Click here for the home page of Mothers Against Drunk Drivers
Friends don't let friends drive drunk. And friends shouldn't let friends publish junk.
To estimate how frequently U.S. adults drive while impaired by alcohol, Liu et al. conducted a telephone survey of 102,263 "non-institutionalized" adults aged 18 years and older. Based on their survey, these researchers reported that 2.5 percent of adults committed 123 million episodes of alcohol-impaired driving in 1993.
These sobering results were based on the following question:
During the past month, how many times have you driven when you've had perhaps too much to drink?
Based on answers to this question, Liu et al. calculated the number of episodes of alcohol-impaired driving per 1,000 adults in 1993. They multiplied the number of self-reported episodes of alcohol-impaired driving per respondent by 12 to obtain the number of episodes per year. In other words, say someone reported they drove while impaired once last month; then the researchers ASSUMED that the respondent drove impaired once every month!
How accurate is this assumption? Just because someone drove while impaired once last month, that does not mean they drive while impaired every month. It also doesn't mean that they didn't drink-and-drive more, or even less, in other months. Plain and simple, shaken or stirred, this assumption (which was not validated) is nonsense.
Moreover, self-reported data is notoriously unreliable. Subject to the phenomenon of "recall bias," people tend not to remember with real accuracy. They also tend to under-report anything, such as drinking and driving, that might have negative social connotations.
Of course, none of this stopped Liu et al. from concluding:
Based on these results, we recommend continued aggressive intervention to prevent alcohol-impaired driving.
While preventing drunk driving is highly desirable, I don't support fabricating facts to accomplish that goal. Science should not become another hit-and-run victim of junk drivers.
Click here for the home page of the National Coalition Against Drunk Driving.
