STAIRS is a European Commission funded study whose aim is to produce a set of guidelines for a harmonised, crash injury database.

The need to evaluate the effectiveness of the forthcoming European Union front and side impact directives has emphasised the need for real world crash injury data-sets that can be representative of the crash population throughout Europe. STAIRS will provide a methodology to achieve this. The ultimate aim of STAIRS is to produce a set of data collection tools which will aid decision making on vehicle crashworthiness as well as providing a means to evaluate the effectiveness of safety regulations. This paper will disseminate the up-to-date findings of the group as they try to harrnonise their methods.

The stage has been reached where studies into the diverse methods of the UK, French and German systems of crash injury investigation have been undertaken. An assessment has already been made of the relationships between the three current systems in order to define the areas of agreement and divergence. The conclusions reached stated that there were many areas that are already closely related and that the differences were only at the detailed level.

With the emphasis on secondary safety and injury causation, core data sets were decided upon, taking into account:​ Vehicle description, collision configuration, structural response of vehicles, restraint and airbag performance, child restraint performance, Euro NCAP, Pedestrian and vehicle occupant kinematics, injury description and causation. Each variable was studied objectively, the important elements isolated and developed into a form that all partners were agreeable on. A glossary of terms is being developed as the project progresses which includes IS0 standards and other definitions from the associated CAREPLUS project, which addresses the comparability of national data sets.

A major consideration of the group was the data collection method to be employed. The strengths and weaknessesof each study were investigated to obtain a clear idea of which aspects offered the best way forward. The quality of this information and transference into a common format, as well as the necessary error checking systems to be employed have just been completed and are described.

In tandem with this area of study the problem of the statistical relationship of each sample to the national population is also being investigated. The study proposes a mechanism to use a sample of crash injury data to represent the national and international crash injury population.