Association of seat belt use with death: a comparison of estimates based on data from police and estimates based on data from trained crash investigators

Inj Prev. 2002 Dec;8(4):338-41. doi: 10.1136/ip.8.4.338.


Objective: Estimates of any protective effect of seat belts could be exaggerated if some crash survivors falsely claimed to police that they were belted in order to avoid a fine. The aim of this study was to determine whether estimates of seat belt effectiveness differed when based on belt use as recorded by the police and belt use determined by trained crash investigators.

Design: Matched cohort study.

Setting: United States.

Subjects: Adult driver-passenger pairs in the same vehicle with at least one death (n=1689) sampled from crashes during 1988-2000; data from the National Accident Sampling System Crashworthiness Data System.

Main outcome measure: Risk ratio for death among belted occupants compared with those not belted.

Results: Trained investigators determined post-crash seat belt use by vehicle inspections for 92% of the occupants, confidential interviews with survivors for 5%, and medical or autopsy reports for 3%. Using this information, the adjusted risk ratio for belted persons was 0.36 (95% confidence interval 0.29 to 0.46). The risk ratio was also 0.36 using police reported belt use for the same crashes.

Conclusions: Estimates of seat belt effects based upon police data were not substantially different from estimates which used data obtained by trained crash investigators who were not police officers. These results were from vehicles in which at least one front seat occupant died; these findings may not apply to estimates which use data from crashes without a death.

Publication types

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

MeSH terms

  • Accidents, Traffic / mortality*
  • Adolescent
  • Adult
  • Bias
  • Cohort Studies
  • Female
  • Humans
  • Male
  • Middle Aged
  • Police / statistics & numerical data
  • Risk Factors
  • Seat Belts / statistics & numerical data*
  • United States / epidemiology