CLASSIFICATION AND ESTIMATION OF ALCOHOL INVOLVEMENT IN FATALITIES

Surrogate measures for alcohol involvement in fatalities are used because blood alcohol content (BAC) is not well reported in general. In addition, determination of alcohol consumption in the absence of BAC testing is often subjective and, therefore, less reliable. This study presents a methodology for solving the problem of determining alcohol-involved fatalities. The objectives were to: (1) develop an appropriate surrogate measure for alcohol-involved fatalities using statistical methods; (2) use the surrogate to classify those fatalities for which alcohol involvement is unknown; and (3) estimate total national alcohol-involved fatalities for 1976-1981. An alcohol-involved fatal accident was defined as one in which an active participant in the accident had a known BAC equal to or greater than 0.05. Multivariate discriminant analysis (MDA) was used on the known cases to develop functions for classifying the unknown fatal accidents and fatalities. Box-Jenkins time series analysis was used to model the estimated total alcohol-involved fatalities from 1976-1981. Nationally, an annual increase of approximately 100 alcohol-involved fatalities per month occurred during 1976-1981. The proportion of alcohol-involved fatalities increased slightly from 68% to 72% during the same period.

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00393623
  • Record Type: Publication
  • Source Agency: National Highway Traffic Safety Administration
  • Report/Paper Numbers: HS-037 521
  • Files: HSL, TRIS, USDOT
  • Created Date: Apr 29 1985 12:00AM