Analysis of precipitation-related motor vehicle collision and injury risk using insurance and police record information for Winnipeg, Canada

J Safety Res. 2011 Oct;42(5):383-90. doi: 10.1016/j.jsr.2011.08.004. Epub 2011 Oct 11.


Introduction: Police records are the most common source of data used to estimate motor-vehicle collision risks, understand causal or contributing factors, and evaluate the efficacy of interventions. The literature notes concerns about this information citing discrepancies between police reports and other sources of injury occurrence and severity data. The primary objective of the analysis was to assess the adequacy of police reports for an examination of weather-related injury collision risk.

Method: Analyses of relative risk were carried out using both police records and comprehensive insurance claim data for Winnipeg, Canada over the period 1999-2001.

Results and conclusions: Both data sets yielded very similar results-precipitation substantially increases the risk of injury collision (police records: RR 1.76, CI 1.55-2.00; insurance: RR 1.80, CI 1.62-1.99) and risk of injury (police records, RR 1.74, CI 1.55-1.96; insurance, RR 1.69, CI 1.55-1.85) relative to corresponding dry weather control periods. Both rainfall and snowfall were associated with large increases in collisions and injuries.

Impact on industry: While relative risks are almost identical, over 64% more injury collisions and 74% more injuries were identified using the insurance data, which is an important difference for evaluating absolute risk and exposure.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Accidents, Traffic / economics
  • Accidents, Traffic / statistics & numerical data*
  • Automobile Driving / statistics & numerical data*
  • Automobiles / economics
  • Automobiles / statistics & numerical data
  • Confidence Intervals
  • Humans
  • Insurance, Accident / economics
  • Insurance, Accident / statistics & numerical data
  • Manitoba / epidemiology
  • Police / statistics & numerical data
  • Rain*
  • Risk
  • Risk Assessment / methods
  • Snow*
  • Visual Perception / physiology*
  • Wounds and Injuries / economics
  • Wounds and Injuries / epidemiology*