About one-fifth of adults in the United States have experienced some form of harm due to someone else's behavior while drinking.
That's according to a study published Monday in the Journal of Studies on Alcohol and Drugs , which found that in 2015, an estimated 53 million adults -- or nearly 1 in 5 -- said they had experienced at least one harm attributable to someone else's drinking in the past year. That harm ranged from property damage to physical injury.
"One thing to think about with the one-in-five number is that it is only limited to a snapshot in time of about a year. So probably more people have actually been harmed by someone else's drinking at other times in their life," said Katherine Karriker-Jaffe, a senior scientist with the Alcohol Research Group at the Public Health Institute in Emeryville, California, who was an author of the study.
"So it might be an underestimate of the negative impacts of alcohol on people other than the drinker," she said.
The study involved analyzing data on 8,750 adults who answered survey questions from two databases : the 2015 National Alcohol's Harm to Others Survey and the 2015 National Alcohol Survey. The surveys were conducted from April 2014 to June 2015. The study did not include children.
Each adult was asked whether they had experienced any of 10 different types of harm in the past 12 months caused by "someone who had been drinking."
The different types of harm included harassment; feeling threatened or afraid; having belongings ruined; having property vandalized; being pushed, hit or assaulted; being physically harmed; being in a traffic accident; being a passenger in a vehicle with a drunk driver; having family or marital problems; and having financial trouble.
The researchers found that 21% of women and 23% of men in the study reported experiencing at least one of those harms in the past year. The most prevalent type of harm was harassment, according to the data.
When it comes to harms other than harassment, "for women, the most prevalent are family and marital problems or financial problems due to someone else's drinking and a close third runner-up would be driving-related harms -- so riding with a drunk driver or actually having a crash caused by someone who had been drinking," Karriker-Jaffe said.
Other than harassment, "for men, the driving-related harms were the most common, followed by property damage and vandalism," she said.
Overall, women were more likely than men to report harm by a spouse, partner or family member who had been drinking, and men were more likely to report harm because of a stranger's drinking, the data showed.
The study had some limitations, including that the data was self-reported, which lends itself to bias if a person in the study was not answering survey questions honestly.
Also, more research is needed to determine whether similar findings would emerge for other years, as the data was collected in 2014 and 2015.
The study findings were "fascinating" for Aesoon Park, an associate professor of psychology at Syracuse University in New York, who has conducted research on alcohol use and misuse but was not involved in the new study.
She noted the study found that younger adults were more likely to experience a broad range of secondhand harms due to someone else's drinking compared to older adults.
"We know now that people who are 18 to 25, they are showing the highest rates of alcoholism," Park said.
"What is interesting about this study is that not only is it about alcohol use disorder, but it shows how the secondhand effect of alcohol is also affecting that same age group," she said. "The second interesting part to this is the gender inequality."
She said both men and women seem to be affected by the secondhand effect of alcohol, even though men are more likely than women to drink excessively . "So it highlights a gender inequality of the secondhand effect of alcohol," Park said.
Dr. Timothy Naimi, a physician and alcohol epidemiologist at Boston Medical Center in Massachusetts, wrote an editorial published alongside the study Monday.
"The underreporting of harms among some individual respondents, coupled with the fact that previous harm leaves some portion of the population unable or less likely to participate in surveys because of premature death, injury, or psychological distress, suggests that even this robust prevalence is likely an underestimate," Naimi wrote.
"This is an emerging area of investigation in its relative infancy and is one that needs nurturing and growth," he wrote in part. "Prevention of secondhand effects from others' drinking at the population level must be driven by structural, environmental interventions that reduce excessive drinking."