The national accident investigation system developed by the National Highway Traffic Safety Administration (NHTSA) has been changing from general multidisciplinary accident investigation projects to mission oriented multi-level, bilevel, and special studies. Currently the research plan for Field Accident Studies calls for incorporating statistical validity into the product of these efforts. Problem areas currently under study or in the planning stage include alcohol, neck injuries, injury causation, vehicle defects, postcrash factors, pedestrians, trucks, recreational vehicles, trends and validation of indices.

Two general categories of findings emanate from the current and past study projects. First are those findings and conclusions which evaluate, measure or describe a phenomenon which in itself generally needs corroboration in frequency or occurrence. This is accomplished by employing bilevel and special study techniques. The second set of findings are those which stimulate some decision actions based on a single occurrence or a small number of occurrences and result in more immediate payoffs in highway safety. Examples of this type of safety benefit, derived from collision investigation activities, are given in this paper.