Belfast, Northern Ireland: Survey results estimating the prevalence of drug use among young people are unreliable because of undetected reporting errors, according to the results of a longitudinal study published in the current issue of the journal Addiction.
A research team at the Institute of Child Care Research at Queens University in Belfast found that adolescents fail to consistently report their drug use over a multi-year period. Authors discovered that many respondents initially admit to having used illicit drugs and then deny their use on subsequent surveys.
“In general, as the social stigma of the drug increased so, too, did the proportion of previous reports that were recanted,” authors wrote, noting that the consistency of positive life-time reporting was highest for more socially accepted substances such as alcohol (93%), tobacco (90%), and cannabis (83%), while it was lowest for psychedelic mushrooms (87%), heroin (85%) and cocaine (82%).
Specific to marijuana, authors found that drug education was associated with increased recanting by respondents.
“The possibility of drug education biasing drug use reporting, via increased recanting independent of actual behavior change, may have substantive implications for the evaluation of drug education itself,” authors wrote. “It could be argued that evaluation studies showing a positive effect from drug education (i.e. a decline in reported drug use in an intervention group relative to a control group) may in fact be reporting differences in the willingness of young people to give truthful answers to the drug use questions rather than changes in willingness to use illicit substances.”
Researchers concluded: “The high levels of recanting uncovered cast doubts on the reliability of drug use reports from young adolescents. Failure to address this response error may lead to biased prevalence estimates, particularly within school surveys and drug education evaluation trials.”
In the United States, politicians and researchers rely primarily on two annual self-report surveys to estimate drug use among the nation’s population: the Monitoring the Future project at the University of Michigan (which measures the prevalence of drug use among adolescents) and the US Department of Health and Human Services’ National Survey on Drug Use & Health (which measures the prevalence of drug use among all age groups).
In the former, results are estimated by administering a questionnaire to students in a classroom setting. In the latter, federal researchers administer questionnaires to residents through face-to-face interviews at their home. In both cases, however, researchers have expressed concern that the social stigma associated with drug use provides a powerful disincentive for respondents to provide truthful self-reports. For example, according to a White House briefing paper analyzing SAMHSA’s figures regarding Americans alcohol and tobacco use, respondents were shown to have under-reported their usage by as much as 30 to 50 percent.
“It is troubling that so many politicians and bureaucrats continue to base the perceived success and/or failure of America’s $40 billion per year drug policies primarily on the basis of these social surveys’ estimates,” said NORML Executive Director Allen St. Pierre, “particularly when the surveys themselves have time and time again shown to be an unreliable way to accurately measure drug use.”
For more information, please contact either Allen St. Pierre or Paul Armentano of NORML at (202) 483-5500.