Influence factors of driving under the influence of alcohol and their regional differences : an exploratory analysis. Since the 1970s, studies are made in the Netherlands into driving under the influence of alcohol during weekend nights. Partly based on the results of these studies, the influence of different measures on the trend of alcohol use in the Netherlands can be estimated. Differences between regions concerning driving under the influence are also made visible by these studies. Whether these differences are systematic or caused by random fluctuations or differences in the sample, has not yet been investigated. Nor was examined which factors may play a role in regional differences concerning the use of alcohol in traffic. This exploratory study first of all investigated whether the data of 2010 indeed support the existence of regional differences concerning driving under the influence. After correction for differences in day of the week and time of day on which the alcohol tests were taken, and for differences in gender and age of the drivers who were tested, a logistic regression analysis indicated differences between provinces. Closer examination indicated that only the provinces Noord-Holland and Groningen differed from the other provinces. In the year 2010, Noord-Holland had a greater share of drivers under the influence (0,5g/l alcohol or more), whereas in the province Groningen the share was smaller than average. After correction the other Dutch provinces no longer show any significant differences. Then a second logistic regression analysis was carried with factors that may play a role in these regional differences. The findings indicate a number of possible factors: * enforcement level (a higher enforcement level coincides with a lower alcohol consumption level in traffic); * urbanization level (a higher urbanization level coincides with a higher alcohol consumption level in traffic); * presence of places of entertainment (higher density of places of entertainment coincides with a higher alcohol consumption level in traffic); and a * factor based on lifestyle, religion and attitudes (a greater share of people who say they never drink and a greater share of followers of certain faiths coincided with a lower alcohol consumption level in traffic). The enforcement level is the only factor that can be influenced by policy. The data used as a basis for this study has been provided by different sources which (with only a few exceptions) are not intended for road safety research. The results of the analyses are largely dependent on the quality and the aggregation level of the available data. This must therefore be considered as an important limitation of the present study. Presently, however, no better data is available. It is advisable to repeat this study in a later year to have more data available to validate and hopefully support the results of the present exploratory study. This can also limit the role of random fluctuations in the data.
Samenvatting