Driving vehicles after drinking alcohol is common, and has been shown to increase the risk of car crashes.1 There is a strong community perception that increased police patrols are effective in reducing dangerous driving practices such as drinkdriving. A recently published review on the Cochrane Database of Systematic Reviews has examined the effectiveness of policepatrols for reducing crashes and crash-related injuries.Police patrols were the main intervention examined in the review, and were defined as an increase in the number of officers or in the frequency and duration of patrols, with the intention of identifying impaired drivers via behavioral cues. Most included studies (91%) assessed increased patrols in combination with other programs such as media campaigns or special training for police officers. Therefore, it was generally not possible to evaluate the independent effect of each element of the intervention. As the interventions were implemented in the community, studies designed to evaluate effectiveness at a community level were included in the review. Study designs included randomized trials, interrupted time series, and controlled before-and-after studies. The authors found many methodological problems with the included studies, including lack of detail, which made quality assessment difficult. Of the 32 eligible studies, two-thirds scored "not adequate" on at leastone feature for assessing methodological quality.uAlthough the studies included in the review were generally consistent in showing positive effectson traffic crashes and deaths, these effects were not always statistically significant at the 95% confidence level, particularly for studies examining fatal crashes. However, 13 of 20 studies showed reductions in total crashes, and two-thirds of these reductions were statistically significant. The review authors concluded that, although the effects were reasonably consistent across studies, methodological weaknesses (including inadequate sample size, failure to match on baseline measures, contamination, and inadequate data analysis) precluded making firm conclusions about the effectiveness of increased police patrols, whether or not the patrols were in combination with other interventions. The review authors commented on the needfor high-quality evaluations of interventions. Substantial resources are spent on crash-reduction programs despite their unproven efficacy. Therefore, government agencies should be encouraged and supported in the design, conduct, and publication of rigorous evaluations for implemented programs.(Author/publisher).
Samenvatting