Collision and violation involvement of drivers who use cellular telephones

Traffic Inj Prev. 2003 Mar;4(1):45-52. doi: 10.1080/15389580309851.


The study sample consisted of 3,869 drivers, split approximately 50/50 between observed cell phone users and those observed not using cell phones (labeled "nonusers"). Cell phone use was determined by a snapshot observation made on city streets. The sample represented 54% of those originally observed, for whom a match was obtained for both vehicle license plate and for gender and estimated age group of the observed driver and that of the driver named in the vehicle policy. Data were obtained from records of insurance claims, police-reported collisions and violations, following a strict protocol to protect individual privacy. The dependent measures were at-fault crash claims and "inattention" violations. A logistic regression model controlled for age, gender, exposure (represented by not-at-fault crash claims), alcohol-related offenses, and aggressive driving offenses. The study also involved a comparison of the contributing factors and collision configurations of police-reported collisions involving the users and "nonusers" in the sample. Drivers observed using cell phones had a higher risk of an at-fault crash than did the "nonusers," although the difference was not significant for males. There was no apparent effect on "inattention" violations. The cell phone users also had a higher proportion of rear-end collisions. The violation pattern of cell phone users suggests that they are, in general, riskier drivers. These differences likely reflect lifestyle, attitude and personality factors. It is essential to control for these factors in assessing the direct risk attributable to cellular telephone use.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Accidents, Traffic / statistics & numerical data*
  • Adolescent
  • Adult
  • Age Distribution
  • Aggression
  • Alcohol Drinking / epidemiology
  • Attention
  • Automobile Driving / statistics & numerical data*
  • British Columbia / epidemiology
  • Cell Phone / statistics & numerical data*
  • Factor Analysis, Statistical
  • Female
  • Humans
  • Insurance / statistics & numerical data
  • Law Enforcement*
  • Logistic Models
  • Male
  • Middle Aged
  • Probability
  • Risk Assessment / methods
  • Sex Distribution