More or Less Behind the Statistics
More or Less Behind the Statistics

WSMoreOrLess: Safe drinking

New alcohol guidelines were issued recently in the UK which lowered the number of units recommended for safe drinking. But are the benefits and harms of alcohol being judged correctly? We speak to Professor David Speigelhalter. Tim Harford presents. Producer: Charlotte McDonald/Richard Vadon

Featured Speakers

BBC Host

Topics Discussed

Episode Summary

Executive Summary: The episode examines the evidence behind UK alcohol guidelines, focusing on why the new 14-unit weekly limit applies equally to men and women. It explains that alcohol may offer small cardiovascular benefits at very low levels but also raises risks of cancer, liver disease, and injury, while research is complicated by biased self-reporting and confounding among non-drinkers. The piece argues that the overall risk at guideline levels is low compared with other health risks, but the evidence is messy and hard to translate into exact safe thresholds.

Main Topics: New UK alcohol guidelines (Priority: 5/5): The programme explains the updated government advice that both men and women should drink no more than 14 units per week, and notes how unusual it is for the same limit to apply to both sexes. Alcohol’s mixed effects on health (Priority: 5/5): Experts discuss the tension between possible protective effects on ischemic heart disease and known harms including breast cancer, liver disease, and alcohol-related injuries. Problems with alcohol research (Priority: 5/5): The episode highlights why comparing drinkers and non-drinkers is difficult, because abstainers are not a random group and self-reported drinking often underestimates actual consumption. Limits of precision in the guidelines (Priority: 4/5): The discussion shows that guideline thresholds are based on imperfect self-report data, making very fine distinctions such as one versus two units per day hard to justify confidently. Risk communication and public health messaging (Priority: 4/5): The segment explores why the protective-heart narrative is scrutinized heavily and how public health authorities may prefer simple anti-drinking messages even when evidence is nuanced. Contextualizing alcohol risk against other behaviors (Priority: 4/5): Risk expert Sir David Spiegelhalter compares alcohol risk at guideline levels with smoking and sedentary lifestyle, suggesting the relative risk is low in broader health terms.

Key Arguments: Alcohol can have some protective effect on ischemic heart disease, but only at very low levels and not as a blanket health benefit. The harms of alcohol are broad and serious, including cancers, liver disease, and injuries, so overall net benefit is limited. Non-drinkers are an unreliable comparison group because many are former heavy drinkers, religious abstainers, or people constrained by poverty or health. Survey data underestimates real alcohol intake; people often report only about 30% of what they actually consume in aggregate comparisons. Because self-reports are inaccurate, research-based thresholds such as 'one unit a day' cannot be interpreted with great precision. The most disputed part of alcohol epidemiology is the claim of cardiovascular benefit, which receives more scrutiny than other alcohol harms. At the new guideline level, lifetime risk of an alcohol-related condition is presented as about 1 in 100, far lower than smoking-related mortality. Compared with sedentary behavior and poor diet, drinking within the guideline appears to be a relatively low health risk. The young do not typically need cardiovascular protection, which is one reason the heart-benefit argument is less relevant for them.

Data Points: UK recommended weekly alcohol limit: 14 units - New guidance applies equally to men and women Men’s new weekly limit: 14 units - Same as women under updated UK guidance Women’s new weekly limit: 14 units - Same as men under updated UK guidance Approximate equivalent: 6 pints of average-strength beer - Used to explain 14 units per week Approximate equivalent: 3.5 litres of average-strength beer - Alternative equivalent for 14 units per week Approximate equivalent: 7 medium-sized glasses of wine - Alternative equivalent for 14 units per week French guideline for men: 26 units/week - Example of higher male limit in Europe French guideline for women: 18 units/week - Example of higher female limit in Europe Irish guideline for men: 26 units/week - Example of higher male limit in Europe Irish guideline for women: 18 units/week - Example of higher female limit in Europe Protective drinking in older women: Up to 5 units/week - Sally Davies said a small group of women over 55 may see some protection

Pivotal Quotes: "A glass of alcohol, and it's not only red wine, has protective effects on the ischemic heart disease and on some other ischemic diseases." — Dr. Jürgen Rehm: Explaining why the heart-protection idea is not merely an old wives' tale "The desire by some to just put across a clear public health message that drinking is bad for you with no complicated caveats." — Dr. Jürgen Rehm: Why the cardiovascular benefits of alcohol may be especially scrutinized "They pegged this guideline at one in a hundred chance of essentially of it killing you." — Sir David Spiegelhalter: Comparing alcohol risk at the new guideline level to other everyday risks

Implications: For listeners, the message is that alcohol at guideline levels is relatively low risk but not risk-free, and exact safe thresholds are uncertain. For policymakers, the episode suggests evidence-based guidelines must balance nuance, imperfect data, and clear public-health messaging.

🔓 Sign Up for Unlimited Episode Search

About More or Less Behind the Statistics

Tim Harford and the More or Less team try to make sense of the statistics which surround us. From BBC Radio 4

View all episodes from More or Less Behind the Statistics