Abstract MP095: Pregnancy as a Window to Racial Disparities in Hypertension
Background: Black women in the United States have 50% higher rates of hypertension (HTN) than white women. This disparity persists after accounting for many known HTN risk factors. Evidence indicates that pregnancy complications may reveal increased risks for later HTN, and some complications, such as preterm delivery, are more common in black vs. white women.
Hypotheses: 1) Adjustment for multiple HTN risk factors measured at mid-life will explain some, but not all, of the black-white disparity in mid-life HTN among women. 2) Markers of pregnancy health will help explain the remaining black-white disparity in mid-life HTN.
Methods: Data came from a Michigan-based cohort study of women enrolled during pregnancy and followed-up at mid-life (n=678, mean age = 37, range 25-58). We included women without pre-pregnancy hypertension (n=652) categorized as African American (AA) (n=242) and white (n=373). We categorized women as hypertensive (systolic BP ≥140, diastolic BP ≥90, or self-reported use of antihypertensive medications, n=126), pre-hypertensive (systolic BP 120-139 or diastolic BP 80-89, n=149), and normotensive (n=340). Mid-life risk factors for HTN were grouped into 4 domains: socioeconomic status ([SES], based on education, occupation, marital status, insurance, and wealth), psychosocial (depression, hostility, job strain), behavioral (current smoking, diet quality score, sedentary hours/day, global sleep score), and physiological (body mass index [BMI], lipids, c-reactive protein [CRP]). We used generalized logit models to assess the degree to which each individual factor attenuated the AA (vs. white) odds ratio (OR) for HTN at mid-life and then sequentially added variables to a multivariable model. We then added indicators of pregnancy health (preterm delivery, pre-pregnancy BMI, and CRP, depressive symptoms, hypertensive disorders, and lipids during pregnancy).
Results: AA women had 3.28(95% CI:1.96,5.51) times the odds of HTN compared to white women after adjusting for age. Adjustment for SES attenuated the OR to 2.52(95% CI:1.46,4.36). Further adjustment for psychosocial factors and behaviors attenuated the OR to 2.30(95% CI:1.29,4.11), and BMI and CRP attenuated it to 2.13(95%CI:1.15,3.93). Adjustment for preterm delivery, pre-pregnancy BMI, and CRP and depressive symptoms during pregnancy reduced the OR to 1.88(95%CI:0.97,3.64), with BMI, CRP, and depressive symptoms playing a larger role than preterm delivery. Adjustment for hypertensive disorders and lipids during pregnancy did not further reduce the race disparity.
Conclusions: SES, psychosocial factors, behaviors, and biomarkers measured at mid-life explained some, but not all, of the race disparity in mid-life HTN. Indicators of pregnancy health, particularly pre-pregnancy BMI, inflammation (CRP), and depressive symptoms contributed substantially to the race disparity in HTN at mid-life.
Author Disclosures: C.E. Margerison-Zilko: None. J.M. Catov: None. C. Holzman: None.
- © 2017 by American Heart Association, Inc.