Teenagers' reported use of fictitious drugs. A study of false positives

Abstract

Aims We have analyzed teenagers' self-reported experiences with a fictitious drug, and explored whether observed characteristics of individuals who report use of real, illegal substances change when overreporters are excluded from the analyses. Methods & Data Data stemmed from a school-based, nationwide survey of 13–19 year olds in Norway (N = 11928, response rate: 92%). Results Two percent reported that they had been offered the non-existent drug “zetacyllin”, whereas 0.5% allegedly had used it. “Zetacyllin users” appeared to be heavy users of existing substances, and therefore rates of GHB-, heroin-, LSD- and cocaine use dropped markedly when they were excluded from the sample. Correlates of such real drug use also changed when analyses without “zetacyllin users” were performed: The statistical effect of gender (i.e. being male) vanished, whereas the impact of age increased. The pattern of correlates with respect to indicators of socio-economic background also changed, but in different directions. Conclusions False positives may seriously distort the validity of survey-data on rare phenomena such as hard drug use. Due to this kind of response error, rates of uncommon behaviors may be inflated and observed correlates may deviate from the true pattern of correlates.