Voas Robert B
Pacific Institute for Research and Evaluation, Calverton, Maryland, USA.
Traffic Inj Prev. 2008 Mar;9(1):22-30. doi: 10.1080/15389580701682114.
Currently, the implementation of sobriety checkpoint programs, which have been demonstrated to be effective in reducing alcohol-related crashes, is limited by the belief that they require large consignments of police officers and result in few arrests. However, one of the earliest evaluations of a checkpoint program in Charlottesville, Virginia, demonstrated that effective checkpoints could be mounted in which police officers made as many arrests as officers on regular patrols. That study was printed by the NHTSA but was not published in a peer-reviewed journal. Because of its significance to current issues in the staffing of and procedures for checkpoint operations, this article reanalyzes the results of that study and describes the procedures implemented in checkpoints.
A before-and-after control design was used to measure the change in nighttime crashes from three baseline years to the program year. Two analyses were conducted: the first on the percentage of all crashes occurring at night in the test city--Charlottesville--and the second on the percentage of all nighttime crashes in the state of Virginia that occurred in the test city. In addition, three waves of random-digit-dialing telephone surveys were conducted: one before and two during the checkpoint program in the test city, and the comparison city, Blacksburg. Finally, the number of impaired-driving arrests per officer hour at the checkpoints was compared with the number of arrests per hour by officers on regular patrol and the effect on arrests of the use of passive sensors was determined.
The monthly percentage of nighttime crashes in Charlottesville was reduced by 17% (p = 000) in relation to the baseline level. The percentage of nighttime crashes in the state of Virginia that occurred in Charlottesville was reduced by 11% (p = .013) from baseline levels. Drivers arrested at checkpoints had lower BACs than those arrested by the regular patrols; however, the conviction rates were the same. The arrest per officer hour did not differ significantly between the two types of enforcement operations. Awareness of the checkpoint activity was high (72%) among nighttime at-risk drivers in the test city. Half reported seeing a checkpoint operation, and a quarter reported being interviewed. Use of a passive alcohol sensor by officers at the checkpoint increased arrests by almost a factor of three.
The results of the evaluation suggest that small-scale sobriety checkpoints can be implemented as part of the regular enforcement program in moderate-sized jurisdictions and that they can be as efficient in producing arrests as standard enforcement patrols, particularly if passive alcohol sensors are used.
目前，清醒检查站计划已被证明在减少与酒精相关的撞车事故方面有效，但其实施受到限制，因为人们认为该计划需要大量警察，且逮捕人数较少。然而，弗吉尼亚州夏洛茨维尔市对一个检查站计划的最早评估之一表明，可以设置有效的检查站，在这些检查站执勤的警察逮捕人数与常规巡逻警察一样多。该研究由美国国家公路交通安全管理局印发，但未发表在同行评审期刊上。鉴于其对当前检查站行动的人员配备和程序问题的重要性，本文重新分析了该研究的结果，并描述了检查站实施的程序。
采用前后对照设计来衡量从三个基线年份到计划年份夜间撞车事故的变化。进行了两项分析：第一项分析测试城市夏洛茨维尔市夜间发生的所有撞车事故的百分比，第二项分析弗吉尼亚州在测试城市发生的所有夜间撞车事故的百分比。此外，还进行了三轮随机数字拨号电话调查：一轮在测试城市和对照城市布莱克斯堡的检查站计划之前，两轮在计划期间。最后，比较了检查站每执勤小时的酒驾逮捕人数与常规巡逻警察每小时的逮捕人数，并确定了使用被动传感器对逮捕人数的影响。
夏洛茨维尔市夜间撞车事故的月度百分比相对于基线水平降低了17%（p = 0.000）。弗吉尼亚州在夏洛茨维尔市发生的夜间撞车事故的百分比相对于基线水平降低了11%（p = 0.013）。在检查站被捕的司机的血液酒精浓度低于被常规巡逻逮捕的司机；然而，定罪率相同。两种执法行动每执勤小时的逮捕人数没有显著差异。测试城市夜间高危司机对检查站活动的知晓率很高（72%）。一半的人报告看到过检查站行动，四分之一的人报告接受过询问。检查站警察使用被动酒精传感器使逮捕人数增加了近两倍。
评估结果表明，小规模清醒检查站可以作为中等规模辖区常规执法计划的一部分实施，并且在逮捕方面与标准执法巡逻一样高效，特别是如果使用被动酒精传感器的话。
