Some intense activity on Twitter today discussing wine and health. Lots of views expressed, although proper discussion of topics like this in 140 characters is quite tricky.
The discussion was prompted by a tweet by Cam Haskell about a meta-analysis looking at the evidence for health benefits of antioxidants. It is here.
This is not a research study itself, but a critical analysis combining all the results from high quality studies to see if there is a general trend.
Most people are under the illusion that dietary antioxidants are good for health. This meta-analysis shows that vitamin A, beta-carotene and vitamin E may even have negative health effects. This isn’t new – those who have followed the literature for a while will have picked this up.
Many wine commentators are under the potential misconception that red wine is good for health because of its antioxidant properties. The evidence suggests that this isn’t the case.
Now there’s certainly good evidence that moderate drinkers live longer than teetotallers. It’s a consistent finding, and is known as the J-shaped curve. This is the shape of the line on the graph that results from plotting mortality against alcohol consumption.
As alcohol consumption increases, mortality decreases. Then, as alcohol consumption increases yet further, mortality starts to increase.
But there are many confounders here. People who tend to drink modestly also tend to exhibit other healthy behaviours. And people who drink wine tend to be richer, which correlates positively with health status.
Good studies control for such confounding factors. But not all studies are good. If you are trawling the scientific literature looking for evidence about wine and health, you need to be able to discriminate the good studies from the less good ones.
The antioxidant story resonates with people’s preconceptions. There are other components in wine that might make it healthful, not least among which is alcohol itself. Many studies show that there’s a health benefit with beer and other alcoholic drinks. But still, despite the evidence against any benefit from dietary antioxidants, that’s the story people cling to.
Do we need wine to be good for health, though? This is another discussion that was worked through on twitter. Most wine lovers drink wine because they enjoy it, not because it’s good for them. But the healthful properties of wine may be recruiting new drinkers, especially in Asia, it seems.
There was also discussion on the negative effects of wine. Alcohol can be harmful to health, and some were suggesting that it it gets off lightly compared with smoking and illegal drugs such as cannabis. Some wine trade figures suggested that if wine were devised de novo today, it would be banned.
I disagree. There is a healthy level of alcohol consumption, whereas with smoking there is no healthy level of consumption. The spectre of David Nutt was raised. Professor Nutt famously suggested that alcohol was more harmful than heroin. But heroin is a class A drug consumed by a few. Alcohol is almost universally consumed by adults.
I quite like the suggestion that wine is a virtuous intoxicant, as argued by Roger Scruton. Wine is quite unlike other mind-altering drugs which are dishonest in nature, because they claim to elevate the perception of the user such that the user enters a transcendental realm. These drugs lie to us because they tell us about another world outside our own. Instead, wine tells us about the true world; the one we live in, revealing more about it. Wine, when drunk in company, leads to an opening out of the self to others.
14 Comments on Wine and health
Well written thread Jamie — as someone who has consistently drunk far more than I am told is healthy, for most of my life, I am firmly of the opinion that wine that is good is good for you.
I actually did a count of my units of alcohol consumed in 2008 and made a note monthly on Wine_Pages. My total units were 3600 odd or the equivalent of roughly three quarters of a bottle of wine a day!! I never binge drink however and rarely have a day off!!
Now I also exercise every day and never eat junk food — factors which I think influence my health much more than my consumption of alcohol.
A life without the pleasure wine gives me, is one I do not wish to contemplate.
Good post with a clear exposition of the the points involved. I agree with the idea that while alcohol is a drug, and a toxic one at that, there is a safe level of consumption and, as far as I am aware, some positive health benefits to drinking wine. At a seminar in Austria a few years ago a Danish doctor presented evidence that for people over 50 a glass of wine a day actually made a real difference to their overall health.
I never fully understood the whole anti-oxidant idea as, even though I’m not a chemist or microbiologist, I don’t quite see how eating a tomato, say, means that any anti-oxidants in the tomato get through the digestive system unchanged! Maybe they do but, as I understand it, the stomach is a pretty hostile environment.
Anyway, a good, clear article outlining the reasons for enjoying wine – not only is it healthy in moderation but it is enjoyable!
I have to admit that the more wine I drink, the more I eat. Comparing quantity of food consumed during a meal without wine (very rare!) with what I can put away with it, is huge – this for me is the main health issue – loads of wine without loads of exercise makes me fat!!
As a doctor I regularly see the harm alcohol does to individuals, families and communities. As a drinker I see the benefits alcohol can bring too. To be a healthy person involves many factors. Some we can do nothing about (eg genetics), others we can (diet, smoking, exercise, alcohol etc). Risk of developing harms from alcohol increases with increasing consumption of alcohol on the wrong side of the J shaped curve. Some can drink excessively and be OK, some are just over the recommended levels and develop harms. I’m afraid there is a spectrum of risk rather than black and white cut off points. Unfortunately alcohol does make some people ill and kill them prematurely, so rather than having a polemic debate, may I suggest, as ever, moderation in everything, including excess.
Wine must be healthy – there’s a very attractive young girl on the ‘Fitness Singles’ tab on the right hand screen of my screen. With a 6pack like hers I can see she likes a little drop of a crisp Awatere Sauvignon now and again!
Where does the Roger Corder theory on wine, as published in “The Wine Diet” fit in to the antioxidant bit Jamie? You imply that there’s pretty much a see-saw effect in published works regarding wine and health, does that make the vinopic.com approach flawed in your opinion? I thought myself the ‘intrinsic quotient’ rating, for obvious reasons, favoured reds over whites for no real scientific basis.
“But the healthful properties of wine may be recruiting new drinkers” isn’t that really the crux of the matter?.
As an aside, ask any Emergency Physician if they have three patients before them: one withdrawing from Heroin, one withdrawing from Cocaine and one withdrawing from Ethanol, which one they will tend to first. Perhaps that is one reason for Nutt’s assertion.
Important subject to bring up and very balanced discussion, I have been an assiduous, silent (until now) follower of your blog, but I would just like to make a comment on this subject. Thanks to Damien I looked up the vinopic website – very interesting – Roger Corder’s studies on red wine polyphenols and their specific physiological effects on humans are well known and a really important contribution. Even though I do not go for the grading of wines according to their health promoting benefits, it is significant that, as far as I can see, there is no reference to “antioxidants” in the arguments which are presented on the site. Polyphenols have been shown to have a number of positive effects on humans which are not via them being antioxidant. This might look like nitpicking, but millions is being spent on looking for “antioxidants” in the diet – including wine, when there really isn’t much credible data showing that this is relevant – even more so for the water-soluble ones, such as would be found in wines.
I was at an event where Roger Corder and other physicians spoke about wine antioxidants, the importance of dietary polyphenols for optimal health and the metabolic, physiologic and psychological effects of alcohol. See the article here: http://enobytes.com/2012/01/06/411-wine-health/
Agreed, there is little evidence that wine polyphenols influence health through an antioxidant action. However, this doesn’t exclude the possibility that specific wine polyphenols confer health benefits due to other molecular mechanisms. Research on dietary polyphenols covers a vast range of different naturally occurring plant chemicals. Although in recent years resveratrol has often been the focus of many media reports about wine and health this really is very misleading because most wines have little or no resveratrol, so it’s irrelevant to this debate. The best evidence that polyphenols can confer a health benefit is from products that have a high content of a specific class of flavonoids called flavanols (which includes procyanidins). Placebo-controlled clinical trials show improvements in vascular function and blood pressure with doses of 500 mg to 750 mg per day. These benefits can be seen even in patients treated optimally with prescription medicines, which highlights how important such actions could be to the overall health of people of all ages whatever their dietary source. http://content.onlinejacc.org/cgi/content/abstract/56/3/218
Grape pips are one source of flavanols/procyanidins, but achieving a daily intake at this level (500 – 750 mg/day) through red wine alone (or orange natural wines) would be difficult (without exceeding moderate alcohol consumption) except for the most extracted tannic red wines. Despite this proviso (and other confounders), it is unduly hasty to attribute the health benefits of regular moderate red wine drinking exclusively to alcohol. My ecological studies found a very clear association between areas in southwest France with the lowest heart disease (and increased longevity) with the production (and drinking) of some of the most flavanol/procyanidin rich red wines that can be found. Studies of alcohol consumption across France have shown average consumption is pretty consistent yet before heart disease was managed more intensively with prescription medicines there were marked regional differences in mortality from heart disease. Traditionally what people drank depended a lot on local habits, which were frequently orientated to regional wines (or beers). So it is important to note that Alsace, as a region where white wine and beer were more frequently consumed, typically had some of the highest levels of heart disease.
As indicate by Jamie, the debate as to whether red wine confers an extra health benefit over other alcoholic drinks has many confounders. Too few studies have examined the incidence of heart disease in people who drink mainly red wine versus other alcoholic drinks. But an important additional confounder that is ignored in studies comparing drinking habits and risk of disease is that even those who drink mainly red wine might not drink flavanol-rich wines, so the data are always likely to be inconclusive. Also, unfined/unfiltered beer made with heavy use of hops can be rich in flavanols/procyanidins, as can traditional ciders, yet these confounders are always ignored in the epidemiological debate.
Despite all the wine and health brouhaha, there is another reason why I like to know the polyphenol levels in wine. I’m sure there are many others with a similar taste in red wine to me. A key thing I discovered on my wine journey is that polyphenol-rich wines are generally the most enjoyable wines particularly with food. Without high levels of polyphenols it is almost impossible to achieve the harmonious complexity of the best red wines. This has certainly been a key factor in terms of establishing the reputation of the great Bordeaux wines over many generations, see: http://www.jancisrobinson.com/articles/winenews050416.html
As an advocate for drinking less but better quality, knowing that a wine has a polyphenol structure due to grape tannins (procyanidins) rather than oak tannins brings you one step closer to finding a great wine every time. Many wines described as full-bodied with firm tannins are just high alcohol wines aged in oak, which I like to avoid. Similarly, many easy-drinking supermarket wines are nothing more than highly alcoholic fruit juices with no extracted grape pip tannins. So if you favour the type of red wine that I do, understanding more about the phenolic structure of a wine can provide the type of insight that enables an informed choice when selecting wines to appreciate and enjoy.
Great post Jamie (as usual), and some great responses. What irks me is that as the ‘wine & health’ stuff can so often be blithely trotted out, much like an awful lot of nutritional (mis)information. That world is so often the preserve of quacks and hucksters – it pains me to see the wine world drifting toward it in pursuit of bucks. As always, consumers deserve clarity and honesty.
As Roger points out, polyphenols are complex things. A question therein: how viable is it to assess those polyphenol levels as Roger points out, given that they’re not static, as the level of polyphenols will change, most particularly as wine ages and tannins polymerise.
Similarly, how easy is it to distinguish grape tannin from the added tannin that is routinely added, vs ellagitanins?
An additional note relates to those who consume moderate amounts of alcohol vs. those who are teetotal. I am inclined to believe that the teetotal figures are skewed by those who don’t drink. Frequently the teetotal numbers can include people suffering significant health problems already, and those who used consume alcohol to great excess, but no longer do. Ideally studies would only include healthy people in each subset. But this is not especially viable when analysing data from large swathes of the population.
Finally, the comparison with smoking is daft. Wine consumption does not necessarily cause one to consume more wine – for alcoholics, perhaps it does. Nicotine does, however, prompt the consumption of more nicotine. And yet for me, Barolo and Burgundy always leads to the pursuit of more Barolo and Burgundy… Where’s the metastudy on that!?!
I have a couple more pertinent comments to add to this discussion, particularly in the light of today’s UK parliamentary Science and Technology Committee recommendation that we should avoid drinking alcohol on at least two days per week. Firstly, in response to Cam’s comments, most recent epidemiological studies of alcohol have excluded the sick quitters (i.e. the ex-drinkers who gave up for health reasons) when comparing non-drinkers with drinkers, and have still concluded that from a heart disease perspective it is better to be a moderate drinker than teetotal.
It is very important that all health advice (nutritional or otherwise) is evidence based. So, concerning safe drinking, there’s no evidence to suggest that 20 g of alcohol per day for a man (2.5 UK units) causes any harm, which casts doubt on the need for two alcohol free days per week. However, getting men to accept that this is moderate drinking is the challenge because that’s only one 175 ml glass of wine (14% ABV), or one pint of beer (4.4% ABV). As far as the relationship between risk of cancer and alcohol is concerned, men drinking moderately have little risk if they are non-smokers and follow a healthy diet. For women the relationship is more complex depending on their age, diet and family history of breast cancer.
Regarding polyphenol analyses of red wine, there is reasonable chromatographic resolution between oak (ellagitannins) and grape tannins (procyanidins); however, it is impossible to separate added grape tannins (e.g. grape seed extract) or grape colour (rubired anthocyanin extracts) from naturally occurring grape flavonoids. So I’m very much in favour of back labels on wine bottles that include a list of all additives used in winemaking.
Polyphenol levels in wine do vary with age, but the rate at which polymerisation occurs is still poorly understood. From personal experience, if you keep wines for 4 to 5 years after vintage some wines will throw a large deposit but many wines with similar initial polyphenol levels may show no sign of a deposit. Nevertheless, analysing polyphenols in red wine can provide a helpful perspective for predicting its potential 3 to 4 years after vintage, and also for its evolution into something really special over the next few years. If you want to be a moderate drinker it is important to know that your chosen glass of a wine is going to be something really pleasurable!
yes, statistics can be made to show just about anything you want, so I’ always skeptical about correlations between health (or lack of) and any given product or substance. Imho, there are so many factors of ones lifestyle that affect your health, that wine consumption is just one of many and so it cant be that significant or important. Intersting stuff, though. (the discussion AND the wine!!!)
“Wine is quite unlike other mind-altering drugs which are dishonest in nature, because they claim to elevate the perception of the user such that the user enters a transcendental realm. These drugs lie to us because they tell us about another world outside our own. Instead, wine tells us about the true world; the one we live in, revealing more about it. Wine, when drunk in company, leads to an opening out of the self to others.”
Wine effects your judgement just as much as, and is as destructive as any alcoholic drink if not used sensibly – don’t kid yourself it’s special just because you enjoy it.