5 UK units of alcohol per day (4).

However, what has remained conclusive evidence and beyond reasonable doubt for many years in demonstrating a cardio-protective effect of alcohol, has now been called into question. A more recent meta-analysis of 44 observational studies pooling nearly 40,000 coronary events (mortality and ill health) found considerable variation across studies, particularly for an average consumption of between 1.5 and 3 UK units of alcohol per day. It concluded that a cardio-protective association between alcohol use and heart disease cannot be assumed for all drinkers, even at low levels of intake (5). Much of the criticism of studies that have found health benefits from moderate drinking have been leveled at the inclusion of ‘sick quitters’ (those people who stop drinking as a consequence of health problems) in control groups (6) and has formed the basis of refuting any conclusive evidence in a similar protective effect of alcohol on the development of dementia (7).

In summary, the ability of even the most robust research design to draw reliable conclusions from the health benefits of alcohol remain inconclusive. What is more, mortality from all causes has been shown to rise above levels of alcohol intake above 1.5 units per day. Although met with this some degree of trepidation, this upper limit was proposed for older people in Our Invisible Addicts, which is based not only on the above evidence but also on existing guidelines from the National Institute of Alcohol Abuse and Alcoholism in the United States (8).

Future studies will need to continue to navigate the complexities involved in exploring the relationship between alcohol and health, particularly if firm recommendations are to be made on the basis of robust methodology.


(1) Artaud F, Dugravot A, Singh-Manoux, Tzourio C, Elbaz A. Unhealthy behaviours and disability in older adults: Three-City Dijon cohort study. BMJ 2013;347:4240.
(2) Crome, I, Dar K, Janikiewicz S, Rao TR. Tarbuck A. Our Invisible Addicts (College Report CR165). London: Royal College of Psychiatrists 2011.
(3) Ronksley PE, Brien SE, Turner BJ, Mukamal KJ, Ghali WA. Association of alcohol consumption with selected cardiovascular disease outcomes: a systematic review and meta-analysis. BMJ 2011;342:671.
(4) Marmot M, & Brunner E. Alcohol and cardiovascular disease: the status of the U shaped curve. BMJ 1991;303:565-568.
(5) Roerecke M, & Rehm J. The cardio-protective association of average alcohol consumption and ischaemic heart disease: a systematic review and meta‐analysis. Addiction 2012;107:1246-1260
(6) Stockwell T. . A reply to Roerecke & Rehm: continuing questions about alcohol and health benefits. Addiction 2013;108:428-429.
(7) Roizen R, Fillmore K, Chikritzhs T, Stockwell T. Light-to-moderate drinking and dementia risk: The former drinkers problem re-visited. Addiction Research & Theory 2012;21:1-13.
(8) National Institute on Alcohol Abuse, & Alcoholism (US). Helping Patients Who Drink Too Much, A Clinician's Guide. Bethesda, Md: National Institue on Alcohol Abuse and Alcoholism 2007

Competing interests: No competing interests

23 August 2013
Rahul Rao
Visiting Researcher in Older People's Alcohol Misuse
MHOA Clinical Academic Group, Insitute of Psychiatry
De Crespigny Park, London SE5 8AF