Abstract
Poor coordination of care across providers and birth settings has been associated with adverse maternal-newborn outcomes. Research suggests that integration of midwives into regional health systems is a key determinant of optimal maternal-newborn outcomes, yet, to date, the characteristics of an integrated system have not been described, nor linked to health disparities.
Methods
Our multidisciplinary team examined published regulatory data to inform a 50-state database describing the environment for midwifery practice and interprofessional collaboration. Items (110) detailed differences across jurisdictions in scope of practice, autonomy, governance, and prescriptive authority; as well as restrictions that can affect patient safety, quality, and access to maternity providers across birth settings. A nationwide survey of state regulatory experts (n = 92) verified the ‘on the ground’ relevance, importance, and realities of local interpretation of these state laws. Using a modified Delphi process, we selected 50/110 key items to include in a weighted, composite Midwifery Integration Scoring (MISS) system. Higher scores indicate greater integration of midwives across all settings. We ranked states by MISS scores; and, using reliable indicators in the CDC-Vital Statistics Database, we calculated correlation coefficients between MISS scores and maternal-newborn outcomes by state, as well as state density of midwives and place of birth. We conducted hierarchical linear regression analysis to control for confounding effects of race.
Results
MISS scores ranged from lowest at 17 (North Carolina) to highest at 61 (Washington), out of 100 points. Higher MISS scores were associated with significantly higher rates of spontaneous vaginal delivery, vaginal birth after cesarean, and breastfeeding, and significantly lower rates of cesarean, preterm birth, low birth weight infants, and neonatal death. MISS scores also correlated with density of midwives and access to care across birth settings. Significant differences in newborn outcomes accounted for by MISS scores persisted after controlling for proportion of African American births in each state.
Conclusion
The MISS scoring system assesses the level of integration of midwives and evaluates regional access to high quality maternity care. In the United States, higher MISS Scores were associated with significantly higher rates of physiologic birth, less obstetric interventions, and fewer adverse neonatal outcomes.
Introduction
The Lancet Series on Midwifery (2014) concluded that “national investment in midwives and in their work environment, education, regulation, and management … is crucial to the achievement of national and international goals and targets in reproductive, maternal, newborn, and child health” [1]. In countries where midwives are integrated into the health care system, the benefits of midwifery care are well-documented [2]. Global health experts recommend scaling up midwifery to improve maternal and newborn outcomes, reduce rates of unnecessary interventions, and realize cost savings [3,4]. However, access to midwifery care in the United States (US) is markedly lower than in most other “Organisation for Economic Co-operation and Development” (OECD) countries, with approximately 10% of US births attended by midwives compared to 50–75% in other high-resource countries [5]. In addition to low density of midwives per state, all midwives are not universally licensed to practice or integrated into regional health care systems. American midwives face multiple challenges to practice, including numerous regulatory barriers and inability to secure third party reimbursement [6]. As a result, women in many states cannot access midwives because of legal or payor restrictions [7,8].
Regulation has been identified by the International Confederation of Midwives as one of the pillars of a strong midwifery profession [9]. Regulation refers to a set of criteria and processes arising from the legislation that describes the scope of midwifery practice (activities which midwives are educated for, competent in, and authorized to perform, consistent with the ICM Definition of the Midwife) [9]. On a global scale, maternal and perinatal outcomes are better in jurisdictions where midwives are regulated and have the legislative authority to practice to their full scope across birth settings, including collaborating with or referring to other health professionals [2]. To date, it has been difficult to examine the impact of variations in midwifery regulation and integration across the United States on perinatal outcomes or on consumer access to maternity care. To address these gaps, a panel of maternity care and health policy experts who were delegates to the Home Birth Summit III [HBS] in 2014 (http://www.homebirthsummit.org/) designed The Access and Integration Maternity Care Mapping (AIMM) Study. The aim of this transdisciplinary, national research project was to examine the impact of state regulatory environments on access to midwives and association with perinatal outcomes across populations in the United States.
Why does integration matter?
There are very few jurisdictions in the United States (US) where all types of midwives, irrespective of practice site, are fully integrated as regulated health professionals into interprofessional care provider networks. However, interprofessional teamwork is essential to the provision of high-quality maternity care [10]. For example, research indicates that, when professionals collaborate on decision-making and when coordination of care is seamless, fewer intrapartum neonatal and maternal deaths occur during critical obstetric events [11]. Poor communication, disagreement, and lack of clarity around provider roles are identified as primary determinants of these adverse outcomes [10–12]. Beliefs about risk, beneficence, non-maleficence and patient autonomy are often discipline-specific and divergent [13,14]. Rates of intervention, and labour management options that facilitate normal, physiologic birth are known to differ by type of provider [15], by birth setting [16,17], and by provider education. When differences around defining risk and responsibility exist among providers, interprofessional cooperation and access to options for care are reduced [18–20]. Moreover, when patients perceive interprofessional conflict, the culture of safety is diminished [21–23].
Conversely, collaboration among health professionals can improve safety and quality, particularly when care is transferred from low to high resource settings [10]. For example, when a woman plans to give birth in a community setting (home or birth center) she benefits when her midwife can facilitate access to specialized hospital personnel, equipment, or medications when necessary. The ability of midwives to function autonomously to their full scope of practice in community settings, in collaboration with other members of the health system, can enhance cost-effectiveness of maternity care [24,25]. Regardless of birth setting, midwife-led care has been linked to significantly improved perinatal outcomes, and maternal experience, in both healthy and at-risk populations [26–28]. In the US, current evidence suggests that scope of practice laws, as well as other aspects of state policy and regulation, may be reducing the maternity care workforce and access to services [26]. An integrated maternity care system facilitates the full exercise of scope of practice, autonomy, self-regulation, and collaboration across disciplines.
The diverse context for American midwifery practice
Over 15 years ago, the American Public Health Association issued a position statement, calling for increased access and integration of midwifery services in the United States. [29](29) Yet, consistent U.S. standards for regulation, scope of practice, and access to reimbursement for midwives are still lacking, resulting in a fragmented system of care.
There are three professional designations for midwives in the United States: Certified Nurse-Midwife (CNM), Certified Midwife (CM) and Certified Professional Midwife (CPM). CNMs/CMs obtain their basic education in midwifery through university-based nursing programs and obtain a master’s degree. Both CMs and CPMs are direct-entry midwives without a prior nursing credential. CPMs have a median of three years of education before attending deliveries as a primary midwife; half gain certification via portfolio review, 40% graduate from an accredited school and others report blended education pathways [7]. CNMs can obtain licensure in all 50 states and DC, and their scope includes well-woman gynecology and primary care, as well as maternity care. They are prepared for practice in any birth setting, but they almost exclusively practice in hospitals [6,30]. CMs are currently licensed in 5 states, and are prepared for an identical scope of practice and settings for care as CNMs. CPMs can currently obtain licensure in 30 states. They provide antepartum, intrapartum, and postpartum/newborn care in community based settings, but typically cannot obtain hospital practice privileges and often have difficulty establishing reliable systems for referral and collaborative care. [7,31]
Wide variations in state regulatory conditions for midwifery practice, especially with respect to birth place, have created an environment of interprofessional hostility in some jurisdictions and interprofessional cooperation in others. Given the emerging evidence on the adverse impact of interprofessional disarticulation on maternal experience and outcomes [31,32], it is important to understand the connections between different regulatory environments and differences in health outcomes, especially when significant disparities exist across populations. Differences in adverse perinatal outcomes between Caucasian women and women of colors are well-documented [33–36], and persist even when controlling for socio-economic status and access to quality prenatal care [33,37]. There is a dearth of information about whether health disparities can be attributed to differences in health insurance coverage, or access to providers, or quality of care [36,38,39].
In 2015, 89.8% of US births were attended by physicians, 8.5% by CNMs/CMs, 0.8% by other midwives (including CPMs), and 0.8% by other providers [40]. In 2014, methods of payment varied by place of birth: 44.2% of hospital births were paid for by Medicaid, 48.0% by private insurance, 3.4% were self-pay, and 4.4% via other sources. In contrast, 16.4% of community births (birth center or home) were paid for by Medicaid, 29.4% by private insurance, 50.0% through self-pay and 4.2% via other sources. Most community births are attended by midwives and half are not covered by insurance [41]. The regulatory environment for payors has been shown to significantly impact the extent of midwifery practice in a state and autonomy of midwives [42].
Such systems-level deficits may have significant, negative impacts on the health and well-being of maternal-newborn populations. Rates of obstetric interventions are on the rise in the United States and adverse maternal and newborn outcomes are high, compared to other OECD countries [43]. Black Americans experience substantially higher rates of maternal and neonatal mortality, preterm birth, and low birth weight [33,34,44]. However, one study found that in states where CNMs have greater professional autonomy (i.e. physician supervision not required), there were lower rates of surgical birth, preterm birth and low birth weight, even when adjusted for maternal age, parity, race, education, marital status, cigarette use and prenatal care utilization [26].
In the Access and Integration Maternity Care Mapping (AIMM) Study, we went beyond CNM autonomy to create an evidence-based scoring system to rank the level of integration of all types of midwives into health systems. We then examined the relationships between state Midwifery Integration Scores, density of midwives, access to midwives across practice settings, rates of obstetric interventions, and maternal and newborn outcomes.
Methods
We convened a multi-disciplinary Task Force with expertise in maternity services research, public health, midwifery, obstetrics, epidemiology, consumer advocacy, and/or roles in midwifery regulation, legislation, and law. They identified the key variables needed to populate a database of published regulatory data across all 50 states and the District of Columbia, detailing rules regarding scope of practice, and requirements for licensure of CNMs, CPMs, and CMs and practice across birth settings. We then employed a formal, process (see Table 1), modeled on the Delphi method [45,46], best practices for transdisciplinary research, and legal epidemiology [47], to identify and validate the most important items for inclusion in a composite measure of midwifery integration.
Table 1
Development of an evidence-based Midwifery Integration Scoring System (MISS).
| Delphi Round 1 –Concept generation |
| • HBS Regulation and Licensure Task Force (Team 1) reviews source documents and identifies 7 domains of midwifery integration • Database populated with state regulations on scope of practice and restrictions • Team 1 agrees by consensus on 110 key items describing midwifery regulation |
| Delphi Round 2 –Expert content validation |
| • HBS Research and Data Task Force (Team 2) defines optimal regulatory conditions that support patient access and collaborative practice–informed by a review of the evidence, and consultation with Team 1 • Database and rubrics translated into format to allow for a ranked composite scoring and comparison across states • State regulatory content experts (N = 92, 1-2/state) review items and scoring rubrics for accuracy and relevance to local implementation of the law • Team 2 harmonizes data and adapts scoring rubrics to reflect state realities • Final scoring system reviewed and confirmed by consensus among Teams 1 and 2, and national midwifery regulators and clinical leaders |
| Delphi Round 3 –Development and application of composite measure |
| • Team 2 selects 50 key indicators of midwifery integration indicating level of autonomy, ability to practice to full scope, and collaboration across birth settings. • Teams 1 and 2 convene to rank order answer options in each of the 50 items (higher scores indicated more favourable access and practice conditions) • Team 1 develops a weighted scoring system based on patient safety and quality. Item level scores are weighted and summed for a total optimal score of 100. • MISS tool generates State Integration Scores (range = 17 to 61 across the US). • Density of midwives (per 1000 state births) and access to midwives across settings (home, birth center, hospital) correlated to MISS scores and outcomes. • Correlation and regression analyses link state MISS scores to selected perinatal outcomes that are reliably reported by CDC Vital Statistics |
| Delphi Round 4 –Development of the AIMM report card |
| • Teams 1 and 2 meet to reach consensus on interpretation and key messages • Creation of Interactive AIMM Maps: ○ MISS scores categorized into four quartiles (very low, low, moderate, high)1 ○ Perinatal outcomes linked to MISS scores and displayed by highest and lowest quartiles ○ 4 base maps to display: level of integration, density, proportion of midwife-attended births in 3 settings, and proportion of black births by state |
1: We categorized MISS scores and outcomes into four equal categories: Values between the 1-24th percentile, the 25th-49th percentile, the 50th to 74th percentile and the 75th to 100th percentile.
Round 1 –Concept generation
The Task Force self-organized into two teams, one with regulatory, law, and consumer access expertise, and another with expertise in public health, legal anthropology, and perinatal epidemiology research methods, including instrument development. Both teams included clinicians, and consumers. Over three rounds of drafts, edits, and consensus-based discussions, Team 1 identified seven relevant domains that were important to identify in state regulations on midwifery. Four domains describe midwifery practice: scope of practice, provider autonomy, governance, access to referral and medications; and three domains describe patient safety, quality, and access to maternity providers across birth setting. The team identified 110 indicators that differentiate the regulatory environment by domain for each type of midwife (CM, CPM, or CNM), and assigned numeric values to describe the diverse conditions, permissions, or restrictions delineated in the state laws (see Table 2).
Table 2
Sample midwifery integration indicators and weighted scores.
| Are CPM/CNM/CMs regulated? • 0 = Prohibited • 1 = Allowed by previous judicial opinion or not mentioned/not prosecuted to date • 2 = Unregulated but allowed by statutory permission • 4 = Licensed |
| Are there statutory limitations/restrictions to site of practice for licensed CPM/CNM/CMs? • 0 = Yes • 1 = Lack of access to hospital privileging or physician referral/signer • 2 = No |
| Consultation/referral required by law for certain conditions? • 0 = Unregulated state • 1 = Required (R) but difficult to access when needed • 2 = Not required (NR) but difficult to access when initiated by midwife • 3 = R or NR but easily accessed when initiated by CPM/CNM/CM |
| Evidence-informed, validated quality assurance (QA)/quality improvement (QI) state system for all sites (home, hospital, birth centers) • 0 = Hospital only • 1 = Hospital and birth center only • 4 = Home/hospital/birth center |
| Is Medicaid reimbursement available for CPM/CNM/CMs? • 0 = No • 2 = Yes, but challenges with reimbursement including birth site • 3 = Yes |
| Do CPM/CNM/CMs have prescription-writing authority? • 0 = Prohibited or not authorized • 1 = Allowed only by physician • 2 = Limited list of medications allowed • 3 = Comprehensive list of medications allowed • 4 = Prescription-writing authority |
We then widened the consultant pool to include experts from national regulatory, legal, payor, professional and perinatal surveillance bodies. These policy leaders noted that the statutory language does not always accurately represent the realities of how rules and laws are interpreted and implemented. Language used in rule-making may be interpreted in more or less restrictive ways, and some rules are not actionable given infrastructure constraints and systems-level limitations. For example, in one state, CPMs have statutory authority to access emergency medications for the management of complications, such as maternal hemorrhage; however, pharmacists in that state are restricted from furnishing these medications to practitioners who are not affiliated with hospitals. Because CPMs cannot gain access to hospital privileges, they must find alternate ways to exercise their authority to carry these lifesaving medications.
Round 2 –Expert content validation
Hence, to verify the realities of implementation of the law within each state, Team 2 identified and recruited state and national regulatory experts (n = 92) to complete an online survey. Participants included 75 state-specific regulatory board representatives; the presidents, regional and chapter chairs for state midwifery associations, state legislative and policy chairs for the American College of Nurse Midwives (ACNM) and National Association of Certified Professional Midwives (NACPM). They evaluated the connections and discordances between theory and practice for each of the identified indicators within the state regulatory environment. In poorly integrated states our national experts (ACNM, NACPM legislative directors) referred us to local midwifery or consumer experts who could reliably speak to ‘on the ground’ conditions. When two state experts disagreed on an indicator or experts did not know the answer, we further consulted with 17 state or national regulators, to resolve discrepancies.
We harmonized expert responses with our regulatory database through a systematic line-by-line comparison. We validated and/or deferred to the statutory language when there were no discrepancies between statutes and local interpretation or implementation. When state experts provided evidence of local interpretation that differed from the apparent intent of laws or rules, we added or adapted response options to reflect the realities of midwifery practice, consumer access, and/or the interprofessional environment.
Round 3—Development and application of composite measure
A final Delphi process (see Table 1), involving both multidisciplinary teams, led to selection of 50/110 indicators of midwifery integration, and the development of a weighted Midwifery Integration Scoring System (MISS) (50 items, maximum summary score 100) that quantifies the potential impact on patient access to high-quality maternity care across birth settings. Both teams reviewed the 110 items and only retained those that were deemed, by consensus, important or very important to the assessment of midwifery integration. In some cases, 2–3 items were combined into one stem query, and response options expanded. Some items were excluded because team members felt that the items were not directly pertinent to midwifery integration. For example, one item (Does informed consent language in statute and/or regulations allow for informed refusal by client?) was excluded because the item relates more to human rights issue rather than quantifying the level of midwifery integration. To create the weighting system, using a scale of 0 (not important), 1 (somewhat important), 2 (important), 3 (very important), 4 (essential), the teams assessed each item for its potential impact on patient access to high-quality maternity care. They assigned higher item-level scores to indicators of greater integration, more interprofessional collaboration, and/or wider consumer access across birth settings. The final list of items describe the range of possible options for scope of practice, regulatory body, prescriptive authority, requirements for physician supervision, access to Medicaid, etc. that vary in both statutory language and implementation across states. See S1 Table for a full list of the indicators and scoring system.
Ranking states by MISS scores and outcomes
We used the MISS composite summary scores to rank states by degree of integration. Then, using the 2014 CDC-Vital Statistics Database, we calculated Spearman’s rho correlation coefficients between the continuous MISS integration scores and selected maternal-newborn outcomes in each state. We used Spearman’s rho because the MISS scores were normally distributed as indicated by the Shapiro-Wilk Test (0.960, p = 0.08), but the outcomes data were not. We selected indicators that represent cost-effectiveness and quality in perinatal care (e.g. rates of spontaneous vaginal birth, exclusive breastfeeding, cesarean, induction, VBAC, preterm birth, low birth weight, neonatal mortality) [43,48], and were available and reliable in the CDCs Vital Statistics database [47]. Finally, based on data from the Area Health Resource File, and Centers for Medicare and Medicaid Services, we calculated correlations between MISS scores; state density of midwives (per 1000 births); and consumer access to midwives across birth settings, defined as the proportion of all births at 1) hospital, 2) home and 3) birth centers for two categories of midwives a) CNMs/CMs and b) CPMs and other direct entry midwives as reported on the birth certificates for each state.
In addition, we calculated the correlations between 1) CM and 2) CPM licensure and perinatal outcomes, to examine the differential effects of licensure versus integration scores by state for all outcomes. We also identified states with the highest increases in community births (at home and birth centers) over the past 8 years and examined correlations with MISS scores.
Finally, appreciating the complex nature of health disparities, to understand the relative importance of midwifery integration on perinatal outcomes, we conducted hierarchical linear regression modelling, to control for the proportion of Non-Hispanic Black births in each state, when examining the relationship of MISS scores with rates of five outcomes: caesarean, preterm birth, neonatal death, low birth weight, and breastfeeding at birth.
