Are female drivers safer? An application of the decomposition method

Epidemiology. 1998 Jul;9(4):379-84.


Using the decomposition method and national data for the year 1990, we examined gender and age differences in involvement rates in fatal motor vehicle crashes. The fatal crash involvement rate per driver is expresses as a multiplicative function of the crash fatality rate (defined as the proportion of fatal crashes involved among all crashes involved), crash incidence density (that is, number of crashes per million person-miles), and exposure prevalence (that is, annual average miles driven per driver). The fatal crash involvement rate per 10,000 drivers for men was three times that for women (5.3 vs 1.7) and was highest among teenagers. Of the male-female discrepancy in the fatal crash involvement rates, 51% was attributed to the difference between sexes in crash fatality rates, 41% to the difference in exposure prevalence, and 8% to the difference in crash incidence density. Age-related variations in the fatal crash involvement rates resulted primarily from the differences in crash incidence density. The results indicate that, despite having lower fatal crash involvement rates, female drivers do not seem to be safer than their male counterparts when exposure is considered. The decomposition method is valuable as both a conceptual framework and an exploratory tool for understanding the contributing factors related to cause-specific injury mortality and the differences in death rates among populations.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Accidents, Traffic / mortality*
  • Adolescent
  • Adult
  • Age Distribution
  • Aged
  • Automobile Driving / statistics & numerical data*
  • Cross-Sectional Studies
  • Databases, Factual / statistics & numerical data
  • Epidemiologic Methods
  • Female
  • Humans
  • Incidence
  • Male
  • Middle Aged
  • Odds Ratio
  • Prevalence
  • Retrospective Studies
  • Sex Factors
  • Statistics, Nonparametric
  • Time Factors
  • Travel / statistics & numerical data
  • United States / epidemiology