Published by Oxford University Press on behalf of the International Epidemiological Association © The Author 2009; all rights reserved.
Commentary: Causal pathways of relative motor vehicle crash fatality risk are hard to estimate from police records
Senior Statistician, Westat, Inc., 1650 Research Blvd, Rockville, MD 20815, USA. E-mail: PaulZador@Westat.com
Accepted 12 May 2009
| The first 10% of the full text of this article appears below. |
Lardelli-Claret et al.1 estimated relative driver and front seat (FS) passenger fatality risks as a function of age, sex and belt use (ASB). Four sets of relative fatality risk (RFR) estimates were derived from Poisson regression models based on Spanish traffic crash registry data for the years 2000–04 (N = 84 338 FS occupant pairs). Each of the models related the probability of an FS driver (Model 1), an FS passenger (Models 2 and 3) or an FS occupant (Model 4) fatality to some combination of covariates for occupant, vehicle, crash and environmental factors. The covariates varied by model. Depending on model specification, the RFR estimates of ASB were conceptualized as accounting for the