A longitudinal study of social, psychological and behavioural factors associated with drunken driving and public drunkenness.

Studies on psychosocial conditions in drunken drivers have generally been cross-sectional and based on rather small selected samples. The objective of this study was to analyse, in a longitudinal perspective, the relationship, in young males, between social and psychological factors and indicators of alcohol abuse on one hand and the risk of subsequent drunken driving and public drunkenness on the other hand, in order to identify similarities and differences in risk factor patterns. Questionnaire information from 8122 military conscripts in 1969/70 was linked to data on drunken driving and public drunkenness for 495 males with offences registered up to 1977. Logistic regression analysis showed that the relative risk (RR) for high alcohol consumption, smoking, use of narcotics and sniffing of solvents had a statistically significant association to subsequent drunken driving and public drunkenness in univariate analyses. In multivariate logistic analyses, RR remained increased for those with fathers belonging to social class II and especially so for those coming from social class III. Smoking (RR 3.3, with a 95% confidence interval of 1.6-6.8) was significantly increased in drunken drivers with a blood alcohol concentration (BAC) of 0.15% or more at apprehension, as was truancy or contact with police or juvenile authorities in drunken drivers with a BAC of 0.05-0.15%, and illicit drug use, intoxication drinking, contact with police or juvenile authorities and hangover with public drunkenness. Thus, we found that early social and behavioural factors, substance abuse and risky use of alcohol were predictors for both drunken driving and public drunkenness, with no marked differences in risk factor patterns.