Objective: To evaluate the mediating role of social and environmental determinants of health (SEDH) in the relationship between self-identified race and preterm birth risk. Methods: Data on birth outcomes and five broad domains of SEDH were obtained from the Pregnancy, Race, Environment, Genes study, a prospective cohort of pregnant women (50% Black American) in Richmond, Virginia. Configural consistency across racial groups for SEDH domains assessed via multiple indicators was evaluated using unidimensional confirmatory factor models. Mediation models tested the extent to which individual variables/constructs and broader SEDH domains accounted for the association between self-identified race and gestational age at delivery (GAD). Results: Economic Stability, Education, Health/Health Care, and Social/Community Context showed evidence of mediation through one or more specific subdimensions. Education and Economic Stability fully mediated the association between race and GAD, indexed by maternal educational attainment and current job duration, respectively. In contrast, consideration of Social/Community Context exposed an even larger association between race and GAD. Conclusions: This study highlights the role of specific and broader SEDH domains in the association between race and spontaneous preterm birth (PTB) risk. Findings underscore the significant impact of educational access and economic security on reproductive outcomes.

A history of abortion is associated with cervical dysfunction during pregnancy, but there remains uncertainty about whether risk can be stratified by the abortion type, the abortion procedure, or number of previous abortions. The objective of this study was to verify the relationship between cervical dysfunction measures in pregnancies with and without a history of termination. Embase and Medline databases were searched from 01 January 1960 to 01 March 2022 resulting in a full-text review of 28 studies. The Newcastle-Ottawa Scale (NOS) was used to assess the quality and risk of bias for non-randomized studies. The meta-analysis consisted of 6 studies that met all inclusion and exclusion criteria and included a combined total of 2,513,044 pregnancies. Cervical dysfunction was defined as either cervical insufficiency/incompetence in 4 of the studies and as short cervix in the others. Results from a random-effects model using reported adjusted odds ratios (aOR) estimated an increase in the odds of 2.71 (95% CI 1.76, 4.16) for cervical dysfunction in the current pregnancy related to a history of induced or spontaneous abortion. Subgroup analyses with only induced abortions (surgical/medical) estimated an aOR of 2.54 (95% CI 1.41, 4.57), while studies limited to surgical abortions had an aOR of 4.08 (95% CI 2.84, 5.86). The risk of cervical dysfunction in the current pregnancy was also found to be dependent on the number of previous abortions. In this meta-analysis, a prior history of abortion, and specifically induced abortions, was associated with cervical dysfunction. The protocol was registered in PROSPERO (CRD42020209723).