Wine and Health: Challenging the 'No Safe Level' Claims

The background

In January 2023, the World Health Organisation (WHO) declared that there is “no safe level of alcohol” and that alcohol causes at least seven different types of cancer. The WHO also aimed squarely at the claim that drinking in moderation has health benefits:

“Moreover, there are no studies that would demonstrate that the potential beneficial effects of light and moderate drinking on cardiovascular diseases and type 2 diabetes outweigh the cancer risk associated with these same levels of alcohol consumption for individual consumers.”

Or to put it another way: while moderate drinking may have some cardiovascular protective effects, drinkers may pay for it with an increased cancer risk.

Prof. Nicolai Worm, Chair of the Wine Information Council, says the WHO recommendations are based on work done by the Global Burden of Disease collaborators, published in 2018. “They came to the conclusion that only zero alcohol is safe,” he said.

And yet, he said, in 2020 they used exactly the same data to “come to a different conclusion”, which is that the J-curve really exists; the J-curve is a graphical representation of the effects of alcohol, that suggests that light-to-moderate levels of drinking have a health-protective effect — particularly on cardiovascular disease, the world's leading killer.

But the WHO’s recommendations didn’t change.
 

What is moderate alcohol consumption?

Prof Ken Mukamal from the Harvard Medical School defines a drink as “something between 11 and 15 grams of alcohol.”

“I’m not going to spend a lot of time talking about the clear array of effects of heavy alcohol consumption,” he said. “Heavy alcohol consumption has been related to cardiomyopathy, stroke, hypertension, cirrhosis, pancreatitis, cancers, falls, violence and dementia.”

He said, “alcohol consumption should not exceed two drinks per day". 

Yet he also said that although “alcohol consumption across the board has dropped globally” —  except in India and China — health authorities have been recommending ever lower amounts of alcohol.

Prof Mukamal said one reason is that the adoption of electronic health records has made it easier for researchers to analyse much larger data sets than in the past and come to what seem to be robust conclusions — but that bigger studies were not only not more reliable, but could actually result in bigger errors if founded on faulty reasoning.

For example, studies based on weekly consumption are inherently flawed, because someone who drinks two drinks a day will have a very different health outcome to someone who drinks 14 drinks over one weekend.

“Just because studies are bigger doesn’t make them better,” he concluded. “We should be measuring multiple domains of drinking that include frequency, quantity and binge drinking. We need to capture drinking with meals and the beverage type.”

The French Paradox revisited

Next up was a star of the wine science world: Prof Curtis Ellison, who shot to fame in 1991 when he was featured in the famous CBS TV 60 Minutes French Paradox story, which suggested that the French have lower levels of heart disease than Americans because they drink red wine. 

The next day, red wine sales soared. Within a year, they were up 40% — a boom that helped to more than double the number of Californian wineries in less than a decade.

Prof Ellison came out swinging in favour of wine, saying that his “interest in the topic began in the 1980s with the Framingham Study,” which has been medically tracking residents of Framingham, Massachusetts, and their descendants since 1948. The study showed something unexpected: smokers who drank moderately had only half the mortality of smokers who didn’t. 

“Now, all papers from the Framingham Study — which is funded by the National Institutes of Health as a special project — have to be approved,” he went on. “When this manuscript was sent to Washington to get approved, they got this response: ‘You are not to submit this manuscript. Refer to only the major risk factors and remove all references to alcohol’. This is political science at its worst.”

He said that many studies over the past 50 years had shown that “moderate drinkers have significantly lower mortality. Why the persistence of the anti-alcohol sentiment?”

Prof Ellison answered his own question by saying that because governments have had to deal with the fallout from alcohol abuse, talking about potential health benefits means mixed messaging. Saying “don’t drink” is just easier. But, he went on, “health policy should not be based on paternalism,” but on evidence.

The second problem, he said, was too many alcohol studies combine different populations in inappropriate ways. Young people who get no benefit from alcohol are mixed in with older people who do, and then binge drinkers are added in on top, leading to false conclusions. “You don’t mix data from young people, and some of the studies continue to do that, including the Global Burden of Disease,” he said.

Another issue is that when meta-analyses are done (where multiple studies are considered together) “the authors are very selective” and only include studies that support their premise.

There’s also the problem of extrapolating beyond available data, and of “tricking people with various statistical manoeuvres”, plus the problem of how to correct for patients under-reporting their alcohol intake. Prof Ellison discussed one study from northern California whose research subjects claimed to be drinking just one to two drinks a day — but their medical records later showed “they had a history of alcoholism and hospitalisation for intoxication”.

And, finally, there’s the problem that researchers exclude patients with diseases like diabetes, so they won’t complicate whatever is being studied. But, said Prof Ellison, small amounts of alcohol could potentially have an “anti-diabetic” effect, “so you’re throwing out the baby with the bath water”.

“The scientific data clearly indicate there is a J-curve,” he concluded. “The dangers of moderate drinking are being exaggerated”.
 

The wine research problem

One problem in alcohol research is that people can’t be observed drinking, so researchers have to trust their claims. But are people really drinking as moderately as they say?

The only way to know for sure would be to conduct a long-term randomised control trial, where people are randomly sorted into different groups and then followed. Prof Mukamal was to lead such a trial, done with the American National Institute of Health, and funded by the alcohol industry to the tune of $100 million; however, after an outcry about the industry’s involvement, it was shut down.

Animals, however, can’t lie about their drinking problems.

Prof Mladen Boban, head of the Department of Basic and Clinical Pharmacology at the University of Split Medical School, Croatia and his team subjected rats to surgically-induced myocardial infarction (heart attack), so they could look at the impact of wine, if any, on the healing process. One group of rats was served a Croatian white wine, vintage 2015, while others were stuck drinking water.

Some rats, like humans, will hit the bottle too hard if they have the chance. The ones who overindulged — drinking double the booze of the moderate rats — died. Twelve of the 23 water-drinking rats also died. The moderate drinkers, however, showed signs of anti-inflammatory markers, and, Prof Boban told the audience, “There was no death observed in the mid-range consumption”.

The J-curve isn’t one-size-fits-all

Professor Giovanni de Gaetano, Head of the Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo Neuromed Pozzilli, says that the J-curve isn’t as simple as it looks.

“We don’t see the J-curve everywhere,” he said, adding it exists for myocardial infarction but not for atrial fibrillation (irregular heart rhythm). “We showed that there were differences between women and men. The effect of higher doses was very different, so the beneficial effect in women disappears more quickly.”

Age is also important: a 40-year-old woman with no cardiovascular risk has nothing to gain from moderate drinking.

There are also national differences. “The J-curve is everywhere, but French and Italian people can drink more safely because the beneficial effect remains longer than in Germany or the UK or in Sweden. But even in countries where moderation is not so frequent we can observe a small but significant J-curve.” This means that blanket statements made by international bodies like the WHO “are not valid for everybody”.

Prof Gaetano said that the authors of the Global Burden of Disease were to be commended because they re-evaluated their own data, and rediscovered the J-curve.

Finally, he said, “People who drink with moderation and regularity may continue to do it with taste, pleasure and, especially, culture.”
 

Wine and cancer

While the discussions on wine and heart health were interesting, they didn’t address the central plank of the new zero-alcohol messaging: cancer.  In January 2023, the WHO issued a statement saying that alcohol is a Group 1 carcinogen — like asbestos or tobacco — that causes seven types of cancer.

“The risk of developing cancer increases substantially the more alcohol is consumed,” the statement went on. “However, latest available data indicate that half of all alcohol-attributable cancers in the WHO European Region are caused by ‘light’ and ‘moderate’ alcohol consumption.”

“I’m getting annoyed a little bit with the current discussion and the lack of scientific evidence about the effects of alcohol and cancer,” he said. “It reminds me a little bit of the Victorian times.”

First, he said, only 4% of cancers in the world, or 740,000 annually, are caused by alcohol. “Of these, the vast majority were caused by heavy drinking. It is only heavy drinking that stands out for the most common cancers.”

Dr Apffelstaedt also said that there was an “inverse relationship for kidney and thyroid cancer and lymphomas, but I wouldn’t recommend trying it for a cure”.

He then presented a Kaiser Permanente study of 1,897 breast cancer survivors, of whom half consumed alcohol. The study concluded that drinking even a small amount of alcohol could lead to a recurrence of breast cancer.

“But if you dig down into the data, this risk of recurrence was limited to obese and mostly post-menopausal women,” he said. “And if you drill down into the data, you actually see that the cardiovascular mortality was reduced in the wine drinkers in this particular study.”

Dr Apffelstaedt noted that many chemotherapy drugs are extremely harmful to the heart, “which may be ameliorated by moderate alcohol use”.

His final conclusion was that when it comes to moderate alcohol consumption, “it is my bet that it doesn’t influence cancer mortality by much” and that “I do not want to deprive my patients of the enjoyment of a glass of wine”.
 

What happens next?

Much of the Congress was given over to discussing the integral role that wine plays in the Mediterranean diet. Based on the eating habits of people in southern Europe, the Mediterranean diet has long been understood as the best diet for heart health.

There were many presentations on the way that the polyphenols in dietary elements like red wine and olive oil interact to lower inflammation and improve insulin sensitivity.

Prof Ramón Estruch, Professor of Medicine in the School of Medicine at the University of Barcelona, said it was important to understand the context of how wine is drunk, because alcohol with food has a different impact than alcohol consumed by itself. “You can see that if you drink wine and follow the Mediterranean diet,” the rate of cancer falls.

It seemed from the conversations happening in the breaks that many wine executives were taking comfort from the presentations, hoping that the Congress was the start of a wider pushback against the WHO’s position.

Unfortunately, it was also clear that the wine industry will struggle to defend itself from medical attacks because the evidence can’t be boiled down to the kind of simple message that's needed for effective political lobbying. Regular moderate consumption is clearly beneficial for some people, depending on their age, their sex, and even their ethnicity, but for others, particularly young women at risk of breast cancer, that same consumption could be risky. It will be extremely difficult for the wine industry to communicate this complex information effectively, much less convince the general public that major health studies, published in highly credible journals like The Lancet, are flawed.

"We have a culture built on the Mediterranean diet, and we want to defend our way of life," said D. Fernando Miranda, Spain's General Secretary of Agriculture and Food at the opening of the Congress. "Wine has a major economic importance for our country." 

Or to put it another way: a $333 billion industry is now at risk. Those executives are right to be worried.