Racial-ethnic disparities in adolescent sexual risk behavior are associated with health disparities during adulthood and are therefore important to understand. Some scholars argue that neighborhood disadvantage induces disparities, yet prior research is mixed. We extend neighborhood-effects research by addressing long-term exposure to neighborhood disadvantage and estimation bias resulting from inclusion of time-varying covariates. Drawing from the Fragile Families and Child Well-Being Study, we compare a point-in-time proximal measure of neighborhood disadvantage with a duration-weighted measure using marginal structural models with inverse probability of treatment weights. Findings indicate that multiracial, non-Hispanic black, and Hispanic youth exhibit significantly higher sexual risk and duration-weighted exposure to neighborhood disadvantage than non-Hispanic white adolescents. Duration-weighted exposure is a better predictor of sexual initiation and number of partners by age 15 than a point-in-time proximal measure of neighborhood disadvantage and accounts for a substantial portion of the race-ethnic differences in sexual risk.

青少年性风险行为中的种族差异与成年期的健康差异有关，因此了解这一点很重要。一些学者认为，邻里劣势会导致差异，但之前的研究结果好坏参半。我们通过解决因包含时变协变量而导致的邻里劣势和估计偏差的长期暴露来扩展邻里效应研究。借鉴脆弱家庭和儿童福利研究，我们将邻里劣势的时间点近端测量与使用具有治疗权重逆概率的边际结构模型的持续时间加权测量进行比较。调查结果表明，多种族、非西班牙裔黑人、与非西班牙裔白人青少年相比，西班牙裔青年表现出更高的性风险和持续时间加权暴露于邻里劣势。与邻里劣势的时间点近端测量相比，持续时间加权暴露可以更好地预测 15 岁时的性开始和伴侣数量，并且占性风险中种族-民族差异的很大一部分。