Pregnancy Preferences and Incident Pregnancy in the US.

Importance: Examining associations of pregnancy and childbearing desires with incident pregnancy is vital to understanding whether individuals attain their reproductive preferences. Much of the existing population-based research is limited by simplistic measurement of pregnancy desires and retrospective study designs. Objective: To assess the prospective association of pregnancy preferences with incident pregnancy and to investigate the degree to which people's attainment of their pregnancy preferences differs by sociodemographic characteristics. Design, Setting, and Participants: This cohort study used longitudinal data from the Surveys of Women, state-representative prospective studies of reproductive-aged females from 9 US states (Alabama, Arizona, Delaware, Iowa, Maryland, New Jersey, Ohio, South Carolina, and Wisconsin). Participants were recruited between November 2016 and July 2020 and followed up for 2 to 3 years, depending on the state. Eligible participants were aged 18 to 44 years and female. Analyses included participants who, at the start of each 1-year observation period, could experience pregnancy (ie, not pregnant, no tubal ligation, and not infertile). Data were analyzed from May 2024 to April 2025. Exposures: The primary exposure was preferences about a potential pregnancy within 3 months and childbearing within 1 year, measured using the Desire to Avoid Pregnancy (DAP) scale. Scores were categorized into low, midrange, and high desire to avoid pregnancy. Main Outcome and Measures: The primary outcome was self-reported incident pregnancy over the year after completing the DAP scale. Differences in incident pregnancy were examined within DAP score grouping by sociodemographic characteristics using model-estimated probabilities of pregnancy. Results: The sample included 9565 unique participants (3256 participants [weighted percentage, 48%] aged <30 years) who contributed 18 603 annual longitudinal observations. A majority of participants were nulliparous (4066 participants [weighted percentage, 51%]) or lived with a romantic partner (5952 participants [weighted percentage, 59%]). An estimated 3% (95% CI, 2%-4%) of participants with high DAP scores reported pregnancy over a year, compared with 8% (95% CI, 7%-9%) of those with midrange DAP scores and 25% (95% CI, 23%-27%) of those with low DAP scores. The association of DAP score with pregnancy differed by age, parity, prior-year birth, cohabitation with partner, education level, employment, and racial and ethnic identity. Conclusions and Relevance: In this population-based cohort study of reproductive-aged females, realization of pregnancy preferences varied by sociodemographic characteristics, highlighting reproductive inequities. These results may be used to inform public health efforts to provide contraception, abortion, preconception, and fertility care.