Primary Submission Category: Reproductive health
Sexual Orientation Disparities in Pregnancy Loss
Authors: Colleen Reynolds Payal Chakraborty Tabor Hoatson Jarvis Chen Lori Chibnik Janet Rich-Edwards Brittany Charlton
Presenting Author: Colleen Reynolds*
Background: Pregnancy loss (e.g., miscarriage, stillbirth) is the most common pregnancy complication–roughly 20% of clinically recognized pregnancies end in a loss. However, little research has examined disparities in pregnancy loss by sexual orientation.
Methods: We used data from all pregnancies, excluding induced abortions, reported in the Nurses’ Health Study 2 (NHS2), a longitudinal cohort (146,652 pregnancies to 52,212 participants). Participants reported whether each pregnancy ended in induced abortion, loss (ectopic/tubal, miscarriage, stillbirth), or live birth. Using weighted log-binomial generalized estimating equation (GEE) models, we compared the risk of pregnancy loss for pregnancies among completely heterosexual participants (reference) to those among four sexual minority (SM) groups: completely heterosexual with same-sex attraction/partners/prior SM identity (heterosexual SM), mostly heterosexual, bisexual, and lesbian participants. Weights were constructed based on (a) inverse probability of treatment conditional on demographic characteristics (addressing confounding); (b) inverse probability of censoring (to account for causes of selection, e.g. induced abortion); and (c) the inverse of the number of pregnancies per participant (to address informative cluster size).
Results: 17% of pregnancies ended in a loss. Pregnancies among SM participants were more likely to end in a loss (RR[95% CI]: 1.19[1.13-1.26]), including ectopic/tubal (1.43[1.15-1.78]) and miscarriage (1.19[1.12-1.26]). Pregnancies among lesbians were most likely to end in loss (1.83[1.34-2.49]), including ectopic/tubal (2.70[1.09-6.69]), miscarriage (1.66[1.17-2.37]), or stillbirth (3.56[1.53-8.30]).
Conclusions: Pregnancies among SMs are disproportionately likely to end in a loss. Future research should evaluate potential mechanisms driving these disparities (e.g., heterosexist discrimination, pelvic inflammatory disease, assisted reproductive technology, prenatal care).