This paper presents a new conceptual model for drinking after driving, that is designed to integrate both epidemiological and experimental approaches. The model was developed at the Vermont Alcohol Research Center, USA, where it provides a guide for the design and conduct of its research, and helps to integrate its many results. Detailed discussions of the Center's studies are given in other papers of this conference (see relevant abstracts). The model focuses on measurements of alcohol-involved driving before, during and after driving. Epidemiological field observations are made before and during driving, phenomenological survey studies during and after driving, and experimental laboratory studies after driving. The model is empirically supported by data obtained from: (1) unobtrusive observation of motorists' drinking behaviour in public bars before driving; (2) the same motorists when they later become drinking subjects in a series of laboratory experiments involving many levels of alcohol from both ad lib drinking and controlled doses. During three summers of roadside surveys of night drinking drivers, about 25,000 motorists were stopped, briefly questioned, and tested for blood alcohol concentration (BAC). The survey studies after drinking involved the bar and roadside drivers, together with samples from the general driving population and drink and driving offenders. Some results are summarised in graphs.
Abstract