School of Social Work and Population Research Center, University of Texas at Austin, Austin, Texas 78712-0358, USA.
J Adolesc Health. 2010 Oct;47(4):399-406. doi: 10.1016/j.jadohealth.2010.02.010. Epub 2010 Apr 28.
Cross-sectional research has demonstrated associations between neighborhood socioeconomic characteristics and adolescents' initiation of heterosexual intercourse. Prospective designs are needed to move toward causal inference, and to identify mediating and moderating influences to inform policies and programs.
Among 5,838 nonsexually active participants in wave I (1994-1995) of the National Longitudinal Study of Adolescent Health, multilevel logistic regression analyses were used to predict initiation of sex by wave II (1996); models were stratified by gender and age group (11-14 and 15-17 yr). Predictors measured at wave I included census tract-level (neighborhood) poverty concentration, family-level income, parental education, race/ethnicity, and family structure. Adolescent college aspirations and life expectations were tested as possible mediators or moderators of the neighborhood poverty-sexual initiation association.
Neighborhood poverty concentration predicted older (15-17 yr) girls' and boys' sexual initiation, after considering individual-level covariates. However, adolescent college aspirations and life expectations were not found to mediate the prediction relationship. Moderating effects were identified for girls (college aspirations) and boys (positive life expectations) in high-poverty neighborhoods, paradoxically reflecting increased risk.
In this longitudinal study, moderating effects generally considered protective against sexual initiation were not protective or were harmful for adolescents living in high-poverty neighborhoods. Subsequent research to understand how to reduce the health risks of living in poor neighborhoods must examine an even wider range of variables and/or use different methodologies.
横断面研究已经证明了邻里社会经济特征与青少年开始异性性行为之间的关联。需要进行前瞻性设计，以推动因果推理，并确定中介和调节影响，为政策和计划提供信息。
在全国青少年健康纵向研究的第一波（1994-1995 年）中，有 5838 名非活跃参与者，使用多层次逻辑回归分析预测第二波（1996 年）的性行为开始；根据性别和年龄组（11-14 岁和 15-17 岁）对模型进行分层。在第一波测量的预测因素包括普查区层面（邻里）贫困集中程度、家庭层面收入、父母教育程度、种族/族裔和家庭结构。青少年对大学的期望和生活期望被测试为邻里贫困与性开始关联的可能中介或调节因素。
在考虑个体水平的协变量后，邻里贫困集中程度预测了年龄较大（15-17 岁）的女孩和男孩的性开始。然而，青少年对大学的期望和生活期望并没有被发现介导预测关系。在高贫困社区中，女孩（对大学的期望）和男孩（积极的生活期望）的调节作用得到了确定，这反映出风险增加的悖论。
在这项纵向研究中，一般被认为可以预防性行为开始的调节作用对于生活在贫困社区的青少年来说并没有起到保护作用，或者是有害的。为了了解如何降低生活在贫困社区的健康风险，后续研究必须检查更广泛的变量范围，或者使用不同的方法。
