"All Alcohol Is Bad" – Esquire
"No Amount of Alcohol Is Safe, Health Experts Warn" – CNBC
"Moderate Alcohol Consumption Is Out" – Forbes
I groaned when I saw these headlines last week.
A new study published in leading science journal The Lancet had the mainstream media warning people against any alcohol consumption.
When news like this hits, we know better than to blindly trust reporting. We know to look past the headlines and dig into the research.
That's what I did last week, and here's what I found...
This study is another example of terrible science and even worse reporting. In fact, I can narrow the biggest problems to three points that explain why this kind of research is a waste of taxpayer dollars.
1. Correlation. My first point is one you're likely familiar with: Correlation does not equal causation.
Correlation means two things happen together. In other words, they rise and fall at the same time and at the same rate. But it does not mean one of those things causes the other. A good example – in his book, Spurious Correlations, Tyler Vigen points out that the divorce rate in Maine correlates closely with the per capita consumption of margarine. They both rise and fall at the same rate. In fact, if you've attended one of our Alliance conferences, you've likely seen me use this example in my presentations. But just because these two factors trend the same way doesn't mean that eating too much margarine leads to divorce.
We've reported on correlation studies before... but we back these up with solid, mechanism of action studies. For instance, many reports on coffee's health benefits appear correlational. But with further digging, we know specifically how molecules in coffee interact with our DNA and with "bad" radicals in our bodies.
Even though the Lancet paper looked at hundreds of studies, the overall conclusions are just that – correlations. There's nothing to prove that alcohol alone causes more deaths, let alone that a single glass will increase your risk of death. Which brings us to the next point...
2. Risk. The stat you've likely seen sounds scary: Folks who have one drink a day had a 0.5% increased relative risk of 23 alcohol-related problems compared with non-drinkers. With a large enough sample size, this can be statistically significant, but it's not practical in the real world.
You know what that really breaks down to? In 100,000 people who had one drink a day, 918 of them experienced a medical "event" like a heart attack or alcohol-related injury. But guess what... In 100,000 non-drinkers, 914 people experienced such an event. That's a four-person difference out of 100,000 people. And that's across 23 different events. I'd hardly call that clinically significant.
What's more, they didn't account for factors we know cause problems like heart attacks and car crashes. That includes a history of family alcoholism and smoking.
Again, this is also relative risk, not absolute risk. Relative risk is simply the difference between the groups in a scientific study. In this case, that's folks who have one drink a day and those who have none.
Your overall lifetime risk is "absolute risk," which is usually much lower. In fact, your absolute lifetime risk of dying from a car crash in Montana is one in 2,302. And yes, we wrote Montana... That's because even trying to find the statistic on absolute risk for this event requires more factors for consideration than this study.
And looking at multiple factors brings me to this last issue...
3. Focus on the wrong outcome. The researchers claim that this tiny (and again, clinically insignificant) rise in risk for one drink a day means we should throw out all our whiskey.
Here's the problem... the rest of the data showed that drinking heavily caused significantly more problems. We already know the most benefit is on a U-shaped curve. Those who don't drink and those who drink too much have the worst health outcomes. Light to moderate drinking is fine. That's what we should focus on here... Maintain your limit of one or two drinks per evening. Better still, drink only on social occasions and use common sense to consume in moderation.
What's more, the study doesn't break down types of alcohol. We know for a fact that wine provides health benefits when consumed in moderation, as do beer and whiskey.
Finally, the authors do a poor job ruling out other factors for things like heart attacks. If you think diet, exercise, and stress management don't contribute more to a heart attack risk than alcohol alone, you need to read more issues of Health & Wealth Bulletin.
My research team dives into topics like correlation studies and why we need to understand the mechanism of action. If you want to read more, we cover these subjects in my monthly newsletter, Retirement Millionaire.
These are just three of the reasons I take issue with massive analysis-based studies like this. Remember, "trust, but verify" is our motto – don't believe everything you read.
- Something different: Alcohol at Disney?
Here's to our health, wealth, and a great retirement,
Dr. David Eifrig and the Health & Wealth Bulletin Research Team
September 6, 2018