Wine and Health Update

For the past 20 years, the data on moderate wine consumption and health has been largely positive. Studies have tended to show that moderate red wine consumption (meaning 1 glass of wine per day for women and 2 glasses for men) has positive effects on mental health, heart health, brain function and resistance to some cancers. As with most studies limited in sample size and with difficulty controlling for all variables, it’s appropriate to be cautious about drawing conclusions. But the connection between wine and health seemed reasonably well established.

Then this past winter studies reported by the UK Department of Health seemed to call all of that into question.

New guidelines for alcohol consumption, produced by the UK Chief Medical Officers, warn that drinking any level of alcohol increases the risk of a range of cancers. This is supported by a new review from the Committee on Carcinogenicity (CoC) on alcohol and cancer risk .

It is now known that the risks start from any level of regular drinking and increase with the amount being drunk, and the new guidelines are aimed at keeping the risk of mortality from cancers or other diseases low. The links between alcohol and cancer were not fully understood in the original guidelines, which came out in 1995.

This review also found that the benefits of alcohol for heart health only apply for women aged 55 and over. The greatest benefit is seen when these women limit their intake to around 5 units a week, the equivalent of around 2 standard glasses of wine. The group concluded that there is no justification for drinking for health reasons….

Dame Sally Davies, Chief Medical Officer for England, said: Drinking any level of alcohol regularly carries a health risk for anyone, but if men and women limit their intake to no more than 14 units a week it keeps the risk of illness like cancer and liver disease low.

This study and the policy prescriptions from the UK received an enormous amount of press. So what’s going on here?

At the Society of Wine Educators Conference earlier this month Matilde Parente (MD, FCAP) gave a presentation that was very skeptical of these studies that the Brits rely on for their policy. Her key point was that the Royal Statistical Society, a professional organization for statisticians and data analysts, thoroughly shredded the argument put forth by the UK Department of Health. A fact that was not widely reported in the press.

In spite of the risks at the revised guidelines being acknowledged as minimal, the communications have also strongly emphasised potential harms of low-level consumption, particularly cancer. This appears to have been built into the commissioned analysis from the start, and it could be argued that the main change from the previous guidelines has been due to a statistically-unjustified assumption imposed by Public Health England1. There has been a continued emphasis on inducing fear through mentions of cancer, and consistent downplaying and even denial of any health benefit – the Press release says that “the protective effect of alcohol against heart disease has now been shown not to apply to men”, which directly contradicts the estimates published in Table 10 of the Guidelines Development Group report (1).

The Royal Statistical Society is not a group of bomb-throwers who toss charges of corruption around lightly. They really, really don’t like these recommendations. Their problem with the UK policy really comes down to three claims:

1. The UK Department of Health has no statistical justification for their recommendations. The increased risk of cancer was assumed from the beginning; not derived from any new studies. The evidence for this is that the tables provided in their report in fact show that “the threshold for males to reach a 1% lifetime risk would have been 21 rather than 14 units, exactly the previous Guideline.”

2. The report assumes a linear relationship between alcohol assumption and risk. “This again serves to increase the apparent harm of low levels of consumption by forcing the statistically unjustified assumption that the number of acute deaths arising from a change in peak daily consumption from, say, 0 to 2 units would be the same as a change from, say, 20 to 22 units.”

3. And perhaps most importantly, the health officials did not consider the well established benefits of alcohol when weighing the risk factors. In other words, instead of weighing the risk of certain cancers against the benefits of alcohol consumption, they simply ignored the benefits

The Royal Statistical Society argues instead that a 1% increase in risk for certain cancers with the consumption of 14 glasses of wine per week for men and women, which is what the data shows, is “broadly acceptable”. However, the risk does increase dramatically with higher consumption levels.

In short, the worthies at the UK Department of Health are anti-alcohol zealots who use their position to push their agenda. In fairness, the UK Department of Health is a public agency charged with protecting public health. And excess alcohol consumption is a huge public health risk. They have a difficult job trying to tame the beast. But public health is not advanced by giving misleading information in an attempt to shock the public into compliance with their puritanical views.

There is surely a lot we don’t know about the relationship between alcohol and health. It may turn out that the positive relationship between wine and health is illusory. But nothing in this report by the UK Department of Health demonstrates that.

And just as an addendum, I should add that this whole controversy brings to light a constant failing on the part of the press who report medical findings. When you read that something increases risk of mortality by, for instance, 20%, it does not mean that 1 in 5 people will die from it. The increased risk is from a base line. If your risk of dying from a malady is for instance 2%, an increase of 20% means your risk increases to 2.4%. This is seldom included in reports and is part of the ignorance of statistics against which the Royal Statistical Society does battle.