Problem driver remediation: a meta-analysis of the driver improvement literature

J Safety Res. 2004;35(4):403-25. doi: 10.1016/j.jsr.2004.06.002.

Abstract

Problem: Given the public safety risk posed by violation and crash repeaters and the substantial costs for state driver improvement programs, it is important that their effectiveness be scientifically demonstrated and that intervention programs are based on sound research findings.

Method: Crash and traffic violation standardized effect sizes (d) representing 106 individual interventions were coded from 35 methodologically sound studies and analyzed using meta-analysis.

Results: Driver improvement intervention in general was associated with small but significant reductions in both crashes (dw = 0.03) and violations (dw = 0.06). Significant effects were found on both measures for warning letters, group meetings, individual hearings, and license suspense/revocation. Of the driver improvement interventions studied, license suspension/revocation was by far the most effective treatment for both crashes and violations (dw = 0.11 and 0.19). Since one of the objectives of license suspension/revocation is to eliminate driving for the period of suspension, it is possible that much or all of the effect is due to reduced exposure and/or more careful driving during the suspension interval. Results were mixed for other types of interventions, although distributing educational or informational material was not associated with any reductions. Interventions associated with violation reduction tended to also be associated with crash reduction, although the relationship was not very strong (r = .30).

Discussion: Although interpretation of the effect size estimates was complicated by almost ubiquitous heterogeneity, the results do suggest an overall positive impact of driver improvement interventions in general.

Impact on industry: The results support the continued use of driver improvement interventions, chiefly warning letters, group meetings, individual hearings, and especially license suspension/revocation. The results also suggest that court-triggered traffic violator programs are less effective than interventions triggered by drivers license agencies.

Publication types

  • Meta-Analysis
  • Research Support, Non-U.S. Gov't
  • Review

MeSH terms

  • Accidents, Traffic / prevention & control*
  • Accidents, Traffic / statistics & numerical data*
  • Automobile Driving / education*
  • Automobile Driving / legislation & jurisprudence
  • Automobile Driving / statistics & numerical data*
  • Behavior Therapy / statistics & numerical data
  • Health Education / statistics & numerical data
  • Health Knowledge, Attitudes, Practice
  • Humans
  • Remedial Teaching / statistics & numerical data
  • United States