Maternal morbidity and mortality ratios continue to increase in the United States, in contrast to other well-resourced nations. The 26.6% increase in the US maternal mortality ratio from 2000 to 2014 (23.8/100,000 live births) was paralleled by a 75% increase in major maternal morbidity.^1,2 The most common causes of maternal mortality are hemorrhage, thromboembolic events, cardiac disease, sepsis, and hypertensive disorders,³ and 40% of maternal deaths are believed to be preventable.⁴

Efforts to reduce the morbidity and mortality ratio in the United States have included education and standardization. For instance, the incidence of maternal hemorrhage was reduced in California after the introduction of team training, simulation drills, and the use of protocols and checklists.⁵ However these efforts do not address the problem of variable resource distribution, such as the case of women living in rural areas of British Columbia, Canada, who were shown to have increased rates of eclampsia, thromboembolism, and uterine dehiscence compared to their urban counterparts.⁶ Efforts to reduce maternal mortality in well-resourced areas should focus on the optimal use of those resources.

In 1976, the March of Dimes published Toward Improving the Outcome of Pregnancy, an opinion statement in response to rising levels of maternal and neonatal mortality. This document called for an “integrated system of regionalized perinatal care” based on a referral network to ensure that high-risk patients received care at centers equipped with the appropriate resources. The report stratified maternal and neonatal care into three tiers and called upon facilities with higher levels of care to collaborate with facilities with lower care levels for transport services, education, data collection and quality initiatives.⁷

After publication of this report, the concept of regionalized perinatal care was widely adopted by the neonatal community, and numerous studies demonstrated improved neonatal outcomes in risk-appropriate care settings.^8–11 For instance, a meta-analysis by Lasswell et al (2010) demonstrated increased odds of neonatal morbidity for low-birth-weight infants (weighing less than 1500 g) born at facilities other than Level III hospitals (38% vs. 23%, with an adjusted odds ratio of 1.62 and a 95% confidence interval [CI] [1.44–1.83]).¹⁰ Menard et al (1998) demonstrated a higher neonatal mortality for low-birth-weight infants at Level II facilities (with a neonatologist present, but lacking a complete multidisciplinary team) vs. Level III.¹¹ Neonatal care regionalization persists, as was recently reaffirmed by the American Academy of Pediatrics.¹²

Regulation of neonatal care levels is formalized through state agencies that survey and grant the appropriate care level through hospital licensing, certification, and approval of proposals for expansion based on need. Clear definitions identify the required personnel, technical abilities, and appropriate infants for each level, which dictate practice patterns. For instance, the Massachusetts Department of Health designates Level I neonatal facilities as those capable of providing care for low-risk neonates at ≥35 weeks gestation and are prepared to provide neonatal resuscitation and stabilization for transport, but do not electively deliver infants <35 weeks. Level II neonatal specialty care is appropriate for infants with problems that are short term and not likely to require urgent subspecialty services, such as uncomplicated preterm labor at ≥32 weeks gestation. Level II nurseries require the availability of a specialized care team for any emergency and care for neonates who have been stabilized at higher centers and are retro-transported to be closer to home. Level III neonatal facilities are prepared to provide for infants born at <32 weeks gestation or weighing <1500 g as well as infants of any gestational age who require pediatric surgical or medical subspecialty care. Level IV neonatal units care for the most critically ill babies, such as those requiring extracorporeal membrane oxygenation (Table 5-1). The March of Dimes continues to promote standard definitions, reaffirmed in 1993 and again in 2010.¹³

The analogous concept of maternal care regionalization has not evolved effectively. Some states, notably Indiana, Arizona, and Maryland, designate levels of maternal care distinct from neonatal care levels with integrated consultative and transport systems, but each state uses different definitions, and no national standard currently exists. A joint American Congress of Obstetricians and Gynecologists/ Society of Maternal Fetal Medicine (ACOG/SMFM) Obstetric Care Consensus document entitled Levels of Maternal Care, published in February 2015 and reaffirmed in 2016, calls for “an integrated regionalized framework to identify when transfer of care may be necessary to provide risk-appropriate maternal care.”¹⁴ Levels of Maternal Care is a guideline, not a mandate; it suggests definitions for five levels of maternal care: birth centers, basic care (Level I), specialty care (Level II), subspecialty care (Level III), regional perinatal healthcare centers (Level IV), and birthing centers (Level V). (Table 5-2). Trauma care is not considered a factor in maternal level of care, as pregnant women are triaged for trauma in the same manner as nonpregnant patients. Home birth is not considered appropriate for any designation.