Study design

This prospective cohort includes a pregnancy study, the POUCH Study, and a maternal follow-up component, the POUCHmoms Study⁽¹¹⁾. The POUCH Study enrolled 3,019 pregnant women at 16–27 weeks’ gestation (1998–2004) from 52 clinics in five Michigan communities⁽¹²⁾ with the aim of studying biological and social factors affecting adverse pregnancy outcomes. Inclusion criteria were singleton pregnancy with no known congenital anomaly, maternal age ≥ 15, maternal serum alpha-fetoprotein (MSAFP) screening, no pre-pregnancy diabetes mellitus, and English speaking. Approval for this study was obtained from institutional review boards at MSU, Michigan Department of Community Health, and nine community hospitals. This study did not include core outcome sets and participants were not directly involved in shaping the research.

To conserve resources, a sub-cohort of 1,371 POUCH Study participants was studied in greater depth (e.g. medical records abstracted, biological samples analyzed, placental pathology). The sub-cohort was constructed by oversampling women with preterm delivery (PTD) and women with a higher risk of pregnancy complications (i.e. African Americans, women with high MSAFP). To account for the cohort and sub-cohort sampling strategy, inverse-probability sampling weights are used in all POUCH and POUCHmoms Study analyses. For example, African-American women, women with high MSAFP and women with PTD who were oversampled into the sub-cohort are assigned a ‘smaller weight’ so that they represent their proportion in the eligible population agreeing to participate in the original POUCH Study.

Of the 1,371 sub-cohort women, 1,280 did not decline future study participation and were presumed alive at the initiation of the POUCHmoms Study (Figure S1), which was designed primarily to assess risk factors for early evidence of CVD at 7–15 years post-POUCH Study pregnancy. Between 2011 and 2014, 678 POUCHmoms Study participants completed interviews and CVD-related assessments, i.e. BP, carotid ultrasounds, anthropometrics, heart rate variability and blood biomarkers. Women’s age at follow-up ranged from 25 to 58 years. For the current analyses we removed women if they lacked complete placenta information, i.e. placentas not saved at delivery or insufficient decidual tissue for assessment (n=160), had chronic hypertension before/during the POUCH Study pregnancy (n=22), or were missing prenatal records (n=6), which led to a final sample of 490 women.