Article Text

Download PDFPDF

Very young drivers in fatal vehicle crashes involving police chases
  1. A Plessinger,
  2. L Frisch
  1. Department of Community Health Sciences, Northeastern Ohio Universities College of Medicine, Rootstown, OH, USA
  1. Correspondence to:
 Dr L Frisch
 Northeastern Ohio Universities College of Medicine, PO Box 95, Rootstown, OH 44272, USA; lfrisch{at}

Statistics from

Motor vehicle crashes associated with high speed police pursuits have received recent attention in both the public health literature and the lay press.1–4 Each year in the US there are approximately 350 fatalities attributable to police pursuits.5 In 2003, 7000 high speed chases occurred in California alone, resulting in 51 deaths.6 Data are more difficult to obtain for countries other than the US, but a recent report by the Police Federation of England and Wales suggested that pursuit related deaths in Britain tripled between 1997 and 2003.7 Though funded by the US Department of Justice, a 1996 publication by the International Association of Chiefs of Police attempted to establish an international model pursuit policy, suggesting that the importance of this problem is widely recognized.8

Drivers under 16 years of age are involved in over 200 US fatal crashes yearly.9–11 However, no previous study has addressed the involvement of very young, typically unlicensed, drivers in fatal police pursuits. The purpose of this paper is to explore the role of such very young drivers in fatal crashes related to police chases.

Methods and materials

This study is based on the US Fatality Analysis Reporting System (FARS).5 FARS is a database of all US fatal motor vehicle crashes occurring on public roads that has been maintained by the National Highway Traffic Safety Administration since 1975. For each crash in which at least one fatality occurs, FARS data are collected from official state sources on all involved people, vehicles, and crash circumstances using standardized reporting forms. For this study, all 1999–2003 FARS cases were selected in which at least one driver was age 15 or less and police pursuit was listed as a contributing factor. Details of each crash were assessed including driver’s sex, number of vehicles and passengers involved, vehicle type, age of all decedents, state and county of crash, crash related factors (three factors), drivers’ licensure, road type and condition, time of accident, and the role of alcohol or drugs. A crash was judged to be related to a police chase if one of the three crash related factors fields contained the entry “police pursuit involved”. All counties in which crashes occurred were classified for rurality according to the Department of Agriculture (USDA) rural-urban continuum codes (RUCC).8 Analysis employed Microsoft Access 2002 and SPSS for Windows version 12.0.1. All data were obtained from fully public sources, and for this reason ethics committee review was not sought.


Over the five year period, there were 49 fatal pursuit related crashes in which the driver was 15 years of age or younger. Drivers’ ages ranged from 9–15 years, with 14 and 15 year olds representing nearly 90% (n = 44) of these crashes. The remaining five drivers were 9, 11, 12, 13, and 13 years of age. Seventy nine percent (n = 39) of drivers were male, and the majority of vehicles involved (76%; n = 37) were passenger cars. Alcohol or drug use contributed to three crashes (6%), though this may be an underestimate because of problems with alcohol and drug data in FARS.5 Seatbelt non-use contributed to fatality in 76% (n = 37) of crashes. Airbags were absent or non-functional in 26 crashes (53%). None of the young drivers involved was validly licensed or driving with a learner’s permit at the time of the crash. Female drivers are overrepresented in this sample when compared with FARS reported crashes involving drivers age 18 years and older (Fisher’s exact t test, p = 0.0066).

These crashes resulted in 69 deaths: 28 drivers age 15 years and younger, 24 other passengers 18 years of age and younger, three adult passengers riding in the fleeing vehicle, and 14 “bystander” drivers, passengers, or pedestrians. Four states (California, Michigan, Florida, and Texas) accounted for nearly half of all crashes (n = 22), a rate disproportionate to their 28% share of the under 18 population. Forty four (90%) of these crashes occurred in metropolitan areas. This is in sharp contrast to the overall pattern of crashes involving young drivers; these are twice as likely to occur in rural as in metro or metro-adjacent counties.10,11


This report is highly dependent on the accuracy and completeness of FARS data on “crash related factors” for which 16 response categories are provided, including “unknown” and “police pursuit involved”. In the 2003 FARS dataset of all crashes, only 5% have an entry in any of the three crash related factors fields. It is possible that some of these crashes may have had unrecorded factors, including police chases, contribute to their fatal outcomes. We were unable to cross check FARS entries against actual police reports, and are unaware of published independent assessments of FARS crash related factor reporting accuracy.

With this important limitation in mind, we have shown that very young, unlicensed drivers are involved in approximately 3% of all fatal crashes involving police pursuit. These crashes tend to occur in metropolitan areas of relatively populous states and are most common in California where police pursuit has become a political and media concern. Although the majority of drivers are boys, girls are significantly overrepresented compared with pursuit related fatal crashes involving older drivers. As few of the drivers pursued were guilty of a violent offence,12 it is difficult to defend the risk involved in the pursuit of untrained and inexperienced drivers. Although we are confident that police pursuits occur outside of the US, we were unable to find crash epidemiology data for other countries. However, young driver “joy riding” has been reported in Europe, often involving high speed operation of stolen vehicles.13,14 Joy riding might be expected to result in police pursuit when this intervention is permitted by police authorities. Limited or judicious authorization of police chases involving fleeing drivers might be expected to reduce deaths attributable to very young, unlicensed driving.



  • This work was supported by a grant made to Northeastern Ohio Universities College of Medicine by Dr and Mrs Robert Blacklow.

  • Competing interests: none.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.