Moderate Drinking and Health: Revisiting the ‘French Paradox’
Americans spend a lot of time thinking about ways to improve our health. Diet, exercise, even a recent foray into the potential of “polypills” — compounds which mimic the desired effects of nutrition and lifestyle interventions but come with none of the hassle, are evidence that the desire to boost wellness is ubiquitous.
A new fascination with the health and longevity benefits of moderate alcohol consumption is something I have been taken notice of and become, for lack of a better word, fascinated with. (This isn’t particularly new, as the “French Paradox — discussed later — was observed around the 1980's). I say fascinated because, initially, it made no sense. A link between drinking alcohol and BETTER health seemed, well, paradoxical. Alcohol is one of the leading causes of preventable death in the US and worldwide. In addition, it seems obvious, anecdotally, that alcohol intake is generally associated with other negative lifestyle behaviors (less sleep, poorer diet, fewer exercise habits, etc).
Nonetheless, low to moderate alcohol intake, as a result of conclusions made in observational studies, is almost being “prescribed” by some to enhance health.
The J-Curve
Where did alcohol begin to gain its reputation as a potential mediator — no less an enhancer — of health?
Multiple epidemiological studies have confirmed the observation of a J-curve shaped association between alcohol intake and adverse health outcomes: coronary heart disease, diabetes, hypertension, chronic heart failure, stroke, dementia, and all-cause mortality. What this means is that, for light to moderate consumption of alcohol (1 drink/day for women and 1–2 drinks/day for men) there exists a “cardioprotective” benefit, whereby increasing one’s consumption above this “threshold” leads to worse health outcomes; total abstinence is associated with greater risk than moderate or low consumption of alcohol.
Above a certain threshold, risk for all-cause mortality increases, suggesting there is some sort of alcohol “sweet spot” where consumption improves health. In this way, researchers have described alcohol as a proverbial “double-edged sword”.
There is speculation as to why this might be; mechanistic speculation, that is. In a review on alcohol and CV health, O’Keefe et al. cite that “ethanol, rather than the specific components of various beverages, appears to be the main factor conferring health benefits.” Indeed, some evidence shows that light to moderate alcohol consumption may beneficially impact insulin sensitivity, elevate HDL cholesterol, reduce inflammation, and even improve endothelial function. All of the above are known risk factors for cardiovascular disease and thus, suggest a causal role for alcohol in primary CVD prevention. In this way, paradoxically, ethanol is “the primary factor for both conferring health benefits and (emphasis mine) causing toxicity.” Toxic or healthful — that is for the individual to decide. Consumption quantity and patterns play a key role here.
A few caveats with epidemiology studies of this kind. Most, if not all of the studies cited in the literature rely on self-reported alcohol consumption. That is, individuals are asked or interviewed on their alcohol consumption patterns. Just like dietary recall, we should assume self-reported alcohol intake is equally as unreliable. Additionally, and here comes the oft-repeated line to “discredit” epidemiology: correlation does not imply causation. It has become a cliche, but rightfully so. All we know is that those people who have a few drinks per week have lower risks of death due to cardiovascular and all-causes — nothing more.
The origins of this alcohol story have roots in two other phenomena: one observational and the other, experimental.
The French Paradox and Sinclair’s Mice
In the 1992 French epidemiologists (epidemiology becoming a motif in our story) coined a term to describe a phenomenon they observed in their home country. This phenomenon, which we now know as the “French Paradox”, related to the fact that the French tended to have a very low incidence of cardiovascular disease despite a diet which was relatively high in saturated fat and cholesterol (the French eat a lot of cheese and butter, apparently).
Detour. “French Paradox” isn’t really a paradox — for a paradox is only a paradox if it can’t be explained. In the case of the “French Paradox”, it relies on the foundation of the diet-heart hypothesis; that saturated fat and cholesterol cause heart disease. This is why it was such a paradox (at the time). The French eat so much heart-disease causing fatty foods, so why don’t they have high levels of CVD compared to other countries? Something must explain this. But…
We now know the diet-heart hypothesis to be basically invalid, at this point. It is unlikely that dietary intake of fat, cholesterol, animal foods, etc. are the cause of most preventable diseases. Our paradox falls apart when analyzed after the fact.
The then-cited reason for the lower rates of CVD among the lipid-centric French was that perhaps their high consumption of red wine, thought then and now to be healthful, was protecting them from developing disease — in spite of their diet.
Indeed, an investigation at the time into lifestyle factors of southern European countries confirmed that wine (red, in particular) was one of the common dietary elements. But why was red wine so beneficial? Researchers acknowledge that the paradox is nothing if not provable and testable. “The most difficult issue is not the description, but the explanation.”
Enter David Sinclair.
If it was true that the wine that explained the “French Paradox” (other countries reporting high levels of wine consumption similarly had higher levels of longevity and disease-free aging), then it seems intuitive that something in the wine must be providing these benefits. The one who isolates this compound, no doubt, will be master of immortality, take hold over death…or market lifespan extension to the masses.
In 2003, Harvard researcher David Sinclair and his group sought to investigate and hopefully discover potential compounds that might modulate lifespan. These compounds, sometimes referred to as “calorie restriction mimetics” are so named because they mimic the effects of one of the only reproducible ways to extend lifespan in a variety of species — chronic caloric restriction. Anything resembling such a mimetic is VERY desirable in the field of aging because, lets be honest, nobody wants to eat 40% less food for the remainder of their 200 year lifespan.
And discover they did. In particular, a compound known as resveratrol, found in many plant species (and, apt to our story, also in red wine) was shown by the group to extend life span in mice by 24% and in species such as flies and fish by as much as 59%.
Sinclair’s lab wasn’t the first to investigate resveratrol. This molecule was first isolated from roots of white hellebore in 1940 and from the plant polygonum cusidatum (used in traditional Chinese medicine) in 1963. It gained fame in 1992 (see: French Paradox).
Sinclair seems the perfect longevity hero on our quest for immortality. Indeed, in his TED talk, he spews certainty and knowledge, hope for the future. His research unsurprisingly exploded in the media and elsewhere. In a NY times article, Sinclair reveals both his drive and his excitement.
“As soon as I realized I was mortal, I started to worry. I set a goal to see if we could make drugs that would target the diseases of aging…I didn’t know it would be possible at all — and I didn't know it would happen so quickly.”
The result? Sinclar, along with an investor, founded Sirtris in 2004, a company founded to commercialize Sinclair’s research and dive head-first into research on specific pharmaceutical compounds that target a well-known aging pathway involving Sirtuin proteins.
Tl;dr: Sirtuins are biologically important molecules regulated by stress as well as nutrient status of the body. Supposedly, they explain many of the anti-aging benefits of caloric restriction, and may also be activated by resveratrol.
And indeed, whether through Sirtuins or voodoo magic, resveratrol has been shown to induce a kaleidoscope of biological changes and improve health. Among the various benefits: reduction in cancer incidence and tumor progression, decreased aggregation of blood platelets, lowering of blood pressure, prevention of atherosclerosis, improved serum cholesterol, prevention/reduction in heart disease, increased antioxidant activity, suppression of inflammation, enhanced immune responses, and neuroprotection. Many of these changes have been shown in mammals (mice, but nonetheless…).
Complexity arises when we begin to talk about how much resveratrol one would actually need to see the above changes. A daily intake of 2 glasses of wine would give a 70 kg individual a dose of 27 ng/kg body weight (this is very low). In contrast, some of the highest (but still non side-effect causing) doses used in experiments with mice — 100mg/kg body weight — equate to around 10,000 bottles of wine. How’s that to drinking for your health.
GlaxoSmithKline (GSK) purchased Sirtris in 2008 for $720 million. The corporation has since closed Sirtris (2013) , but still hopes to move clinical research assets into practice. Whatever that means.
Drinking thresholds, and and the “social factor”
A recent article in the Lancet challenged the current “drinking thresholds” that are established to advise on the number of drinks/week that denote the lowest “risk” (of heart disease, all-cause mortality, etc.).
This was a big, and important study. For once, it seems, a group wanted to take the current “dogma” (current drinking thresholds are the best available) and provide evidence to the contrary — possibly counter the argument once and for all that “moderate alcohol consumption is beneficial”.
The group looked at 83 long-term prospective studies in 19 high-income contries, characterizing the drinking threshold for all-cause mortality and a variety of other known-diseases in current alcohol drinkers.
What did they find? For all-cause mortality (that is, death from any cause), alcohol consumption showed a positive and curvilinear association — with the lowest risk appearing at a drinking threshold of 100g alcohol per week (this was the lowest intake they could report). This risk threshold is lower than the current US guideline of 196 g/week (for men) and right around the recommended intake for women, which is 98g.
Findings on life expectancy proved to be equally interesting. Those drinking less than the “threshold” of 100g/week had a longer life expectancy at age 40 (i.e. more “years remaining) than those drinking above the threshold — and this increased linearly with dose. Drinking 100–200, 200–350, and above 350g/week of alcohol reduced life expectancy at age 40 of .5, 1–2, and 4–5 years, respectively.
Put another way: Men who reported drinking 196g of alcohol per week (current US threshold for low-risk) had about 1–2 years LOWER life expectancy at age 40 than did men drinking less than 100g per week.
These data put into perspective the cumulative effect of excessive alcohol consumption, and the fact that maybe one drink per day might not be acutely detrimental, but over time, might accumulate some negative health benefits. Indeed, “data support adoption of lower limits of alcohol consumption than are recommended in most current guidelines”
In here lies my beef with the current paradigm that moderate alcohol consumption might not be detrimental, but in fact beneficial. We know, based on experimental evidence, that ethanol is hepatotoxic (harms the liver) as well as neurotoxic in certain doses in conditions. Long term alcohol use increases the risk for many malignancies including those of the GI tract and the liver. Even light to moderate alcohol intake among women is associated with increased risk for breast cancer.
Are we willing to sacrifice the asset “cardioprotection” for the well-known liabilities that alcohol imposes on a plethora of other risk factors. So what if alcohol prevents you from developing heart disease at the expense of breast cancer.
In this, I believe that those “recommending” the moderate intake of alcohol for health are misled and misleading. The advice to “enjoy a glass of wine or two with dinner” to improve heart health is, in my opinion, fantastically naive. Fortunately, not all are this irresponsible.
Organizations such as the American Heart Association, and I’m sure others, provide statements such as “if you DO drink, do so in moderation”. This is acknowledges the known proclivity of most Americans to consume alcohol, however regularly. However, it is more a statement to “reduce alcohol consumption” rather than “increase consumption to low to moderate amounts.” Authors of the Lancet study support this notion in saying that “we are unaware of any guidelines that encourage non-drinkers to consume alcohol.”
My take — ever since hearing of this alcohol-health association has been this. I DO, obviously buy into the fact the moderate alcohol consumption is associated with better health — for the data tell me so. The reasons I believe this diverge from the “its the alcohol or something in it” idea.
Who drinks alcohol “in moderation”, and where do they do it? Often, those who engage in social situations of a few number of occasions per week, who are having drinks with friends, in social atmospheres. Who also consumes alcohol “in moderation?” Those most likely aware of the harms of excessive alcohol who are similarly taking care of themselves in every other aspect of life (healthy-user bias). I think, however, the social aspect might explain most of our story. I find that I am probably most happy when engaging in a few social events per week, going out on the town, and generally winding down while not thinking about work, school, life. In these situations, I’m generally consuming “low to moderate” amounts of alcohol. But not for my health.
The impact that this happiness and enjoyment probably has on my own health, well-being, and longevity, cannot be understated.
Perhaps the French and those in Mediterranean regions got it right for reasons we don’t understand. Perhaps, the benefits of moderate red wine consumption are entirely separate from the beverage itself, but rather, from the traditions surrounding its consumption, which involve moderate consumption before and during the largest meal of the day, which it turns out, happens to be the one shared with family and friends. Wine and its benefits, both longevity and happiness related, may in fact be mediated through “enhanced social bonding, with an emphasis on moderation generally espoused by this tradition.”
A toast to health
References
O’Keefe et al. Alcohol and cardiovascular health: the razor-sharp double-edged sword. Journal of the american college of cardiology. 50 (11): 1009–1014 (2007)
O’Keefe et al. Alcohol and cardiovascular health: the dose makes the poison…or the remedy. Mayo clinic proceedings. 89(3): 382–393 (2014)
Ferrieres J. The French Paradox: lessons for other countries. Heart. 90: 107–111 (2004)
Bau J., Sinclar D. Therapeutic potential of resveratrol: the in vivo evidence. Nature Reviews drug discovery. 5: 493–506 (2006)
Wood et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. The Lancet. 391:1513–1523 (2018)
