Evaluation of State-Level Sexual and Gender Minority Laws for the Primary Prevention of Sexual and Intimate Partner Violence Among Sexual and Gender Minority Youth in the U.S.

Quantitative Component (Aims 1-2): SM youth are identified using the YRBSS sexual identity item, which asks respondents "Which of the following best describes you?" Respondents indicating gay, lesbian, bisexual, describing their sexual identity some other way, or questioning are categorized as SM. Respondents indicating heterosexual (straight) are categorized as non-SM. Primary outcomes are three individual-level YRBSS items measuring past 12-month sexual violence (SV) victimization and past 12-month intimate partner violence (IPV) victimization (physical and sexual). Policy exposures are constructed as state-level binary indicators (present versus absent) for three types of SGM laws: gender-affirming care laws, conversion therapy bans, and enumerated identity-based anti-bullying laws. Exposure data are drawn from legal epidemiology datasets available through LawAtlas.org for years 2016, 2018, 2020, 2022, and 2024 with a one-year lag between exposure measurement and outcome assessment to ensure policies are implemented before outcome measurement. Outcome data are drawn from YRBSS survey including 29 states for the years 2017, 2019, 2021, 2023, 2025 representing 98 state-years with a total sample of approximately 357,040 youth. Aim 1 employs a difference-in-differences (DiD) approach in a quasi-experimental framework using mixed effects DiD Poisson models to estimate the effect of SGM law adoption on SV/IPV risk among SM youth. Separate models are estimated for each law type and each outcome. Time-varying state-level covariates include population size, percentage below poverty line, and state-level demographic composition of race/ethnicity and age. Sensitivity analyses include Goodman-Bacon decomposition, doubly robust estimators, and imputation-based estimators to address potential treatment effect heterogeneity across adoption timing. Parallel pre-trends are assessed using lag and lead models with bounds adjusted for any non-parallel pre-trends. Exploratory analyses examine effects among specific racial/ethnic SM youth subgroups (Black, Hispanic/Latino, Native American/Alaska Native, and mixed race SM youth), reporting point estimates with 95 percent confidence intervals without formal hypothesis testing to provide directional evidence. Aim 2 builds on the Aim 1 models by adding an interaction term between state-level SGM law adoption and social climate to test whether policy effects on SV/IPV risk differ based on state-level stigma toward SGM populations. Social climate is measured using publicly available survey data from the Public Religion Research Institute (PRRI) American Values Atlas collected from a nationally representative sample of approximately 22,000 adults annually from all 50 states 2016-2024. Respondents are asked the degree to which they favor allowing same-sex couples to marry legally, with responses dichotomized as favorable or unfavorable and summarized as the proportion with favorable attitudes at the state level (continuous). State-level estimates are derived using a multilevel regression and poststratification (MRP) approach with maximum likelihood modeling and poststratification based on demographics and state characteristics from the American Community Survey. Using model output, relative risks are calculated for meaningful cut-offs in SM social climate (for example, quartiles) using average marginal effects. Qualitative Component (Aim 3): Aim 3 participants are purposively sampled from states representing three strata of SGM-affirming policy adoption: high, medium, and low, with up to 10 youth per stratum. Within each stratum, constructed sampling quotas aim to include approximately 4 cisgender boys, 4 cisgender girls, and 2 transgender or nonbinary youth, and to ensure that at least half of participants identify as a racial or ethnic minority. This approach supports exploration of intersectional mechanisms and addresses known limitations of surveillance data in capturing diverse SGM youth experiences. Qualitative data are analyzed using thematic analysis to identify mechanisms through which state-level policies and social climates may affect SV/IPV risk among SGM youth. These findings contextualize and interpret the population-level quantitative results from Aims 1 and 2.