New research has called recent findings on the protective effect of moderate drinking into question, suggesting that the results may be due to errors in the participant selection process and inadequate controlling for potentially confounding factors.
Instead, the new study suggests that the apparent benefits of moderate drinking disappear when these factors are properly accounted for, with only a minor protective effect remaining in women age 65 and older. While we might want to believe that moderate drinking isn’t bad or is even good for us, the data don’t really suggest that’s the case.
Benefits of Drinking Debunked
Previous findings have suggested that drinking a glass of wine is good for your heart and that moderate or light drinkers live longer than teetotalers. However, the new study points out that there are many issues with making such statements. First, the authors argue that the common decision to lump “non-drinkers” into one group is misleading because this will include (alongside genuine teetotalers) many former drinkers, who are already known to have poorer health, to be more likely to suffer from depression and to be at increased risk of death in comparison to never-drinkers. Re-performing the calculations with the former drinkers excluded significantly reduces the observed effect, according to the new study.
Additionally, there are many factors that impact mortality risk, and these weren’t always well-controlled for in past studies—for example, some only took smoking status and age into account, and others only took smoking status and body mass index into account. Of course, many other factors can impact mortality risk, so without ruling more of them out, the apparent protective effect of alcohol could easily be an illusion.
Another related issue is that many studies discount people with health problems at the beginning of the study. There are reasons for this (for example, somebody who develops a health problem as a result of heavy drinking may start to drink less before the start of the study and consequently could be misclassified as a light drinker), but it may have the effect of selectively focusing on only the healthy drinkers, effectively ignoring the participants experiencing consequences from moderate drinking. Some people won’t die sooner because of moderate drinking, but only looking at these people doesn’t give a reliable picture of all moderate drinkers.
Mortality Data and Drinking Status
The new study aimed to re-address these issues by looking at 10 years’ worth of data from the Health Survey for England, looking for associations between alcohol consumption and all-cause mortality. The analysis focused on those 50 and over and used two samples: one of over 18,000 participants (including around 4,100 deaths over the 10-year follow up) and one of over 34,500 participants (including over 4,200 deaths over a six-year follow up).
The initial findings looked a lot like those of studies reporting a benefit to moderate or light drinking, but when the researchers adjusted for personal, socioeconomic and lifestyle factors known to affect mortality risk, the apparent “protective” effects were reduced. When the researchers discounted former drinkers from the “non-drinker” category, the apparent effects were reduced further still.
By this point, the only remaining protective associations were seen in younger men (aged 50 to 64) and older women (those over 65). For the men, the apparent benefits were present only in those limiting their consumption to 15 to 20 units per week (a typical drink contains about 2 to 3 units of alcohol.) For older women, the effects were significant for those drinking 10 units a week or less, but they were still fairly minor (only around a 15 percent to 25 percent reduction in risk).
The new findings have also been the subject of criticism, with some arguing that all groups drinking less than 20 units per week experienced lower mortality rates than teetotalers. However, looking at the data, it’s clear that such criticism is unfounded, based either on findings that lacked statistical significance or the analysis that included former drinkers in the same category as teetotalers.
Drinking Really Isn’t Good for You, Even in Moderation
While some may cling to the notion that for certain groups there is a protective effect to moderate drinking; the data don’t support the idea strongly, and the observations still persisting after adjustments in the new study could easily be due to other factors, like the small, unreliable samples of lifelong teetotalers. This definitely won’t be the last piece of research to investigate this question, but it shows the importance of thinking about the potential issues with any study making blanket claims about the benefits of a substance that kills almost 100,000 people each year in the U.S. alone.