About That Study That Says No Amount of Alcohol is Safe

NPR reports today:

No amount of alcohol is safe, according to The Global Burden of Diseases study, which analyzed levels of alcohol use and its health effects in 195 countries from 1990 to 2016.... 

"Our results show that the safest level of drinking is none," the report states. "This level is in conflict with most health guidelines, which espouse health benefits associated with consuming up to two drinks per day."

Well. The findings here are accurate but weird and misleading. If you click through to the Lancet study, you find that the authors are using incredibly broad measures to arrive at this conclusion, both in terms of what constitutes a "health" outcome and which populations they include. One could easily assume that the authors are suggesting that taking alcohol into your body, even at very low doses, will harm it. But their data don't show that. Instead, they take all harm caused by alcohol--drunk driving, suicides, violence caused by drunkeness--in all parts of the world, and call these health effects. Let's have a look.

Globally, alcohol use was the seventh leading risk factor for both deaths and disability-adjusted life-years in 2016, accounting for 2.2% female deaths and 6.8% male deaths. The three leading causes of attributable deaths in this age group [15-49] were tuberculosis (1.4% of total deaths), road injuries (1.2%), and self-harm (1.1%).

I want to stop here before we get to deaths of the over-fifty set to point out how preposterous these findings are. Two of the top three causes of death aren't even health-related. The third is, but I don't know how relevant it would be to an American drinker. According to the CDC, "A total of 9,272 TB cases were reported in the United States in 2016." I assume this is a statistical quirk that comes from TB problems in other parts of the world, but drinking is definitely not a major cause of TB here.

It is associated with greater risk for older drinkers. "For populations aged 50 years and older, cancers accounted for a large proportion of total alcohol-attributable deaths in 2016, constituting 27.1% of total alcohol-attributable female deaths and 18.9% of male deaths."

The authors acknowledge that there are positive benefits to heart health from alcohol, but render this judgment: the 'protective effects for ischaemic heart disease and diabetes in women are offset by monotonic associations with cancer." This indicative of the thrust of this study, which my old colleagues in politics would have called "advocacy research." The authors have judged alcohol to be a scourge on society and when assessing "health" effects, figure that into the calculation. This is not my gloss of the report, it is their explicit finding:

"The conclusions of the study are clear and unambiguous: alcohol is a colossal global health issue and small reductions in health-related harms at low levels of alcohol intake are outweighed by the increased risk of other health-related harms, including cancer. There is strong support here for the guideline published by the Chief Medical Officer of the UK who found that there is 'no safe level of alcohol consumption.'”

This is a political document masquerading as epidemiology. I might even agree with the findings, but I find them deceptive and unhelpful when cast as health research. It's fair to look at the whole of alcohol's negative societal effects and decide how to regulate it. It is misleading in the extreme to suggest that an individual consuming modest amounts of alcohol is endangering her life. Put another way, you driving into a tree does not damage my health. They are categorically different things and should be treated as such. Mashed together this way they are designed to conflate separate issues to support a preferred conclusion.