A Retrospective Cohort Study of Race and Ethnicity, Pre-pregnancy Weight, and Pregnancy Complications

Objective: To examine relationships between race/ethnicity, pre-pregnancy overweight/obesity status, and pregnancy complications: gestational diabetes (GDM), preeclampsia, and cesarean-delivery. Methods: We conducted a large retrospective population-based cohort study among mothers who delivered a singleton live birth in hospitals that participated in the Obstetrics Clinical Outcomes Assessment Program database (N=57,144). Race/ethnicity was categorized into mutually exclusive groups of Non-Hispanic White (NH-W), NH African-American (NH-AA), Hispanic, NH Asian (NH-A), NH American Indian/Alaskan Native (NH-AI/AN), and NH Native-Hawaiian/Other Pacific Islander (NH-NHP). Pre-pregnancy overweight/obesity status was defined as body mass index (BMI) ≥ 25kg/m2. Information on pregnancy complications was ascertained from medical records. We used unadjusted and adjusted stratified Poisson regression models with robust standard errors to estimate relative risk (RR) and associated 95% confidence intervals (CI). We used stratified models and models with interaction terms to examine interactions between race/ethnicity and overweight/obesity status on pregnancy complications. Results: Majority of women were NH-W (52.2%). Similarly, most were overweight/obese, pre-pregnancy (52.1%). Compared with NH-W, NH-AA, Hispanics, NH-A and NH-NHP had higher risk of GDM (aRR:1.36, 1.44, 2.12, and 2.42 respectively) while Hispanics, NH-A, and NH-NHP had a lower risk (aRR:0.89, 0.53 and 0.64 respectively) of preeclampsia (all p-values<0.05). Women with pre-pregnancy overweight/obesity had higher risk of cesarean-delivery, GDM, and preeclampsia (aRR:1.58, 1.92 and 2.57 respectively, p-values<0.05), when compared with normal-weight women. In stratified models, among overweight/obese women, NH-AI/AN women had an elevated risk of preeclampsia as compared to NH-W (aRR:1.36; 95%CI:1.04-1.76). This association was not present among normal-weight women (aRR:0.64; 95%CI:0.24-1.70). Similarly, among overweight/obese women, but not normal-weight women, Hispanic and NH-NHP had a lower risk of preeclampsia (aRR:0.79; 95%CI:0.70-0.89 and aRR:0.60; 95%CI:0.39-0.94, respectively) and NH-AA had a greater risk of GDM (aRR:1.29; 95%CI:1.09-1.53) when compared with NH-W women. Among all race/ethnicities, but not AI/AN and NH-NHP, overweight/obese women had an increased risk of GDM and preeclampsia, respectively, when compared with normal-weight women (p-values<0.05). The multiplicative interaction terms between race/ethnicity and pre-pregnancy overweight/obesity status were significant for all three pregnancy complications (interaction p-values<0.05). Conclusion: Race/ethnicity modifies associations of pre-pregnancy overweight/obesity status with pregnancy complications. Conversely, pre-pregnancy overweight/obesity status modifies associations of race/ethnicity with pregnancy complications.