Differences in medically assisted reproduction use by sexual identity and partnership: a prospective cohort of cisgender women.

study questionDoes medically assisted reproduction (MAR) use among cisgender women differ among those with same-sex partners or lesbian/bisexual identities compared to peers with different-sex partners or heterosexual identities? SUMMARY ANSWER: Women with same-sex partners or lesbian/bisexual identities are more likely to utilize any MAR but are no more likely to use ART (i.e. IVF, reciprocal IVF, embryo transfer, unspecified ART, ICSI, and gamete or zygote intrafallopian transfer) compared to non-ART MAR (i.e. IUI, ovulation induction, and intravaginal or intracervical insemination) than their different-sex partnered and completely heterosexual peers. WHAT IS KNOWN ALREADY: Sexual minority women (SMW) form families in myriad ways, including through fostering, adoption, genetic, and/or biological routes. Emerging evidence suggests this population increasingly wants to form genetic and/or biological families, yet little is known about their family formation processes and conception needs. STUDY DESIGN, SIZE, DURATION: The Growing Up Today Study is a US-based prospective cohort (n = 27 805). Participants were 9-17 years of age at enrollment (1996 and 2004). Biennial follow-up is ongoing, with data collected through 2021. PARTICIPANTS/MATERIALS, SETTING,

methodsCisgender women who met the following criteria were included in this sample: endorsed ever being pregnant; attempted a pregnancy in 2019 or 2021; and endorsed either a male- or female-sex partner OR responded to questions regarding their sexual identity during their conception window. The main outcome was any MAR use including ART (i.e. procedures involving micromanipulation of gametes) and non-ART MAR (i.e. nonmanipulation of gametes). Secondary outcomes included specific MAR procedures, time to conception, and trends across time. We assessed differences in any MAR use using weighted modified Poisson generalized estimating equations. MAIN RESULTS AND THE ROLE OF CHANCE: Among 3519 participants, there were 6935 pregnancies/pregnancy attempts and 19.4% involved MAR. A total of 47 pregnancies or pregnancy attempts were among the same-sex partnered participants, while 91 were among bisexual participants and 37 among lesbian participants. Participants with same-sex, compared to different-sex partners were almost five times as likely to use MAR (risk ratio [95% CI]: 4.78 [4.06, 5.61]). Compared to completely heterosexual participants, there was greater MAR use among lesbian (4.00 [3.10, 5.16]) and bisexual (2.22 [1.60, 3.07]) participants compared to no MAR use; mostly heterosexual participants were also more likely to use ART (1.42 [1.11, 1.82]) compared to non-ART MAR. Among first pregnancies conceived using MAR, conception pathways differed by partnership and sexual identity groups; differences were largest for IUI, intravaginal insemination, and timed intercourse with ovulation induction. From 2002 to 2021, MAR use increased proportionally to total pregnancies/pregnancy attempts; ART use was increasingly common in later years among same-sex partnered and lesbian participants. LIMITATIONS, REASONS FOR CAUTION: Our results are limited by the small number of SMW, the homogenous sample of mostly White, educated participants, the potential misclassification of MAR use when creating conception pathways unique to SMW, and the questionnaire's skip logic, which excluded certain participants from receiving MAR questions. WIDER IMPLICATIONS OF THE

findingsPrevious studies on SMW family formation have primarily focused on clinical outcomes from ART procedures and perinatal outcomes by conception method, and have been almost exclusively limited to European, clinical samples that relied on partnership data only. Despite the small sample of SMW within a nonrepresentative study, this is the first study to our knowledge to use a nonclinical sample of cisgender women from across the USA to elucidate family formation pathways by partnership as well as sexual identity, including pathways that may be unique to SMW. This was made possible by our innovative approach to MAR categorization within a large, prospective dataset that collected detailed sexual orientation data. Specifically, lesbian, bisexual, and same-sex partnered participants used both ART and non-ART MAR at similar frequencies compared to heterosexual and different-sex partnered participants. This may signal differential access to conception pathways owing to structural barriers, emerging conception trends as family formation among SMW has increased, and a need for conception support beyond specialized providers and fertility clinics. STUDY FUNDING/COMPETING INTEREST(S): The research reported in this publication was supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health (NIH), under award number R01MD015256. Additionally, KRSS is supported by NCI grant T32CA009001, AKH by the NCI T32CA057711, PC by the NHLBI T32HL098048, BM by the Stanford Maternal Child Health Research Institute Clinical Trainee Support Grant and the Diversity Fellowship from the American Society for Reproductive Medicine Research Institute, BGE by NICHD R01HD091405, and SM by the Thomas O. Pyle Fellowship through the Harvard Pilgrim Health Care Foundation and Harvard University, NHLBI T32HL098048, NIMH R01MH112384, and the William T. Grant Foundation grant number 187958. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The first author recently had a leadership role in the not-for-profit program, The Lesbian Health Fund, a research fund focused on improving the health and wellbeing of LGBTQ+ women and girls. The fund did not have any role in this study and the author's relationship with the fund did not bias the findings of this manuscript. TRIAL REGISTRATION NUMBER: N/A.