Factors associated with giving birth at advanced maternal age in the United States

Abstract

Objective: In 2018, 17 percent of all births in the United States occurred to women of advanced maternal age (AMA.) While the outcomes of AMA pregnancies have been examined extensively, the drivers behind increasing rates of AMA pregnancies in the United States are less understood. Some scholars have asserted that women are increasingly delaying their first birth in favor of educational and career aspirations. Yet birth trends in the United States do not support this as the primary explanatory factor of AMA births. Other factors may also contribute to high rates of AMA in the United States. This study sought to identify main predictors of AMA birth using a cross-sectional retrospective sample.

Design: We employed a multivariate logistic regression analysis on a cross-sectional retrospective sample to identify significant independent predictors of giving birth at advance maternal age (AMA) in the United States.

Setting: Data was obtained from the Unites States Pregnancy Risk Assessment Monitoring System (PRAMS) Phase 7 Core Questionnaire and linked birth certificates. Questions are designed to assess maternal attitudes and experiences before, during and just after pregnancy. Surveys for Phase 7 were completed in 2017 and 2018. The study population represents mothers from 34 states that met the CDC's 55% response rate threshold and Puerto Rico.

Participants: The PRAMS dataset comprises self-reported data and linked birth certificate data from women who recently gave birth. A total of 38,549 mothers are included in the Phase 7 dataset. State sample sizes ranged from 503 mothers in Wyoming to 1,897 mothers in Michigan. All mothers gave birth in the year 2017.

Measurements and findings: The outcome of interest was AMA birth, defined as conceiving and subsequently giving birth to a baby at age 35 or older. Predictors for AMA birth were selected a priori and included: pregnancy intention, history of previous live birth, insurance status, income, education, race/ethnicity, marital status, and urban location. Previous live birth to at least one child was a significant independent predictor for AMA birth. Mothers with high parity, defined as 6 or more previous live births, were 17 times more likely to give birth at advanced maternal age. Mothers with an unwanted pregnancy were 1.9 times more likely to have an AMA birth. College attainment, high income, marital status, urbanicity, and race/ethnicity were also independent predictors of AMA birth. Health insurance was not a significant predictor of AMA birth after accounting for other factors.

Key conclusions: Delayed and late childbirth may not be intentional for a significant group of older mothers. Converse to popular assumptions that women delay childbearing in favor of career aspirations, the majority of AMA mothers have previous children. Half of AMA mothers have two or more previous children. The findings in this paper suggests that multiple factors predict AMA births. There may be several subtypes of women who enter pregnancy at advanced maternal age.

Implications for practice: As women weigh personal desire to bear children against competing social expectations, they may find themselves navigating their own unique path shaped in part by the region in which they live. Better characterization of the circumstances that lead to advanced maternal age in the United States, including exploration of unintended and unwanted AMA pregnancy, is necessary to develop policies and interventions that meet women's needs. This work should utilize a reproductive justice framework to ensure that women's preferences, particularly women of color, are upheld while promoting health and wellbeing for women.

Keywords: Contraception behavior; Maternal age; Maternal health; Pregnancy; Pregnancy unwanted; Reproductive health; Unplanned.