The Availability of Midwifery Care in Rural United States Communities

Abstract

Introduction

Access to pregnancy‐related and childbirth‐related health care for rural residents is limited by health workforce shortages in the United States. Although midwives are key pregnancy and childbirth care providers, the current landscape of the rural midwifery workforce is not well understood. The goal of this analysis was to describe the availability of local midwifery care in rural US communities.

Methods

We developed and conducted a national survey of rural US hospitals with current or recently closed childbirth services. Maternity unit managers or administrators at 292 rural hospitals were surveyed from March to August 2021, with 133 hospitals responding (response rate 46%; 93 currently offering childbirth services, 40 recently closed childbirth services). This cross‐sectional analysis describes whether rural hospitals with current or prior childbirth services had midwifery care with certified nurse‐midwives available locally and whether rural communities with and without midwifery care differed by hospital‐level and county‐level characteristics.

Results

Among hospitals surveyed, 55% of those with current and 75% of those with prior childbirth services reported no locally available midwifery care. Of the 93 rural communities with current hospital‐based childbirth services, those without midwifery care were more likely to have lower populations (37% vs 33%); majority populations that were Black, Indigenous, and people of color (24% vs 10%); and hospitals where at least 50% of births were Medicaid funded (77% vs 64%), compared with communities with midwifery care. Conversely, communities with midwifery care more often had greater than 30% of patients traveling more than 30 miles for hospital‐based childbirth services (38% vs 28%).

Discussion

More than half of rural hospitals surveyed reported no locally available midwifery care, and availability differed by hospital‐level and county‐level characteristics. Efforts to ensure pregnancy and childbirth care access for rural birthing people should include attention to the availability of local midwifery care.

INTRODUCTION

Approximately 1 in 7 births in the United States occurred in a rural community in 2021, totaling more than half a million births. ¹ Although the annual number of rural births has not changed substantially over the last 2 decades, more than 400 hospital‐based maternity units in the United States have closed since 2004, ² , ³ , ⁴ leaving more than half of rural counties without a hospital‐based maternity unit. ³ Prior research suggests that health care workforce shortages, coupled with other financial and safety challenges to operating low‐volume maternity units, contribute to the decline in access to care for rural residents during pregnancy and childbirth. ⁵ , ⁶ Such workforce shortages impact rural residents throughout the perinatal period; rural birthing people also face barriers to accessing prenatal and postpartum care ⁷ , ⁸ , ⁹ in addition to challenges accessing labor and birth services at the time of childbirth.

Continuing education (CE) is available for this article. To obtain CE online, please visit http://www.jmwhce.org. A CE form that includes the test questions is available in the print edition of this issue.

QUICK POINTS

• ✦Expanding midwifery care has been identified as one strategic response to the rural maternal health crisis in the United States.
• ✦This study addresses a gap in the literature on midwifery care access in rural US communities, as existing studies are either outdated or are limited in scope to one state.
• ✦This analysis found that more than half of surveyed rural hospitals reported not having locally available midwifery care with certified nurse‐midwives in their hospitals or communities.
• ✦Among surveyed rural hospitals with current childbirth services, this analysis found that midwifery care access differed by hospital‐level and county‐level characteristics; many hospitals in medically underserved and vulnerable rural communities reported not having locally available midwifery care.
• ✦Policy efforts to ensure equitable access to pregnancy and childbirth care should include attention to the availability of local midwifery care in rural US communities.

Prior rural health care workforce research, and many policy initiatives, have focused on physicians who provide care during pregnancy and childbirth (primarily obstetricians as well as family physicians). ⁴ , ⁵ The United States is unique among high‐income countries in that the vast majority of births are attended by physicians, ¹⁰ , ¹¹ but that is changing; midwives hold a longstanding, important, and now‐expanding role in the prenatal and childbirth care workforce in many US communities, including rural areas. ¹² Growing demand for midwifery care and entrance into the profession is reflected in the 29% increase in enrollment in certified nurse‐midwifery and certified midwifery programs between 2016 and 2021. ¹⁰ Furthermore, the US Health Resources and Services Administration's Bureau of Health Workforce forecasts a 32% net growth in the number of midwives in the workforce nationally between 2018 and 2030; this is in contrast to a projected 7% decrease in the number of obstetricians during the same period, due to attrition. ¹³

Although national‐level and state‐level data about the midwifery workforce at‐large are published with regularity, recent literature describing access to midwifery care in rural US communities is limited. The most recent multistate study examining the availability of rural midwifery care was conducted in 2014; this study reported that, across 9 US states, midwives attended births in one‐third of surveyed rural hospitals with maternity units, and 14% of hospitals hoped to expand midwifery care. ¹⁴ A 2019 survey conducted in Kansas found that only 1 of 56 rural Kansas hospitals had a certified nurse‐midwife (CNM) attending births. ¹⁵ These studies have contributed important knowledge about the practice of midwifery in rural hospitals but are now either a decade old or are limited in scope to one state.

The dwindling capacity of rural health care systems to provide pregnancy and childbirth care may be among the forces driving rural‐urban disparities in maternal health outcomes, including higher risks of pregnancy‐related death and severe complications for rural birthing people. ⁵ , ¹⁰ , ¹⁶ The federal government, ¹⁷ , ¹⁸ health care leaders, ¹⁹ , ²⁰ and advocacy organizations ¹⁹ , ²⁰ have proposed the expansion of midwifery care as one strategy to address maternal health workforce shortages in rural communities and the pregnancy‐related and birth‐related health inequities experienced by rural residents. ¹² Furthermore, a more robust rural midwifery workforce may offer rural birthing people a broader range of clinician options for pregnancy‐related and childbirth‐related care, facilitating birthing people's agency in choosing care approaches aligned with their preferences, such as the midwifery model of care. ²¹ Information about the current landscape of local access to midwifery care for rural birthing people in the United States is needed to inform policy efforts around the expansion of the midwifery workforce. This study seeks to address this need by describing the availability of midwifery care in rural US communities with current or prior hospital‐based childbirth services.

METHODS

Data

This cross‐sectional analysis used data from a survey of rural US hospital administrators that we conducted in 2021. First, we used the American Hospital Association (AHA) Annual Survey data and the Centers for Medicaid & Medicare Services (CMS) Provider of Services file to identify hospitals that were providing childbirth services in 2018, using a previously published, verified method. ⁴ We restricted the study population to hospitals in rural counties as defined by the Office of Management and Budget (OMB) as nonmetropolitan core‐based statistical areas. ²² We surveyed a random sample of hospitals in rural counties with a majority non‐Hispanic White population (n = 194) and all hospitals in rural counties where a majority identified as Black, Indigenous, or people of color (BIPOC; n = 98), due to significant racial inequities in pregnancy‐related outcomes and care access for BIPOC birthing people and their infants. ²³ , ²⁴ , ²⁵ County‐level racial and ethnic proportions were identified using data from the American Community Survey. ²⁶

With input from rural health care clinicians and administrators, we created and administered a 47‐question web‐based survey via the Qualtrics platform (Provo, Utah) between March and August 2021. Survey questions addressed 4 topic areas related to rural hospital‐based childbirth care services: local factors, training and clinical safety, workforce and staffing, and finances. More information about the survey design can be found in prior publications. ⁶ , ²⁷ Maternity unit managers or administrators at 292 rural hospitals were surveyed, and 133 responded (46% response rate). Of the 133 hospitals, 93 provided childbirth services at the time of the survey, and 40 had stopped providing childbirth services in the decade prior to the survey.

Measures

Local midwifery care access was identified by responses to the question: “Is midwifery care with CNMs available in your community?” Respondents could answer “In the community and affiliated with my hospital,” “In the community but not affiliated with my hospital,” or “Not available in the community.”

In addition, we examined a variety of characteristics of the sampled rural hospitals and communities. Rural counties were dichotomized as either micropolitan (counties with at least one town of 10,000 or more residents) or noncore (counties that have no town of 10,000 or more residents) per OMB definitions. Using AHA data, we examined Critical Access Hospital status. Finally, using data from our survey, we examined the proportion of hospital births paid by Medicaid (<50% vs ≥50%), whether a hospital had fewer births than the median of the sampled hospitals in 2019 (274 births), and whether a hospital reported that greater than 30% of their patients traveled more than 30 miles to access childbirth care.

Analysis

We compared rural hospitals with and without local midwifery care by the characteristics described above. We calculated proportions of rural communities in our sample with and without local midwifery care stratified by whether the surveyed hospital had current or prior childbirth services. Among hospitals with current childbirth services, we tested differences in midwifery care availability across the characteristics listed above using Pearson χ² tests. Finally, we estimated predicted probabilities and 95% CIs of not having locally available midwifery care in rural communities with current hospital‐based childbirth services, adjusted for all above characteristics.

Analyses were conducted using Stata version 18.0 (College Station, Texas). This study was exempted from review by the University of Minnesota's institutional review board. The Strengthening the Reporting of Observational Studies in Epidemiology checklist for cross‐sectional studies was used to guide the writing of this article (Supporting Information: Appendix S1).

RESULTS

Figure 1 presents the proportions of surveyed rural hospitals with current or prior hospital‐based childbirth services by whether midwifery services with a CNM were reported to be available in the hospital, in the community, or not available locally. Among surveyed hospitals that had current childbirth services, 27% reported having midwifery care available in the hospital, and 18% reported having midwifery care available in their community. In comparison, 10% of rural hospitals that had prior childbirth services reported having midwifery care available in the hospital, and 15% reported having midwifery care available in their community. Overall, 55% (n = 51) of rural hospitals that had current childbirth services and 75% (n = 30) of rural hospitals that had prior childbirth services reported that midwifery care was not available locally.

Figure 1.

Local Midwifery Access in Rural US Communities With Current or Prior Hospital‐Based Childbirth Services, 2021 (n = 133)

This figure pertains to rural US communities with current or prior hospital‐based childbirth services (n = 133).

Figure 2 presents the proportions of surveyed rural hospitals with current hospital‐based childbirth services by hospital‐level and county‐level characteristics, stratified by whether midwifery care with a CNM was or was not locally available. Among rural communities with current hospital‐based childbirth services, 10% of those with locally available midwifery care and 24% of those without midwifery care were in counties with a majority BIPOC population. Approximately 64% of rural hospitals in communities with local midwifery care and 77% of those without local midwifery care had at least 50% of their births financed by Medicaid. More rural hospitals in communities without local midwifery care (53%) had less than the median number of births in 2019 (<274 births) compared with those in communities with midwifery care (43%). Finally, 38% of rural hospitals in communities with local midwifery care reported that more than 30% of their patients traveled more than 30 miles to access childbirth care at the hospital, compared with 28% of those in communities without local midwifery care. None of the differences between surveyed rural hospitals with and without locally available midwifery care were statistically significant (P > .05).

Figure 2.

Characteristics of Rural US Communities With Current Hospital‐Based Childbirth Services, Based on Whether or not There Is Local Midwifery Care, 2021 (n = 93)

Abbreviation: BIPOC, Black, Indigenous, and people of color.

No differences in availability of midwifery care were statistically significant at P < .05. This figure pertains only to rural US communities with current hospital‐based childbirth services (n = 93).

Figure 3 presents adjusted predicted probabilities of not having locally available midwifery care with a CNM among rural communities with current hospital‐based childbirth services, by hospital‐level and county‐level characteristics. Among surveyed hospitals with current childbirth services, the predicted probability of not having locally available midwifery care was approximately 73% (95% CI, 51%‐96%) for hospitals in majority BIPOC counties, compared with 49% (95% CI, 37%‐60%) among those in counties that were majority non‐Hispanic White or had no majority racial or ethnic group. The predicted probability of not having locally available midwifery care was approximately 55% (95% CI, 43%‐67%) among rural communities where the majority (≥50%) of births at the surveyed hospital were Medicaid funded, compared with 49% (95% CI, 30%‐67%) among those where less than 50% of births were Medicaid funded. Predicted probabilities of not having locally available midwifery care were 59% (95% CI, 44%‐75%) and 46% (95% CI, 30%‐62%) among surveyed hospitals with less than the median (<274) number of births and at least 274 births in 2019, respectively. Finally, the predicted probability of not having locally available midwifery care was approximately 57% (95% CI, 45%‐69%) among rural hospitals that reported that more than 30% of patients traveled 30 or fewer miles to access childbirth care at the hospital, compared with 45% (95% CI, 27%‐62%) among hospitals that reported that more than 30% of patients traveled at least 30 miles to access childbirth care. Among rural communities with current hospital‐based childbirth services, the predicted probabilities of not having locally available midwifery care were not significantly different (P > .05) across any of the measured characteristics.

Figure 3.

Predicted Probability of not Having Locally Available Midwifery Care in Rural US Communities With Current Hospital‐Based Childbirth Services, 2021 (n = 93)

Abbreviation: BIPOC, Black, Indigenous, and People of Color; CAH, Critical Access Hospital.

Predicted probabilities have been multiplied by 100%. No predicted probabilities were statistically significant at P < .05. This figure pertains only to rural US communities with current hospital‐based childbirth services (n = 93).

DISCUSSION

Many rural communities, —including those with hospitals that provide childbirth services, lack local access to midwifery care with a CNM. More than half of rural communities that have hospitals with current childbirth services had no locally available midwifery care. Three‐quarters of rural communities whose hospitals recently stopped providing childbirth care had no locally available midwifery care. These findings suggest that access to a full range of maternity care clinicians is limited for pregnant rural residents in these communities. ³ , ⁵ , ²⁷

This analysis found that there was considerable variation in access to local midwifery care among rural US communities. Rural communities without local midwifery care were more likely to be in noncore counties (those rural counties without a town of at least 10,000 residents) compared to those with local midwifery care. Additionally, local midwifery care was less common in rural communities with hospitals where at least half of births were Medicaid funded and in counties with majority BIPOC populations. These findings highlight the need for greater investment in midwifery care access for pregnant residents of rural communities who are particularly impacted by challenges related to urban bias, poverty, and structural racism. ²⁸ , ²⁹ These findings build on previous research reports that BIPOC rural residents and individuals living in socioeconomically disadvantaged rural communities have limited access to pregnancy‐related and birth‐related care, including midwifery care and other evidence‐based birth support services. ²⁷ , ²⁹ , ³⁰ , ³¹

This analysis also found that locally available midwifery care was more common in rural communities where surveyed hospitals reported that a substantial portion of patients (>30%) traveled more than 30 miles to access childbirth care at the hospital, compared with those reporting that more than 30% of patients traveled 30 or fewer miles to access hospital‐based childbirth care. This finding holds promise for the idea of midwifery care as a model for delivering maternity care services in more remote, less‐populated rural areas. Additionally, it could potentially imply a willingness of rural residents to travel further for hospital‐based childbirth care services if they desire midwifery care. ³² , ³³

In this study, 27% of surveyed rural hospitals that provided current childbirth services reported that midwifery care was available in their hospital. This is similar to the findings of a prior study of midwifery care in rural communities across multiple US states, conducted in 2014; this study found that midwives attended births at one‐third of rural hospitals with maternity care units. ¹⁴ Taken together with the current findings, this suggests that the proportion of rural hospitals with current childbirth services that offer midwifery care may have changed little in the last decade, despite trends of maternity units closing in rural communities across the United States. ³

This study adds to the body of evidence that rural residents face barriers to accessing local pregnancy and childbirth care in their communities; barriers to care may contribute to elevated rates of poor pregnancy‐related and birth outcomes for rural residents in the United States. ¹⁶ Residents of rural communities that do not have locally available midwifery care are less able to access person‐centered ³⁴ , ³⁵ and lower‐intervention care ³⁶ , ³⁷ that is associated with the midwifery model of care. In addition, residents of rural communities that do not have locally available midwifery care are less able to choose from a full range of maternity care clinicians for their pregnancy‐related and childbirth‐related care, impacting rural people who would prefer to receive pregnancy‐related and childbirth‐related care in the midwifery model. Finally, the use of midwifery care for birthing people with low‐risk pregnancies is associated with cost savings at the hospital level; rural hospitals and health plans serving rural communities that do not have locally available midwifery care miss out on the potential cost savings that could come from providing midwifery as a more cost‐effective model of care for residents with low‐risk pregnancies. ³⁸

Clinical and Policy Implications

Clinicians, researchers, and policymakers have identified expanding midwifery care as one strategic response to the rural maternal health crisis. ¹² , ¹⁸ This study's findings indicate that the majority of rural communities lack access to local midwifery care, including rural communities that have hospital‐based childbirth services. This underscores the importance of targeted workforce initiatives to expand the availability of midwifery care in rural US communities. Policies could focus on expanding the midwifery care workforce in Maternity Care Target Areas experiencing the largest shortages of maternity care clinicians. ³⁹ The newly announced Transforming Maternal Health model from CMS will focus in part on helping state Medicaid agencies improve access to midwifery care throughout the perinatal period for beneficiaries of Medicaid and the Children's Health Insurance Program. ⁴⁰ Finally, existing midwifery workforce research suggests that workforce‐based efforts are most effective when supported by a broad, multipronged approach, including attention to midwifery scope of practice, reimbursement rates for midwifery care, and targeted investments in midwifery workforce growth. ⁴¹ , ⁴² , ⁴³

In addition, there are a variety of different types of midwives, including CNMs/certified midwives (CMs), certified professional midwives (CPMs), and direct entry midwives (DEMs), who have distinct educational and licensing requirements, scopes of practice, and breadths of care; for example, CNMs/CMs typically practice in hospital settings, whereas CPMs and DEMs work in freestanding birth centers or attend home births. ¹⁰ Initiatives to expand the midwifery workforce in rural communities should include attention to these different types of midwifery care in order to offer rural birthing people a more robust set of options from which to choose for pregnancy‐related and childbirth‐related care. ²¹

Perinatal telehealth service models developed or expanded during the COVID‐19 pandemic may also play a role in bridging access to midwifery care for rural residents, especially those living in rural communities that do not have access to local midwifery care. ⁴⁴ , ⁴⁵ Telehealth can mitigate access barriers for patients and may improve prenatal and postpartum care visit attendance. ⁴⁶ , ⁴⁷ The ability to access midwifery care via a phone or video call may be especially salient for rural residents who would otherwise need to travel long distances to access prenatal appointments and childbirth‐related care. ⁴⁵ Moreover, midwives have reported reductions in job stress and increases in job satisfaction from providing care via telehealth, suggesting a potential role for telehealth in mitigating high clinician burnout rates. ⁴⁵ Furthermore, the decrease in commuting days associated with telehealth could broaden the hiring pool for rural hospitals seeking to recruit midwives from other communities. ⁴⁵

Limitations

This survey analysis provides new information on the midwifery workforce in rural US communities, but it is not without limitations. One limitation was sample size (n = 133); as such, some differences detected in this analysis—between proportions of rural communities with and without locally available midwifery care, and between predicted probabilities of not having locally available midwifery care—did not reach statistical significance at conventional levels (P < .05), despite being potentially clinically and substantively important. In addition, although the hospitals that participated in the survey span all 4 US Census regions and are in counties of varying racial and ethnic demographics, this study is not representative of all rural hospitals or communities. Furthermore, the survey asked only about CNMs, and not other types of midwives. Although the majority of midwives in the United States (especially those who work in hospitals) hold the CNM or CM credential, ⁴⁸ approximately 10% of midwife‐attended births in the US are attended by CPMs and DEMs ¹⁰ ; this study does not capture the care provided by these clinicians. Another limitation is that this analysis relies on the knowledge and recall of survey respondents. Although the rural hospital administrators and maternity unit managers who responded to this survey were likely aware of midwives attending births within their facility, they may not have had comprehensive knowledge about the types of midwives providing maternity‐related care in their communities, or midwives providing primary care or other services. Additionally, respondents may not have had knowledge about all birth resources available in their communities at the time of the survey. As such, the proportions yielded from this analysis may be overestimated (if respondents reported care availability by more types of midwives than just CNMs) or underestimated (if respondents were unaware of midwifery services available in their broader communities). It is therefore important to interpret findings in the context of the knowledge base of the respondents. It should also be noted that since data for this survey were collected (2021), policymakers in several states with substantial rural populations (eg, Arkansas, New Hampshire) have increased Medicaid reimbursement rates for midwifery care and/or expanded midwifery scope of practice laws in response to COVID‐19 and rural workforce shortages. ⁴⁹ , ⁵⁰ Any corresponding impacts on the availability of midwifery care in rural communities would not be reflected in these data. Finally, these data included limited information on the local availability of prenatal care and other types of care that midwives can provide in rural communities and on community‐based midwifery services that are not hospital‐affiliated (eg, freestanding birth centers). Future research is needed to identify the specific types of care that CNMs/CMs and other types of midwives are providing in rural US communities in order to more fully understand the current landscape of rural maternity care and the rural midwifery workforce.

CONCLUSION

More than half of rural hospitals surveyed reported no locally available midwifery care. Furthermore, availability differed by hospital‐level and county‐level characteristics, with many hospitals in medically underserved and vulnerable rural communities reporting no local access to midwifery. As policymakers contemplate strategies to ensure equitable access to pregnancy and childbirth care in rural US communities, careful attention should be directed toward addressing the availability of local midwifery care.

CONFLICT OF INTEREST

The authors have no conflicts of interest to disclose.

Supporting information

Appendix S1. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Checklist