Evaluating the effects of state-level school-based sex education policies and practices on teenage pregnancy prevention

Project Details


This proposal addresses the following area of investigation: 'Identifying factors that improve the quality, access, and equity of pregnancy prevention programs for adolescents or young adults, or that reduce existing disparities.' School-based sex education has the potential to promote adolescent sexual and reproductive health by giving large numbers of youth the information and skills to prevent sexually transmitted infections (STIs) and teenage pregnancy. Policies related to school-based sex education are often formulated at the state level and may impact the effectiveness of teenage pregnancy prevention efforts. There is some evidence that variations in the content of these policies affect sexual risk behaviors and subsequent teenage birth and sexually transmitted infection (STI) rates. However, there is a dearth of research assessing how these policies are put into practice in schools, and how these practices are in turn related to sexual behaviors, health outcomes, and disparities. The overarching goal of this project is to longitudinally evaluate the effects of state-level school-based sex education policies and practices on adolescent sexual risk behaviors and sexual health outcomes. To achieve this goal, we will conduct secondary data analyses using a novel combination and integration of publicly available reports and datasets merging existing state-level, school-level, and individual-level data from 2001 to the present. State policy variables from the Guttmacher Institute and the Sexuality Information and Education Council of the United States will include sex education mandates and content requirements. Indicators of school sex education practices from the Centers for Disease Control and Prevention (CDC) School Health Profiles will include topics taught such as condom efficacy and how to obtain contraception as well as sex education topics on which health education teachers have received professional development such as pregnancy preve ntion. Variables from the CDC High School Youth Risk Behavior Survey will include students' self-reports of ever having sex, age at first sex, use of condoms/contraception/alcohol/drugs at last sex, number of partners, ever testing for HIV, ever getting pregnant/getting someone pregnant, and ever being diagnosed with an STI. Sexual health outcomes from the CDC AtlasPlus and vital statistics data include rates of chlamydia, gonorrhea, syphilis, and births among 15-19-year-olds. The behavior and outcomes measures will be broken down by gender and racial, ethnic, and sexual minority status when possible to identify effects of policies and practices on those who have suffered historic disparities. This harmonized dataset will be used to test our own hypotheses and will be made available to other researchers. Our longitudinal analyses will include a) difference-in-differences approaches to compare each state to itself and other states prior to and after policy changes; b) structural equation modeling to examine whether sex education practices mediate the relationship between sex education policies and youth sexual risk behaviors and adverse sexual health outcomes; and c) multilevel modeling to assess the relationships between state-level policies and outcomes, school-level practices, and individual-level behaviors. Results of the project will have a positive impact on the field of teenage pregnancy prevention by providing evidence that state education policymakers, school districts, principals, and teachers can use as they deliberate how best to teach sex education. As improved and evidence-based school-based sex education policies and practices are enacted, adolescents may likewise make healthy behavior choices and benefit from reduced disparities in teenage pregnancies and other adverse sexual health outcomes.

Effective start/end date1/1/22 → …


  • DHHS Office of the Secretary


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.