Objectives
Sexual and racial/ethnic minority people face unique obstacles accessing abortion care. Yet, little data exist on abortion utilization at these intersections. Using a novel quantitative intersectionality method — multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) — we assessed abortion utilization by sexual orientation and race/ethnicity.
Methods
We analyzed pregnancy-level data from the Nurses’ Health Study 3 and Growing Up Today Study cohorts (n=46,241 pregnancies, 1979-2024). We conducted an intersectional MAIHDA, nesting pregnancies within participants, who were nested within 15 strata defined by sexual orientation (heterosexual, heterosexual with same-sex experience/mostly heterosexual, bisexual/lesbian/gay) and race/ethnicity (Non-Latine Asian, Non-Latine Black, Non-Latine multiple/another race, Non-Latine White, Latine). We fit two multilevel logistic models — one with random intercepts for strata (simple intersectional) and another adding fixed effects (intersectional interaction) — using Bayesian methods. Analyses using informative priors to address abortion underreporting are forthcoming.
Results
Overall, 9.0% of pregnancies ended with induced abortions. Pregnancies to sexual and racial/ethnic minority participants had higher odds of ending in abortion than those to Non-Latine White heterosexual participants. Non-Latine White heterosexual participants had the lowest percentage of pregnancies ending in abortion (6.2%, 95% credible interval: 6.1-6.3%); Non-Latine Black bisexual/lesbian/gay participants had the highest (39.4%, 35.5-43.1%). Intersectional effects explained a substantial portion (14%) of the between-stratum variance.
Conclusions
An intersectional perspective is crucial for addressing abortion care needs of multiply minoritized groups — especially as access has declined — as some groups face disproportionate reproductive health barriers and rely more heavily on abortion to achieve reproductive autonomy.
扫码关注我们
求助内容:
应助结果提醒方式:
