Perinatal Health in U.S. Communities Without Maternity Care

This is the authors' abstract. We don't add key points for this paper.

Katy Backes Kozhimannil, PhD, MPAJulia D. Interrante, PhD, MPH.Obstetric care access is declining in communities across the United States, and it is essential to understand how living in areas without maternity care affects maternal health outcomes, including maternal and pregnancy-related mortality. Prior research has shown elevated risks of maternal morbidity and mortality among people living in the communities with the least access to maternity care, including Black, Latino, Indigenous, and rural families and those with lower incomes.1,2 In this month's issue of Obstetrics & Gynecology, Dr. Atwani et al3 tackle this important topic, describing the association between maternity care access and maternal health in counties across the United States. This analysis adds to a growing body of literature that can help inform efforts to improve maternal health and address inequities in access to care. Conducting such analyses is nuanced and complex, requiring careful attention to understanding and measuring exposure (maternity care access) and outcomes (maternal and infant health) as well as the limitations of data available for such measures.4 Identifying whether a county has facilities and personnel to support childbirth is incredibly challenging.4 Dr. Atwani et al used the March of Dimes measure5 to evaluate availability of maternity care facilities and personnel, which is widely known and commonly used but has important limitations that should be considered when interpreting study findings, especially in rural contexts. The authors point out that this measure did not include information about family physicians who provide maternity care (March of Dimes started including family physicians in 2024). Family physicians attend births at 41% of rural hospitals with maternity care,6 so the lack of data on these clinicians results in disproportionate underreporting of maternity care access in rural communities. Additionally, not all obstetrician–gynecologists (ob-gyns) attend births or provide maternity care, but the Area Health Resources Files7 group all ob-gyns together and may incorrectly indicate maternity care access in the March of Dimes measure for clinicians who provide only gynecologic care but do not attend births. Further, measurement of hospitals that provide maternity care also came from Area Health Resources Files data, extracted from the American Hospital Association’s Annual Survey. Our prior validation research shows that using American Hospital Association data capture only 88% of obstetric unit closures.4 Hospital mergers and acquisitions, which are more common among rural hospitals, are also not addressed. The lack of recent, accurate, and widely available measures of all maternity care providers and facilities offering maternity care is a long-standing problem not unique to the March of Dimes measure, these authors, or any of us who aim to study these important topics. Rather, these limitations stem from a lack of data infrastructure specific to maternity care availability in the United States. We suggest shifting away from the term “maternity care deserts” and instead more precisely describing these areas as “counties without maternity services.”2,8 Whereas deserts are naturally occurring, the absence of maternity care in some communities is not—access to health care in the United States is shaped by policy and business decisions. The term “deserts” in this context is used to imply emptiness, whereas desert environments are thriving ecosystems and are home to vibrant Indigenous communities. Using the term “maternity care deserts” may contribute to erasure of Indigenous Peoples and the issues they face, including high risks of maternal mortality and long distances to maternity care. We agree with the authors' conclusion that, “Expanding access to comprehensive maternity services in these counties is a critical step toward improving maternal health outcomes and narrowing persistent inequities.” Using terminology that reflects a naturally occurring phenomenon suggests inevitability and detracts from the fact that new policy directions could change whether communities have local maternity care. Referring to “counties without maternity services” helps draw focus to the potential for strengthening local access to these services. Dr. Atwani et al used the Centers for Disease Control and Prevention (CDC) WONDER data to ascertain the study's main outcome, maternal mortality, as well as pregnancy-related mortality. The CDC WONDER data are publicly available and easy to access, making them useful for many purposes. However, one challenge with using these data at the county level is the suppression of small numbers, which may disproportionately affect counties without maternity care that are frequently rural and less populated. CDC WONDER uses a suppression rule that does not reveal counts below a population threshold. Even when data are pooled across years (2018–2022), data for uncommon (though still too frequent) outcomes such as maternal or pregnancy-related mortality are suppressed for more than one-third of counties. Ensuring that measurement of key outcomes is rural-relevant (eg, by examining effects of data practices such as suppression) is essential in studies of maternity care access.1 Moving forward, it is essential to recognize, measure, and discuss a full range of maternal and infant health outcomes beyond mortality, because the effects of lack of access to maternity care extend to many aspects of maternal health and well-being.8 For example, evidence indicates high levels of anxiety and stress for pregnant patients who lose local maternity care access or patients who need to travel long distances to give birth.1,9 Whenever possible, research should include outcomes related to the maternal–infant dyad rather than just the pregnant patient or the infant in isolation.10 Every tragic outcome in maternal and child health reverberates through families and communities for generations. Access to maternal health is part of basic community infrastructure2,9 and should be treated as such. Maternity care access should be prioritized in public health data systems so that we can accurately measure the effects of lack of access, but also in clinical and policy efforts to combat the maternal health crisis.