Multilevel risk profiles and reproductive health across adolescence and young adulthood.

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ABSTRACT Sexual and reproductive health continues to be a public health concern. In this United States, rates of sexually transmitted infections and unintended pregnancy are highest among adolescents and young adults, making prevention programs to this group to be particularly important. Longitudinal data now exists that can give greater insight into the factors that underlie sexual and reproductive health outcomes develop across adolescence and young adulthood, but new analytic methods are necessary to unlock the full potential of this data. We propose the use and integration of two innovative analytic methods, the time-varying effect modeling (TVEM) and latent class analysis (LCA) to long-term longitudinal data from the National Longitudinal Study of Adolescent to Adult Health (Add Health). This project has three specific aims. The first aim will elucidate age trends in SRH outcomes (e.g., sexual risk behaviors, STIs, unintended pregnancy) across adolescence through young adulthood at the population level. We will examine how these age-varying trends differ by demographic subgroups (biological sex, race/ethnicity, and sexual minority status). The second aim will examine how profiles of multilevel early risk factors (e.g., parent, peer, neighborhood) predict sexual and reproductive health outcomes across adolescence through young adulthood, and how these age-varying associations differ by demographic subgroup. Finally, the third aim will examine how age-varying individual factors differentially predict SRH outcomes across adolescence and young adulthood, and how these associations differ by demographic subgroups and membership in the early risk profile classes determined in Aim 2. This project will provide population-level knowledge on the developmental course of SRH outcomes, processes which underlie these outcomes, and how they differ by subgroup. Results will allow prevention scientists to design interventions targeting the most relevant risk factors at particular ages for specific subgroups. To maximize the project’s overall impact, published manuscripts will include details of the statistical models and programming syntax so that researchers can use these approaches to answer new questions about age-varying processes.