BEHAVIOUR CHANGE Vol. 16, No. 2 1999 pp. 111–126 P ersistent driving offending is a substantial problem for law-enforcement and correc- tional agencies. The relatively high rates of con- victions for traffic offences and the subsequent legal and social costs point to the need for effective intervention. The absolute numbers are also large; for example, in New Zealand in 1992, over 9,000 drivers were convicted of driving while disqualified (DWD). Approximately 90% of these individuals were initially disqualified because of excess blood- or breath-alcohol levels (Bailey, 1993). Approximately 16% of these people failed to abide by their disqualification and were subse- quently incarcerated. These individuals repre- sented almost one fifth of all those sentenced to imprisonment in the 1992 year in New Zealand (Spier, Norris, & Southey, 1993). It appears that overseas figures are similar to those of New Zealand although, given different sentencing practices, they can be hard to interpret (Bakker, Ward, Cryer, & Hudson, 1997). Many of these offenders have long histories of driving while disqualified. For example, of the 311 disqualified drivers in prison during the 1991 New Zealand Penal Census, 169 (54%) had at least six previous convictions (Braybrook & Southey, 1991). Moreover, 88% of these indi- viduals had previously been imprisoned. Although these imprisoned drivers had also A Descriptive Model of the Relapse Process in Disqualified Drivers Leonie Wilson University of Canterbury Tony Ward University of Melbourne Leon Bakker Department of Corrections, Christchurch, NZ Driving-while-disqualified (DWD) offenders represent a significant proportion of the driving offender population, and contribute to a significant drain on governmental and societal resources. The reoffence or relapse process of recidivist DWD offending has been somewhat neglected in the past, and this paper represents a first attempt at developing an empirically gen- erated model of this phenomenon. In this study, we undertook a grounded theory analysis of offence chain information obtained from 28 male recidivist DWD offenders. The grounded theory analysis resulted in the development of a 15-stage model of the relapse process of recidivist DWD offenders. The model identifies and describes the sequence of psychological, behavioural, and environmental factors, including mediating variables that contribute to DWD reoffending. The model was then applied to 14 new offence chains to examine its level of satu- ration, content validity, and inter-rater reliability. The results suggest that the model has provi- sional validity and adequate inter-rater reliability, and achieved a high level of saturation. The strengths and value of the model are discussed, as are its clinical and research implications. Address for correspondence: Dr Tony Ward, Department of Criminology, University of Melbourne, Parkville VIC 3052, Australia. Email: t.ward@criminology.unimelb.edu.au 111