Aim: To examine the consistency of illicit drug use estimates from a telephone survey compared to an anonymous self-complete survey.
Background: Non-anonymous interviewing was considered best practice when seeking information on sensitive behaviours; although most studies making these recommendations arose during the 1980s and 1990s. Subsequent attitude shifts towards illicit drugs, mobile phone availability, and mass social media uptake challenges the existing advice on methods for eliciting information on sensitive behaviours.
Method: Ever and recent use of ecstasy, marijuana, methamphetamine and cocaine was collected on the NSW Population Health Survey (NSWPHS) and the National Drug Strategy Household Survey (NDSHS) in 2016. Methods for both surveys have been described elsewhere. Prevalence estimates and odds ratios were calculated using R’s ‘Survey’ package. Respondents that answered questions on illicit drug use, lived in NSW, and were aged between 16 and 44 years were analysed (NSWPHS: N=2,367; NDSHS: N=2,299).
Results: State-wide estimates of recent ecstasy and methamphetamine use were consistent between the two surveys, while recent marijuana and cocaine use were slightly lower on the NSWPHS. Ever use estimates in the NSWPHS were consistently lower compared to the NDSHS. The results remained the same after adjusting for age and sex.
Conclusion: Estimates of illicit drug use were generally lower for the telephone survey compared to the anonymous self-complete survey. However, recent ecstasy and methamphetamine use were consistent between the two survey approaches. Observed differences between the surveys have implications for the measurement and interpretation of statistics on illicit drug use, which needs further exploration.