Shakya Holly B, Perkins Jessica M, Traeger Margaret, Tsai Alexander C, Bangsberg David R, Kakuhikire Bernard, Christakis Nicholas A
Center on Gender Equity and Health, School of Medicine, University of California San Diego, San Diego, CA, United States.
Department of Human and Organizational Development Peabody College, Vanderbilt University, PMB 90, 230 Appleton Place, Nashville, TN 37203, United States.
SSM Popul Health. 2018 Feb 6;4:236-243. doi: 10.1016/j.ssmph.2018.02.001. eCollection 2018 Apr.
We investigated the household-level social network correlates of acceptance of intimate partner violence (IPV) in rural, agrarian settings of Honduras and Uganda, two low-income countries with unequal access to resources based upon gender. We collected complete social network data in each location (Honduras in 2014 and Uganda in 2012), across a diverse range of relationships, and then created a measure of household cohesion by calculating the degree to which members of a household nominated each other as important social connections. Our measure of IPV acceptance was based on 4 questions from the Demographic Health Survey to assess the conditions under which a person believes that it is acceptable for a man to perpetrate physical violence against his wife or partner and we coded a person as positive on IPV acceptance if they answered positively to any of the four questions. We used logistic regression to calculate the odds that an individual accepted IPV given (1) household level cohesion and (2) the proportion of the household that accepts IPV. We found individuals from more cohesive households were less likely to accept IPV controlling for the overall level of IPV acceptance in the household. Nevertheless, those in households more accepting of IPV were more likely to personally accept IPV. In stratified analyses, when household IPV acceptance was especially high, the benefit of household cohesion with respect to IPV was attenuated. The design and implementation of interventions to prevent IPV should consider household structure and norms rather than focusing only on individuals or couples.
我们调查了洪都拉斯和乌干达农村农业地区亲密伴侣暴力（IPV）接受度的家庭层面社会网络关联因素，这两个低收入国家在资源获取方面存在基于性别的不平等。我们在每个地点（2014年在洪都拉斯，2012年在乌干达）收集了涵盖各种关系的完整社会网络数据，然后通过计算家庭成员将彼此提名视为重要社会关系的程度来创建家庭凝聚力指标。我们对IPV接受度的衡量基于人口健康调查中的4个问题，以评估一个人认为男性对其妻子或伴侣实施身体暴力在何种情况下是可以接受的，如果他们对这四个问题中的任何一个回答为肯定，我们就将其编码为对IPV接受度呈阳性。我们使用逻辑回归来计算在给定（1）家庭层面凝聚力和（2）家庭中接受IPV的比例的情况下，个体接受IPV的几率。我们发现，在控制家庭中IPV接受度总体水平的情况下，来自凝聚力更强家庭的个体接受IPV的可能性较小。然而，那些所在家庭对IPV接受度更高的人个人接受IPV的可能性更大。在分层分析中，当家庭对IPV的接受度特别高时，家庭凝聚力在IPV方面的益处就会减弱。预防IPV干预措施的设计和实施应考虑家庭结构和规范，而不是只关注个人或夫妻。
