Maternal Medical History, Psychosocial Factors, and Birth Outcomes
[Thesis]
Haviland, Miriam Joan
Wise, Lauren A.
Boston University
2020
137 p.
Ph.D.
Boston University
2020
Major depressive disorder, anxiety, and psychological stress are common co-occurring morbidities in pregnancy. Psychotropic medications are commonly used to treat these conditions, though many women discontinue use in pregnancy due to concerns for adverse pregnancy outcomes. Prior investigations into the effect of psychotropic medications on preterm delivery may be prone to confounding by indication. In the first study of this dissertation, we compared mean gestational age at delivery between women who used psychotropic medications during pregnancy and women who never used medications, stratifying by severity of preconception depressive symptoms and perceived stress (measured before and during pregnancy). We used data from the Boston University Pregnancy Study Online (PRESTO), an ongoing prospective cohort study of pregnancy planners. We found that women who used medications during pregnancy delivered slightly earlier (37.2 weeks; 95% CI: 36.4, 37.9) than women who never used medications (38.1 weeks; 95% CI: 37.5, 38.6). We observed these associations among women with both high and low levels of depressive symptoms and perceived stress. Our results suggest that psychotropic medication use during pregnancy may be associated with slightly shorter gestations. Mental health symptoms (depression and anxiety), stress, and low psychosocial resources (social support and resilience) are associated with preterm delivery. Many of these psychosocial factors tend to co-occur and women who experience more than one of these factors are more likely to deliver preterm than women who experience only one. Understanding what combinations of adverse psychosocial factors women experience during pregnancy may help clinicians more effectively identify women at risk of preterm delivery. In our second study, we identified three latent classes of adverse psychosocial factors (few, some, and many factors) using data from Spontaneous Prematurity and Epigenetics of the Cervix (SPEC), a prospective cohort study of pregnant women at Beth Israel Deaconess Medical Center. Participants with both some (RR: 1.50; 95% CI: 0.86, 2.62) and many adverse psychosocial factors (RR: 1.29; 95% CI: 0.36, 5.00) were more likely to deliver preterm than participants with few factors, though these associations were imprecisely estimated. Our findings suggest that screening for multiple adverse psychosocial factors may help providers better identify women at risk of preterm delivery. Despite advances in in vitro fertilization (IVF) technology, less than half of IVF cycles result in a pregnancy. These low pregnancy probabilities may be due to chromosomal nondisjunction, which causes nonviable aneuploid embryos that are naturally rejected by the body. Preimplantation genetic testing for aneuploidy (PGT-A) was developed to identify euploid embryos prior to implantation. Prior evaluations of PGT-A have produced mixed results, and may be prone to confounding by indication. In our third study, we evaluated the effect of PGT-A on the cumulative incidence of live birth, controlling for important confounders using a propensity score for PGT-A. We found that women ≥ 38 years old who used PGT-A were more likely to have a live birth than women ≥ 38 years old who did not use PGT-A (RR: 1.67; 95% CI: 1.31, 2.13). We also observed that PGT-A increased the likelihood of having a live birth among women 35-37 years old (RR: 1.27; 95% CI: 1.05, 1.54). Among women < 35 years old, those who used PGT-A were no more likely to have a live birth than those who did not (RR: 0.91; 95% CI: 0.78, 1.06). Our findings suggest that PGT-A may be beneficial for older women.