Worldwide 1.25 million people die in road crashes every year and it is widely recognized that drink driving is an important risk-increasing factor. Official statistics of alcohol-related crashes are likely to underestimate alcohol-related crashes and casualties, because official statistics are affected by underreporting. This study aims to obtain a good insight into the definitions, legislations and reporting procedures of alcohol-related road casualties to reach an accurate estimate of the drink driving problem and recommendations on how to improve the reliability and comparability of official statistics. A total of 45 countries, represented by road safety experts, responded to an online questionnaire. The questionnaire consisted of questions divided into four main categories: drink driving legislation, definitions of alcohol-related road casualties, recording methods, quality of official statistics and experts’ best estimates. A weighted average of 21.8% alcohol-related road deaths is found in official statistics in the group of 45 countries and this proportion remains constant between 2000 and 2010. There are first signals that the proportion in 2015 is lower than in the period 2000-2010. If the proportion of 21.8% applies worldwide and is based on 1.25 million road fatalities per year, the annual number of alcohol-related road deaths will be around 273 000. However, this number is an underestimate of the real problem because strong indications of underreporting of alcohol-related crashes in official crash statistics are found. Most countries (89%) still base their official data upon only one single data source and in most cases (87.5%) these are the police records for which this study found evident shortcomings. Furthermore, countries use different definitions which makes international comparison difficult.
Recording of alcohol in official crash statistics: underreporting and procedures to improve statistics