Purpose:
Australia is currently in the lowest prevalence group of OECD countries for tobacco
consumption. But how comparable are these estimates, given they are derived from varying surveying methods, and in particular varying survey modes?
Methods:
The 2007 National Drug Strategy Household Survey was the ninth survey in a current series,conducted every 3 years. Since the 2001 survey, both the computer-assisted telephone interview (CATI) and drop-and-collect (D&C) pen-and-paper method have been employed simultaneously. These method-specific samples are combined to provide a single national prevalence estimate.
Results:
There was a 2.5 percentage point variation in prevalence of daily smoking by survey mode; ven higher for some population groups. The prevalence variation by survey mode was statistically significant, and materially significant in terms of rankings of OECD countries. Australia's choice for an official estimate reflects a balance among various surveillance principles, cost and consistency. Notwithstanding this balance, the resulting estimate pulls up short of being the lowest among OECD countries.