This paper presents the methodology and results of an analysis of the available information on motor vehicle safety which could be used to provide a basis for establishing priorities for future Government and private sector efforts directed at enhanced crash protection. The work was stimulated by several factors: (1) 5 years have elapsed since the National Highway Traffic Safety Administration (NHTSA) published a plan for motor vehicle safety research and development, (2) motor vehicles have changed substantially over the past several years, (3) the quantity and quality of accident data and vehicle crash performance information have increased dramatically over the past 5 years, and (4) Government policies and the amount of Government and private sector resources available for future efforts are changing. This study takes the agency's automated data files on personal injury as the baseline information on injuries sustained by restrained and unrestrained occupants of all types of vehicles, and by pedestrians and cyclists. This information is analysed according to the magnitude of harm sustained, the potential for reduction in harm, and the relative adequacy of data and methods available for the exploration of solutions. The primary parameters used for this analysis are: crash severity, injury severity, injured body region, and source of injury in the vehicle. Secondary parameters are: car impact direction, occupant seating position, ejection path (if ejected), and restraint use. The human harm is quantified as a cost-weighted sum of the number of people injured or the number of injuries sustained, whether fatal or not. Fatality and fatality reduction estimates are also made. Priorities suggested by these data and analyses are discussed extensively.