Safer Bars: A Cluster-Randomized Effectiveness Evaluation of Alcohol-related Sexual Violence Prevention through Bar Staff Bystander Training

摘要 背景和意义：性暴力的发生与社会背景密切相关， 支持高水平的酒精消费。从2007年开始，亚利桑那州卫生服务部 （ADHS）在疾病预防控制中心强奸预防和 教育（RPE）资助计划。更安全的酒吧是一个旁观者预防培训的工作人员的酒精服务 机构。本申请提出了一个理论驱动的，准实验的有效性评估， 亚利桑那大学的独立团队。我们的方法很重要，因为很少有评价 测量的行为结果以及解释观察到的旁观者变化的介质 行为，或与此类干预相关的社区水平结果。具体目标：具体目标 （1）测试一个假设的中介模型的个人层面的行为结果的基础上， 计划行为理论：（2）评估该计划对酒吧级风险管理安全政策的影响; 以及（3）检查更安全的酒吧计划对减少警察人数的社区影响 在酒精供应场所密度高的地区进行调度。方法：随机整群试验 交叉设计，将标准水平的参与者随机分为干预组和延迟干预组 本文提出评估是在基线、培训完成以及培训后三个月和六个月进行的。 将利用地理信息系统绘制的“热点”图来选择样条， 警方在亚利桑那州三所主要大学的三英里半径内派遣，所有这些大学都有很高的 校园附近的酒精供应场所的密度。ADHS将把他们的计划交付集中在酒吧 在这些地理参数中。我们的分析模型将评估一个假设的温和的 计划行为理论（Theory of Planned Behavior，TPB）。有关成果包括：（a） TPB的核心结构行为意图在多大程度上介导了酒吧员工的预期绩效 三个月随访时的旁观者行为以及这种关系在多大程度上受到以下因素的调节： 个人特征和酒吧环境;（B）该计划对酒吧级风险管理的影响 （c）社区层面的影响，如警方在2009年派遣的警察人数所显示的。 在基线和项目结束时，使用地理信息系统，酒精服务设施密度高的地区 数据有证据表明，安全棒的疗效是可行的，可扩大规模，而不会造成过度的财政负担，在其他 这些国家都获得了RPE的资助。

Abstract Background and Significance: Sexual violence perpetration is robustly associated with social contexts that support high levels of alcohol consumption. Beginning in 2007, the Arizona Department of Health Services (ADHS) developed and field-tested the Safer Bars curriculum with support from the CDC Rape Prevention and Education (RPE) funding program. Safer Bars is a bystander prevention training for staff of alcohol serving establishments. This application proposes a theory-driven, quasi-experimental effectiveness evaluation by an independent team at The University of Arizona. Our approach is significant because few evaluations have measured behavioral outcomes as well as the mediators that explain the observed changes in bystander behavior, or community-level outcomes associated with such interventions. Specific Aims: the specific aims are to: (1) Test a hypothesized mediational model of individual-level behavioral outcomes based on the Theory of Planned Behavior; (2) assess the effect of the program on bar-level risk management safety policies; and (3) examine the community-level impact of the Safer Bars program on reducing the number of police dispatches in areas with high densities of alcohol-serving establishments. Methods: A cluster-randomized trial design with crossover to randomize participants at the bar level into intervention and delayed intervention arms is proposed. Assessments are at baseline, completion of training, and at three and six months post-training. The sample bars will be selected using Geographical Information System (GIS) mapping of “hot spots” for police dispatches within a three-mile radius of the three major Arizona universities, all of which have a high density of alcohol-serving establishments adjacent to campus. ADHS will focus their program delivery on bars within these identified geographic parameters. Our analytical model will assess a hypothesized moderated mediational model based in the Theory of Planned Behavior (TPB). Outcomes of interest include: (a) the extent to which Behavioral Intent, the central construct of TPB, mediates bar staff performance of intended bystander behaviors at three-month follow-up and the extent to which this relationship is moderated by personal characteristics and bar environment; (b) the effects of the program on bar-level risk management patron safety policies; and, (c) the community-level effects as demonstrated by number of police dispatches in areas with a high density of alcohol-serving establishments at baseline and at the end of the project using GIS data. With evidence of efficacy, Safer Bars is feasible for scale-up without excessive financial burden in other states, which all receive RPE funding.