Abstract P2046: Gestational Hypertension in the Latine Population by Race, Ancestry, and Nativity: An Intersectional-MAIHDA Approach

Objectives: Gestational hypertension (GH), hypertension during pregnancy, is associated with greater risk of cardiovascular disease later in life. Aggregating Latine populations may mask important heterogeneity in the burden of this adverse pregnancy outcome. Differences have been examined by nativity, but few have accounted for potential heterogeneity by race or ancestry, and none have accounted for all three simultaneously. Thus, the purpose of this research is to use a novel intersectional modeling approach to estimate the prevalence of GH among Latine birthing people by race, nativity, and ancestry. Methods: 2020 Natality Public Use Data of 863,907 Latine birthing people’s live births were analyzed. Data on GH (yes/no), birthing person’s nativity (U.S. Born/Not U.S. Born), race (White, Black, Native American/Native Alaskan [Native A/A], Asian, Native Hawaiian/Pacific Islander [NOPI], and more than 1 race) and ancestry (Mexican American, Puerto Rican, Dominican, Cuban, Central South American [C/S American] and Other) were extracted from birth certificates. Intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (I-MAIHDA) was used to examine variation in GH across strata of nativity, race, and ancestry. This modeling approach allows for assessment of intersectionality in a way that maximizes statistical efficiency. Results: Mean GH prevalence among Latine birthing people was 7.1%. Black birthing people were more likely to develop GH than White birthing people, and non-U.S. born individuals were less likely to develop GH than U.S. born (Table 1). Puerto Rican, Dominican, and Cuban birthing people were more likely to develop GH than Mexican birthing people (Table 1). I-MAIHDA produced 72 nativity-race-ancestry strata, ranging in size from 10 to 278,584 birthing people (mean: 11,998.7). Birthing people with the lowest predicted probability of GH were predominately non-U.S. C/S Americans, while those with the highest predicted probability were predominately Black and Indigenous U.S. Born Puerto Rican and Dominican (Table 2). Approximately 14% of the between-stratum variance in GH was explained by nativity-race-ancestry intersectional effects (versus the main additive effects of each). Conclusion: We found evidence of variation in GH prevalence within the Latine birthing population. Future research should consider which factors drive heterogeneity in GH prevalence among Latine birthing people.