Alcohol-related road casualties in official crash statistics

IRTAD Research Report
Vissers, L.; Houwing, S.; Wegman, F.

What we did

This study examines how improving insights regarding the real number of alcohol-related road casualties worldwide can help to save lives. Every year 1.25 million people die in road crashes according to the World Health Organization. It is widely recognised that drink driving is an important riskincreasing factor and contributes to many road deaths. With great certainty, the real number of alcoholrelated road casualties is higher than reported in the official statistics. Better insights into reporting procedures is of the utmost relevance to arrive at comparable and reliable data.

For this study, a total of 45 countries were surveyed with the help of an online questionnaire. The survey was facilitated through the members of the International Transport Forum’s permanent working group on road safety, known as the International Traffic Safety Data and Analysis Group (IRTAD), the Ibero-American Road Safety Observatory (OISEVI) and the International Center for Alcohol Policies (International Alliance for Responsible Drinking (IARD), formerly ICAP), Washington.

The questionnaire looked at drinking and driving legislation and at definitions of alcohol-related road fatalities and serious injuries. Based on the information provided, the methods of recording alcoholrelated road casualties and the quality of the data were reviewed, with specific attention to the issue of underreporting.

What we found

Previous research found large differences in the share of alcohol-related road fatalities for different countries, ranging between 2% and 38% of the total. This is confirmed by this study, which found a share of alcohol-related fatalities ranging from approximately 5% to 35%. The official data of the countries surveyed for this study show that a weighted average of 21.8% among road deaths are alcohol-related. This proportion remains constant over the years (2000-2010). Accepting this figure as a reasonable estimate for all countries in the world and based on 1.25 million annual road fatalities worldwide, the alcohol-related deaths among fatally injured road users can be put at around 273 000 people every year.

There are indications, however, that this number underestimates the problem, because official statistics are affected by underreporting of alcohol-related crashes and casualties. As a result, the actual number of alcohol-related fatalities is probably higher still. (Serious injuries are even more prone to underreporting). The vast majority of countries surveyed (89%) still base their official data upon only a single data source. In most cases these are the police records (87.5%) for which this study found evident shortcomings. These deficiencies in data collection negatively influence the accuracy and reliability of the official data regarding alcohol-related road casualties, which complicates meaningful international comparisons. Official statistics do not give a complete picture of alcohol-related fatalities and serious injuries if only based police records of fatal crashes.

Furthermore, different countries use different definitions of what constitutes a road crash casualty. Definitions regarding alcohol-related serious road injuries in particular differ substantially between countries. This reinforces the distortions created through inaccurate recording of crash data when comparing countries. Therefore relying on official statistics will often be misleading. To enable more accurate analyses with a few to addressing the problem, improvements are needed.

What we recommend

Review how data on alcohol-related road crashes is collected

In order to come to more reliable and comparable data on alcohol-related serious road injuries and fatalities, countries should begin by assessing their current status on the recording of data on alcoholrelated road fatalities and serious injuries. In this assessment we suggest to check for compliance with the following good-practices:

Aim for a systematic alcohol testing of every road user actively involved in a serious crash

Ideally, 100% of active road users that are involved in a road crash that resulted in death or serious injury should be tested for alcohol. If a systematic alcohol testing at this level is not possible, countries should apply additional methods for adjusting the official numbers of alcohol-related road fatalities and serious road injuries.

Use statistical analysis methods to better estimate the number of alcohol-related road fatalities

Additional statistical analysis methods as described in this study can help to obtain better estimates of alcohol-related serious road crashed. Methods should be developed and applied that align with the legal system and data collection framework of individual countries, rather than harmonise methods internationally.

Harmonise definitions of alcohol-related road casualties

To make official country statistics comparable, definitions of alcohol-related road casualties should be harmonised. A number of European countries have already adopted the definition proposed by the 2009 SafetyNet project, as “any death occurring within 30 days as a result of a fatal road crash in which any active participant was found with a blood alcohol level above the legal limit”. A similar approach should be used to define a person seriously injured in an alcohol related crash, based for example on the severity level of 3+ on the Maximum Abbreviated Injury Scale (MAIS3+), so that it would be defined as “any serious injury at MAIS3+ that occurred as a result of a road crash in which any active participant was found with a blood alcohol level above the legal limit”. If countries are unable to apply these recommended definitions, developing algorithms to allow for conversion of these definitions is recommended.

Conduct future research on how to measure alcohol-related road crashes involving pedestrians and cyclists

In order to make sure pedestrians and cyclists are also counted as any active participant in the definitions on alcohol-related fatalities and serious injuries further research is needed regarding the possibilities of measuring blood alcohol concentration for pedestrians and cyclists involved in road crashes.


SWOV publication

This is a publication by SWOV, or that SWOV has contributed to.