Since random breath testing (RBT) began in Adelaide, South Australia, its effectiveness has been monitored by a series of roadside breath alcohol surveys, during which the level of illegal drink-driving there has decreased. Some results are presented of a roadside survey conducted from February to May 1991. The survey was conducted between 10pm and 3am at 20 signalised intersections. As each light changed to red, the driver of the first car to stop was asked to do a breath test and given a reply-paid mail questionnaire. 6789 drivers were approached, of whom 94.8% did a breath test and 41.9% answered the questionnaire. Tables are given of: (1) the characteristics of respondent and non-respondent drivers; (2) the proportions of groups of drivers who perceived that they were likely to be caught; (3) the effect of experience of RBT on driver perception; and (4) the relationships between perception and blood alcohol content (BAC) and between perception and intention. There was good evidence that RBT has been associated with a decrease of drink-driving in Adelaide, but probably because much of the influence of RBT is to create and maintain an awareness of the drink-driving problem. A case can be made for applying more RBT to drivers unwilling to stop drink-driving.
Abstract