It's easy to imagine how second-hand smoke can impact an innocent bystander - most of us have probably inhaled the smoke from someone else's tobacco without wanting to. But it's not the only habit that can affect others; 'second-hand drinking' is a thing, too.
It's not quite as literal, as it doesn't mean the alcohol you drink will end up in the bloodstream of another. But over the years it's become increasingly clear that second-hand drinking is a significant public health issue. Now, a study in the United States has placed its effects on par with second-hand smoking.
Using US national survey data from 2015, research has found that each year, one in five American adults - which is an estimated 53 million people - experience harm because of someone else's drinking. These harms include things like threats or harassment, which was the most commonly cited, as well as physical aggression, drunk driving, and even financial or family problems.
What's more, the authors found that the burden of second-hand drinking was not experienced equally. Young people under the age of 25 were mostly on the receiving end, and in the year previous to the survey, some 21 percent of women and 23 percent of men said they experienced harm because of someone else's alcohol use.
For men, this harm was usually from a stranger and resulted in ruined property, vandalism, or physical aggression. But for women, the drinker tended to be a family member, causing financial issues or problems in the house.
The authors of the study conclude that their findings substantiate previous research "documenting the considerable risk for women from heavy, often male, drinkers in the household and, for men, from drinkers outside their family."
"Further," they add, "our findings are consistent with recent data from outside the United States that highlight the significance of the proximity of male harmful drinkers for women's victimisation by others who have been drinking."
The results, which were based on two telephone surveys of 8,750 adults, found that even people who drank but not heavily were more than doubly at risk of experiencing harassment, threats, and driving-related harm from someone else's drinking. (In this case, heavy drinking was defined as five or more alcoholic beverages at a time at least monthly for men, and four or more alcoholic beverages for women).
Most of us know that alcohol comes with its fair share of health problems for the individual, but in a commentary published alongside the new paper, Sven Andréasson, a physician at the Karolinska Institutet of Stockholm, says that's only a part of the impact.
"When researchers compare the relative impact of different substances on the users and others, alcohol stands out as the substance that by far causes the most harm to others," he writes.
"When combining harms to the users with harms to others, alcohol again scores highest, ahead of tobacco, heroin, cocaine, and other substances."
In 2015, the US Centres for Disease Control and Infection announced that 58 million nonsmokers in the US are still exposed to secondhand smoke. This is around the same level of people estimated to have experienced harm from second-hand drinking, a number which is likely an underestimate.
Just as research on second-hand smoke led the way to better controls on smoking, the authors hope their research will spur legislation to reduce the health burden of alcohol.
"Among policies, taxes have strong evidence of reducing not only excessive drinking but outcomes in which second-hand effects are prevalent," writes Timothy Naimi, a physician at the Boston Medical Center in another accompanying commentary.
"And yet recently federal alcohol taxes have been cut, specific excise taxes in states have eroded by more than 30 [percent] in inflation-adjusted terms during the past 25 years, and current tax revenues only account for a fraction of alcohol-related costs."
It's about time we started taking second-hand drinking more seriously.
The research has been published in the Journal of Studies on Alcohol and Drugs.