Comments on the Agency's Plan To Evaluate the Occupant Crash Protection Standard
CED-80-70: Published: Feb 28, 1980. Publicly Released: Feb 28, 1980.
- Full Report:
Comments were provided on the Evaluation Plan for Federal Motor Vehicle Safety Standards 208. The review was limited to an overview of the potential for making the evaluation and of the data sources to be used in the evaluation plan. The National Highway Traffic Safety Administration proposes to use three accident databases to evaluate the effectiveness of Federal Motor Vehicle Safety Standard 208. Included are (1) the National Accident Sampling System (NASS); (2) the Fatal Accident Reporting System (FARS); and (3) selected State accident data files. Of these three systems, the Safety Administration has the most confidence in NASS. Although NASS will be the primary data source for the overall plan and will be used to estimate the effectiveness of different automatic restraint systems, not enough data from that system will be available until late in the evaluation period for measurements of injury reductions and even later for measurements of fatality reductions. Meanwhile, FARS data and State accident data will be used to obtain initial effectiveness estimates and interim answers to some of the basic questions. The Safety Administration believes that the importance of Standards 208 demands that data collection on the worth of the standard should begin as soon as possible in spite of the fact that this information will be less detailed and less desirable. FARS will be used to provide interim answers of automatic restraint effectiveness in reducing fatalities. The FARS data file contains a census of all fatal traffic accidents which occur in the United States. The information in this data file is derived primarily from State accident report forms, supplemented with additional information made available by the States. Additionally, effectiveness estimates calculated from State data will be used initially to calculate the injury reduction effectiveness of the air bag system and the automatic belt system.
The success of the plan is highly dependent upon the expansion and full implementation of NASS. A major difficulty which must be faced in using NASS data to evaluate Standard 208 is sample size. Additionally, the plan fails when describing the when and how, nor does the plan provide an estimate of when NASS will have data available to estimate the fatality-reducing effect of specific restraint systems. Though NASS is the major data system used, FARS will be used to provide interim answers of automatic restraint effectiveness. While FARS contains data on fatal accidents involving vehicles equipped with active restraints, it does not provide a direct measure of the exposure of those vehicles to accidents. Rather, some indirect measure exposure must be derived for the automatic-restraint-equipped and the active-restraint-equipped vehicles. Once this measure of exposure is defined, then fatality rates for vehicles equipped with automatic restraints and active restraints can be compared, and an estimate of the relative effectiveness of automatic restraints in reducing fatalities can be calculated. Further, because FARS data are difficult to estimate the effectiveness of automatic restraints in reducing fatalities, additional detail needs to be provided to resolve, or at least explain, the difficulties. Another system which will provide data in the interim is state data which are necessary to calculate injury reduction effectiveness estimates. This system creates problems because of its lack of uniformity. Thus, action must be taken to create consistent and reliable data.
Recommendation for Executive Action
Comments: Please call 202/512-6100 for additional information.
Recommendation: NHTSA should: (1) publish the NASS Advisory Committee's comments and recommendations along with its disposition of them in a public document; (2) separate itself from the analysis and conclusion phases of the evaluation by having independent contractors perform this work; and (3) have general agreement from all parties affected by Standard 208 concerning the types of data to be collected and the methods of analyzing that data.