Severe Maternal Morbidity among Low-Income Patients with Hypertensive Disorders of Pregnancy.

OBJECTIVE: Hypertensive disorders of pregnancy (HDP) contribute significantly to the development of severe maternal morbidities (SMM), particularly among low-income women. The purpose of the study was to explore the relationship between maternal characteristics and SMM, and to investigate if differences in SMM exist among patients with HDP diagnosis. STUDY DESIGN: This study utilized 2017 Alabama Medicaid administrative claims. SMM diagnoses were captured using the Centers for Disease Control and Prevention's classification by International Classification of Diseases codes. Maternal characteristics and frequencies were compared using Chi-square and Cramer's V statistics. Logistic regression analyses were conducted to examine multivariable relationships between maternal characteristics and SMM among patients with HDP diagnosis. Odds ratios and 95% confidence intervals (CIs) were used to estimate risk. RESULTS: A higher proportion of patients experiencing SMM were >34 years old, Black, Medicaid for Low-Income Families eligible, lived in a county with greater Medicaid enrollment, and entered prenatal care (PNC) in the first trimester compared with those without SMM. Almost half of patients (46.2%) with SMM had a HDP diagnosis. After controlling for maternal characteristics, HDP, maternal age, county Medicaid enrollment, and trimester PNC entry were not associated with SMM risk. However, Black patients with HDP were at increased risk for SMM compared with White patients with HDP when other factors were taken into account (adjusted odds ratio [aOR] = 1.37, 95% CI: 1.11-1.69). Patients with HDP and SMM were more likely to have a prenatal hospitalization (aOR = 1.45, 95% CI: 1.20-1.76), emergency visit (aOR = 1.30, 95% CI: 1.07-1.57), and postpartum cardiovascular prescription (aOR = 2.43, 95% CI: 1.95-3.04). CONCLUSION: Rates of SMM differed by age, race, Medicaid income eligibility, and county Medicaid enrollment but were highest among patients with clinical comorbidities, especially HDP. However, among patients with HDP, Black patients had an elevated risk of severe morbidity even after controlling for other characteristics. KEY POINTS: · Patients with SMM were more likely to have a HDP diagnosis.. · Among those with HDP, Black patients had elevated risk of SMM.. · Differences in care delivery did not explain SMM disparities..

Infant Mortality among Adolescent Mothers in the United States: A 5-Year Analysis of Racial and Ethnic Disparities.

OBJECTIVE: This study was aimed to examine differences in infant mortality outcomes across maternal age subgroups less than 20 years in the United States with a specific focus on racial and ethnic disparities. STUDY DESIGN: Using National Center for Health Statistics cohort-linked live birth-infant death files (2009-2013) in this cross-sectional study, we calculated descriptive statistics by age (<15, 15-17, and 18-19 years) and racial/ethnic subgroups (non-Hispanic white [NHW], non-Hispanic black [NHB], and Hispanic) for infant, neonatal, and postneonatal mortality. Adjusted odds ratios (aOR) were calculated by race/ethnicity and age. Preterm birth and other maternal characteristics were included as covariates. RESULTS: Disparities were greatest for mothers <15 and NHB mothers. The risk of infant mortality among mothers <15 years compared to 18 to 19 years was higher regardless of race/ethnicity (NHW: aOR = 1.40, 95% confidence interval [CI]: 1.06-1.85; NHB: aOR = 1.28, 95% CI: 1.04-1.56; Hispanic: aOR = 1.36, 95%CI: 1.07-1.74). Compared to NHW mothers, NHB mothers had a consistently higher risk of infant mortality (15-17 years: aOR = 1.12, 95% CI: 1.03-1.21; 18-19 years: aOR = 1.21, 95% CI: 1.15-1.27), while Hispanic mothers had a consistently lower risk (15-17 years: aOR = 0.72, 95% CI: 0.66-0.78; 18-19 years: aOR = 0.74, 95% CI: 0.70-0.78). Adjusting for preterm birth had a greater influence than maternal characteristics on observed group differences in mortality. For neonatal and postneonatal mortality, patterns of disparities based on age and race/ethnicity differed from those of overall infant mortality. CONCLUSION: Although infants born to younger mothers were at increased risk of mortality, variations by race/ethnicity and timing of death existed. When adjusted for preterm birth, differences in risk across age subgroups declined and, for some racial/ethnic groups, disappeared. KEY POINTS: · Infant mortality risk was highest for adolescents <15 years old across racial/ethnic groups.. · Racial/ethnic disparities in timing of death were present even among the youngest adolescents.. · Infants of NHB adolescents had greatest risk of mortality, especially as age increased.. · Preterm birth influenced infant mortality risk, especially among NHB adolescents..

Testing a Tailored Social-Ecological Model for Autism Spectrum Disorders.

OBJECTIVE: To propose a tailored social ecological model for Autism Spectrum Disorders and explore relationships between variables in a large nationally-representative dataset. METHODS: A tailored social-ecological model was developed and examined across variables in the 2016/2017 National Survey of Children's Health. A series of iterative multivariable logistic regressions were performed including individual, family, and community/neighborhood variables. A multivariable logistic regression using state-level fixed effects was performed to understand dynamics related to macro-level policies. RESULTS: In the full model, gender, disability severity, certain types of insurance coverage and household income were significantly related to ASD diagnosis. Females had lower odds of a diagnosis compared to males (aOR: 0.27; CI:0.18-0.41). Children with at least one other moderate/severe disability had odds 7.61 higher (CI:5.36-10.82) of a diagnosis than children without moderate/severe disabilities. Children with public insurance only (aOR:1.66; CI:1.14-2.41) or both private and public insurance coverage (aOR: 2.62; CI:1.6-4.16) had higher odds of a diagnosis compared to children with private insurance only. For those who reported it was "somewhat" or "very often" hard to cover basics with their income, odds of a diagnosis were higher compared to those who reported it was "never" or "hardly ever" hard to cover basics (aOR: 1.676; CI:0.21-2.56). CONCLUSIONS FOR PRACTICE: Patterns of ASD diagnosis are related to individual and family characteristics. There is some evidence that a child's environment has some relationship to reported ASD diagnosis. Professionals should be aware of an individual's environmental factors or context when assessing for ASD.

The weathering hypothesis and stillbirth: racial disparities across the life span.

Objective: Given that studies have consistently demonstrated increased risk of stillbirth with increasing maternal age as well as race-ethnic disparities in stillbirth, the primary research question is how stillbirth risks within and between race-ethnic groups change with age.Design: Using fetal death (stillbirth) and live birth data from the National Center for Health Statistics 2007-2014. We calculated crude stillbirth rates (per 1,000 deliveries). Using multivariable logistic regression, we calculated adjusted odds ratios (aOR) and 95% confidence intervals (CI) for stillbirth across age groups to examine disparities in stillbirth among non-Hispanic blacks (blacks) and non-Hispanic whites (whites).Results: Rates and risks of stillbirth increased as maternal age increased; however, the magnitude of risk varies by race-ethnicity. Compared to women less than 20, black women age 40+ had a risk of almost 3.5 times (aOR = 3.47, 95% CI = 3.24-3.70) whereas among white women age 40+, the risk was more than 2.5 times (aOR = 2.68, 95% CI = 2.55-2.82). The risk of stillbirth among blacks compared to whites increased, peaking at ages 30-34 (aOR = 2.64, 95% CI = 2.56-2.73). At age 35 and beyond, the disparity in risks declined.Conclusion: As age increased, the risk of stillbirth increased for both blacks and whites. Because the disparity in risk between blacks and whites did not continue to increase with age, stillbirth does not seem to reflect the weathering hypothesis as other perinatal outcomes do.

Autism Spectrum Disorder Among US Children (2002-2010): Socioeconomic, Racial, and Ethnic Disparities.

OBJECTIVES: To describe the association between indicators of socioeconomic status (SES) and the prevalence of autism spectrum disorder (ASD) in the United States during the period 2002 to 2010, when overall ASD prevalence among children more than doubled, and to determine whether SES disparities account for ongoing racial and ethnic disparities in ASD prevalence. METHODS: We computed ASD prevalence and 95% confidence intervals (CIs) from population-based surveillance, census, and survey data. We defined SES categories by using area-level education, income, and poverty indicators. We ascertained ASD in 13 396 of 1 308 641 8-year-old children under surveillance. RESULTS: The prevalence of ASD increased with increasing SES during each surveillance year among White, Black, and Hispanic children. The prevalence difference between high- and low-SES groups was relatively constant over time (3.9/1000 [95% CI = 3.3, 4.5] in 2002 and 4.1/1000 [95% CI = 3.6, 4.6] in the period 2006-2010). Significant racial/ethnic differences in ASD prevalence remained after stratification by SES. CONCLUSIONS: A positive SES gradient in ASD prevalence according to US surveillance data prevailed between 2002 and 2010, and racial and ethnic disparities in prevalence persisted during this time among low-SES children.

Disparities in Infant Mortality by Race Among Hispanic and Non-Hispanic Infants.

OBJECTIVES: U.S.-born Hispanic infants have a well-documented health advantage relative to other minority groups. However, little published research has examined racial heterogeneity within the Hispanic population, in relation to health outcomes. The current study aims to explore possible implications of racial identification for the health of U.S. born Hispanic compared to non-Hispanic infants. Methods Data were drawn from 2007 to 2008 NCHS Cohort Linked Live Birth-Infant Death Files, restricted to deliveries of Hispanic black, Hispanic white, non-Hispanic black (NHB) and non-Hispanic white mothers (NHW) (n = 7,901,858). Adjusted odds ratios for first week mortality, neonatal, postneonatal, and overall infant mortality were calculated for each group, using NHW as the reference group. RESULTS: A distinct health gradient was observed in which NHB infants (n = 1,250,222) had the highest risk of first week (aOR 2.29, CI 2.21-2.37), neonatal (aOR 2.23, CI 2.17-2.30), postneonatal (aOR 1.74, CI 1.68-1.81), and infant mortality (aOR 2.05, CI 2.00-2.10) compared to NHW infants (n = 4,578,150). Hispanic black infants (n = 84,377) also experienced higher risk of first-week (aOR 1.28 (1.12-1.47), neonatal (aOR .27, CI 1.13-1.44), postneonatal (aOR 1.34, CI 1.15-1.56), and infant mortality (aOR 1.30, CI 1.18-1.43) compared to both NHW and Hispanic white infants (n = 1,989,109). Conclusions for Practice: Risk of infant mortality varies among Hispanic infants by race, with poorer outcomes experienced by Hispanic black infants. Compared to non-Hispanic infants of the same race, Hispanic black infants experience a smaller health disadvantage and Hispanic white infants have better or similar infant health outcomes. Our findings suggest implications of racial heterogeneity on infant health outcomes, and provide insight into the role of race as a social construct.

Prevalence of Cerebral Palsy among 8-Year-Old Children in 2010 and Preliminary Evidence of Trends in Its Relationship to Low Birthweight.

BACKGROUND: The public health objective for cerebral palsy (CP) in the United States is to reduce the percentage of children with CP who were born low birthweight (LBW, <2500 g) by 10% between 2006 and 2020. This study reports the prevalence of CP in a constant surveillance area for the years 2006, 2008, and 2010 and describes initial progress towards the CP public health objective. METHODS: Data on children with CP at age 8 years were ascertained by the Autism and Developmental Disabilities Monitoring (ADDM) Network, a population-based surveillance system that monitored CP in four areas of the United States. RESULTS: CP prevalence in 2010 was 2.9 per 1000 [95% confidence interval (CI) 2.6, 3.2], down from 3.5 (95% CI 3.2, 3.9) in the same surveillance area in 2006. Among CP cases with no documented postneonatal aetiology, 49.1% (95% CI 42.9, 55.2) were born LBW in 2010 compared with 54.3% (95% CI 48.4, 60.1) in 2006. In 2010, 28.1% (95% CI 22.9, 30.4) were born very low birthweight (VLBW, <1500 g) compared with 35.4% (95% CI 30.0, 41.2) in 2006. The relative risks for associations between CP and both LBW and VLBW also declined, though not significantly, during the study period. CONCLUSIONS: Declines in the associations between CP and LBW categories may have contributed to declines during the study period in both the prevalence of CP and the percentage of children with CP who were born LBW or VLBW. Ongoing monitoring of these trends is warranted.

Identifying Continuous Quality Improvement Priorities in Maternal, Infant, and Early Childhood Home Visiting.

OBJECTIVE: The purpose of this article was to describe a methodology to identify continuous quality improvement (CQI) priorities for one state's Maternal, Infant, and Early Childhood Home Visiting program from among the 40 required constructs associated with 6 program benchmarks. The authors discuss how the methodology provided consensus on system CQI quality measure priorities and describe variation among the 3 service delivery models used within the state. DESIGN: Q-sort methodology was used by home visiting (HV) service delivery providers (home visitors) to prioritize HV quality measures for the overall state HV system as well as their service delivery model. RESULTS: There was general consensus overall and among the service delivery models on CQI quality measure priorities, although some variation was observed. Measures associated with Maternal, Infant, and Early Childhood Home Visiting benchmark 1, Improved Maternal and Newborn Health, and benchmark 3, Improvement in School Readiness and Achievement, were the highest ranked. CONCLUSIONS: The Q-sort exercise allowed home visitors an opportunity to examine priorities within their service delivery model as well as for the overall First Teacher HV system. Participants engaged in meaningful discussions regarding how and why they selected specific quality measures and developed a greater awareness and understanding of a systems approach to HV within the state. The Q-sort methodology presented in this article can easily be replicated by other states to identify CQI priorities at the local and state levels and can be used effectively in states that use a single HV service delivery model or those that implement multiple evidence-based models for HV service delivery.

Perinatal Disparities Between American Indians and Alaska Natives and Other US Populations: Comparative Changes in Fetal and First Day Mortality, 1995-2008.

To compare fetal and first day outcomes of American Indian and Alaskan Natives (AIAN) with non-AIAN populations. Singleton deliveries to AIAN and non-AIAN populations were selected from live birth-infant death cohort and fetal deaths files from 1995-1998 and 2005-2008. We examined changes over time in maternal characteristics of deliveries and disparities and changes in risks of fetal, first day (<24 h), and cause-specific deaths. We calculated descriptive statistics, odds ratios and confidence intervals, and ratio of odds ratios (RORs) to indicate changes in disparities. Along with black mothers, AIANs exhibited the highest proportion of risk factors including the highest proportion of diabetes in both time periods (4.6 and 6.5 %). Over time, late fetal death for AIANs decreased 17 % (aOR = 0.83, 95 % CI 0.72-0.97), but we noted a 47 % increased risk over time for Hispanics (aOR = 1.47, 95 % CI 1.40-1.55). Our data indicated no change over time among AIANs for first day death. For AIANs compared to whites, increased risks and disparities persisted for mortality due to congenital anomalies (ROR = 1.28, 95 % CI 1.03-1.60). For blacks compared to AIANs, the increased risks of fetal death (2005-2008: aOR = 0.60, 95 % CI 0.53-0.68) persisted. For Hispanics, lower risks compared to AIANs persisted, but protective effect declined over time. Disparities between AIAN and other groups persist, but there is variability by race/ethnicity in improvement of perinatal outcomes over time. Variability in access to care and pregnancy management should be considered in relation to health equity for fetal and early infant deaths.

Time trends and payer differences in lengths of initial hospitalization for preterm infants, Arkansas, 2004 to 2010.

OBJECTIVE: The objective of this study was to examine the time trend in length of stay (LOS) and explore potential differences in neonatal LOS by insurance type for preterm infants in Arkansas between 2004 and 2010. STUDY DESIGN: There were 18,712 preterm infants included in our analyses. Accelerated failure time models were used to model neonatal LOS as a function of insurance type and discharge year while adjusting for key maternal and infant characteristics, and complication/anomaly indicators. RESULTS: Before adjusting for the complication/anomaly indicators, the LOS for preterm infants delivered to mothers in the Medicaid group was 3.2% shorter than those in the private payer group. Furthermore, each subsequent year was associated with a 1.6% increase in the expected LOS. However, after accounting for complications and anomalies, insurance coverage differences in neonatal LOS were not statistically significant while the trend in LOS persisted at a 0.59% increase for each succeeding year. CONCLUSION: All of the apparent differences in LOS by insurance type and more than half of the apparent increase in LOS over time are accounted for by higher rates of complications among privately insured preterm infants and increasing rates of complications for all surviving preterm infants between 2004 and 2010.

Prevalence of cerebral palsy, co-occurring autism spectrum disorders, and motor functioning - Autism and Developmental Disabilities Monitoring Network, USA, 2008.

AIM: The aim of this study was to report the prevalence and characteristics of children with cerebral palsy (CP). METHOD: Children with CP (n=451) were ascertained by the Autism and Developmental Disabilities Monitoring (ADDM) Network, a population-based, record-review surveillance system monitoring CP in four areas of the USA. Prevalence was calculated as the number of children with CP among all 8-year-old children residing in these areas in 2008. Motor function was categorized by Gross Motor Function Classification System level and walking ability. Co-occurring autism spectrum disorders (ASD) and epilepsy were ascertained using ADDM Network surveillance methodology. RESULTS: The period prevalence of CP for 2008 was 3.1 per 1000 8-year-old children (95% confidence interval 2.8-3.4). Approximately 58% of children walked independently. Co-occurring ASD frequency was 6.9% and was higher (18.4%) among children with non-spastic CP, particularly hypotonic CP. Co-occurring epilepsy frequency was 41% overall, did not differ by ASD status or CP subtype, and was highest (67%) among children with limited or no walking ability. INTERPRETATION: The prevalence of CP in childhood from US surveillance data has remained relatively constant, in the range of 3.1 to 3.6 per 1000, since 1996. The higher frequency of ASD in non-spastic than in spastic subtypes of CP calls for closer examination.

Assessing needs and resources for the home visiting system in Alabama: a mixed methods approach.

The purpose of this article is to describe the initial assessment for the development of a home visiting (HV) system in a state with no existing system. We outline a mixed methods process where the quantitative component was used to identify the communities that possess "at-risk" profiles, and the qualitative component explored the resources and gaps in existing HV services. We employed a mixed methods approach, using six categories of indicators from quantitative secondary data sources to identify "at-risk" profiles for Alabama's 67 counties. A weighted score for each indicator was calculated and counties were ranked. Surveys and focus groups were conducted to further define resources and gaps of existing HV programs. The composite indicator scores identified 13 counties as having the highest level of risk. Five of these 13 communities had no HV home visitation services. Areas of focus for future HV system development include trust, communication, availability, cost, and timeliness. In this assessment related to the Alabama HV system, we used quantitative data to apply criteria to the indicators being measured and qualitative data to supplement the quantitative findings. We examined resources, gaps, program quality, and capacity of the existing HV programs in order to assist in the future development of the HV system and early childhood system. The methods presented in this paper have potential applications beyond HV programs and systems, including broader examinations of complex systems for service provision to the maternal and child health populations.

Underinsurance in children with special health care needs: the impact of definition on findings.

To identify differences in groups of children with special healthcare needs (CSHCN) identified as underinsured by two alternate definitions and discuss implications for policy decisions based on using one definition over another. Secondary data from the National Survey of CSHCN 2005/2006 were analyzed. Only CSHCN who were continuously-insured for 12 months were included in analyses. We identified groups of underinsured CSHCN using two general definitions ("economic" and "attitudinal") and three mutually-exclusive groups (identified by both definitions, identified by attitudinal but not economic, and identified by economic but not attitudinal). Key variables included demographics and condition characteristics. Different underinsurance rates were identified [attitudinal = 30.9 % (n = 11,470); economic = 22.7 % (n = 8,447)] with fair agreement by kappa score (κ = 0.3194; Z = 65.91; p > 0.0001). Differences across mutually-exclusive groups included family income ≥400 % FPL (attitudinal only = 34.2 %, economic only = 16.3 %, both = 18.4 %; p < 0.001) and high severity (attitudinal only = 42.5 %, economic only = 68.5 %, both = 69.9 %; p < 0.001). CSHCN who needed equipment/supplies/home health (OR = 1.31, p < 0.001) had increased odds of being identified as underinsured by the economic, but not attitudinal definition. CSHCN with private insurance had increased odds of being identified by attitudinal only or both definitions, but not by economic only (AO: OR = 1.41, p < 0.001; BOTH: OR = 2.36, p < 0.001). Despite overlap between the two definitions, choosing either one excludes some CSHCN, potentially underestimating the extent of underinsurance and masking important findings related to specific conditions characteristics. A definition that comprehensively identifies and describes underinsurance is vital to translating health insurance coverage expansion into benefit packages that meet complex health and service needs.

Achieving Consensus Among Stakeholders Using the Nominal Group Technique: A Perinatal Quality Collaborative Approach.

BACKGROUND AND OBJECTIVES: Public health systems exhibiting strong connections across the workforce experience substantial population health improvements. This is especially important for improving quality and achieving value among vulnerable populations such as mothers and infants. The purpose of this research was to demonstrate how Alabama's newly formed perinatal quality collaborative (Alabama Perinatal Quality Collaborative [ALPQC]) used evidenced-based processes to achieve consensus in identifying population quality improvement (QI) initiatives. METHODS: This multiphase quantitative and qualitative study engaged stakeholders (n = 44) at the ALPQC annual meeting. Maternal and neonatal focused QI project topics were identified and catalogued from active perinatal quality collaborative websites. The Delphi method and the nominal group technique (NGT) were used to prioritize topics using selected criteria ( impact , enthusiasm , alignment , and feasibility ) and stakeholder input. RESULTS: Using the Delphi method, 11 of 27 identified project topics met inclusion criteria for stakeholder consideration. Employing the NGT, maternal projects received more total votes (n = 535) than neonatal projects (n = 313). Standard deviations were higher for neonatal projects (SD: feasibility = 10.9, alignment = 17.9, enthusiasm = 19.2, and impact = 22.1) than for maternal projects (SD: alignment = 5.9, enthusiasm = 7.3, impact = 7.9, and feasibility = 11.1). Hypertension in pregnancy (n = 117) and neonatal abstinence syndrome (n = 177) achieved the most votes total and for impact (n = 35 and n = 54, respectively) but variable support for feasibility . CONCLUSIONS: Together, these techniques achieved valid consensus across multidisciplinary stakeholders in alignment with state public health priorities. This model can be used in other settings to integrate stakeholder input and enhance the value of a common population QI agenda.

Classifying children with ASD by service utilization and treatment type: A cluster analysis of a nationally representative United States survey.

BACKGROUND: If there are patterns of the distribution of services and treatments across the population of people with ASD, these patterns should be based along clinical characteristics or other service needs and not sociodemographic characteristics unrelated to evidence-based care. We examined how individuals in a broad, nationally representative sample "grouped together" based on service utilization and services needed but not covered by insurance. By understanding various treatment patterns, clinicians, researchers, policymakers, and self-advocates and their families can better advocate for high-quality, evidence-based services to be provided equitably. METHODS: Using the 2011 Survey of Pathways to Diagnosis and Services, a cluster analysis was performed to explore patterns in this population based on medication use, private services use, school-based service use, and services not covered by insurance. Differences in clusters were then explored through multinomial logistic regression. RESULTS: Six clusters emerged, showing differences in the level of service/medication usage and insurance coverage. Differences across clusters were associated with the level of functional limitation and age at ASD diagnosis. Disparities by insurance type, functional limitation, and age at diagnosis exist among patterns of ASD service provision. CONCLUSIONS: Our analysis showed that intervention for children with ASD can be across several scales - high and low users of services (both private and school-based), high and low users of medications, and high and low levels of reported non-covered services. The differences were clustered in multiple ways. Further research should incorporate longitudinal and nationally representative data to explore these relationships further.

Community and Systems Contributors and Strategies to Reduce Racial Inequities in Maternal Health in the Deep South: Provider Perspectives.

Purpose: Black pregnant individuals in Alabama are disproportionately affected by severe maternal morbidity and mortality (SMM). To understand why racial disparities in maternal health outcomes persist and identify potential strategies to reduce these inequities, we sought perspectives from obstetric health care providers, health administrators, and members of local organizations who provide pregnancy, delivery, and postpartum care services in Alabama. Methods: We conducted qualitative in-depth interviews with stakeholders (n=20), purposively recruited from community-based organizations, clinical settings, government organizations, and academic institutions. Interview guides were based on Howell's conceptual model of pathways to racial disparities in maternal mortality. Data were coded using a modified framework theory approach and analyzed thematically. Results: Racism, unjust laws and policies, and poverty/lack of infrastructure in communities emerged as major themes contributing to racial disparities in maternal health at the community and systems levels. Inadequate health insurance coverage was described as a strong driver of the disparities. Service providers suggested strategies for Alabama should be community focused, evidence based, and culturally sensitive. These should include Medicaid expansion, expanded parental leave, and removal of laws restricting choice. Community- and systems-level interventions should include community infrastructure improvements, choice in maternity services, and provision of digital communication options. Conclusions: Providers shared perspectives on community and structural areas of intervention to reduce racial inequities in SMM. These results can inform discussions with health system and community partners about Alabama and other Deep South initiatives to improve maternal health outcomes in black communities.

Contribution of Health Care Practitioner and Maternity Services Factors to Racial Disparities in Alabama: A Qualitative Study.

OBJECTIVE: Black patients are three times more likely to die of pregnancy-related causes than White patients in the United States, and Alabama has the third worst maternal mortality rate in the nation. We sought to identify health care practitioner and maternity service factors contributing to disparities in Alabama, as well as potential strategies to address these contributors. METHODS: We conducted key informant interviews with obstetricians, nurses, doulas, lactation counselors, health system administrators, and representatives of professional organizations who deliver maternity care to racially and ethnically diverse patients in Alabama. The interview guide was developed using Howell's conceptual framework on racial and ethnic disparities in severe maternal morbidity and mortality. Adopting a thematic analysis approach, we coded and analyzed transcripts using NVivo 12 software. Open coding and selective coding were conducted to identify themes related to health care practitioner- and maternity services-level determinants. RESULTS: Overall, 20 health care practitioners or administrators were interviewed. Primary themes related to health care practitioners included implicit bias and explicit racism, lack of communication and lack of positive patient-health care practitioner relationships, lack of cultural sensitivity, and variation in clinical knowledge and experience. Primary themes related to maternity services included lack of accessibility, inadequate quality and content of care, lack of continuity of care, discriminatory facility policies, and workforce shortages and lack of diversity. Strategies suggested by participants to address these factors included bias trainings for health care practitioners, improvements in racial and interdisciplinary diversity in the maternity workforce, and evidence-based interventions such as group prenatal care, disparities dashboards, simulation trainings, early warning signs criteria, and coordinated care. CONCLUSION: We gained diverse perspectives from health care practitioners and administrators on how maternity health care practitioner and maternity health services factors contribute to inequities in maternal health in Alabama. Strategies to address these contributing factors were multifaceted.

Children with cerebral palsy: racial disparities in functional limitations.

BACKGROUND: Previous studies of the frequency of cerebral palsy in the United States have found excess prevalence in black children relative to other groups. Whether the severity of cerebral palsy differs between black and white children has not previously been investigated. METHODS: A population-based surveillance system in 4 regions of the United States identified 476 children with cerebral palsy among 142,338 8-year-old children in 2006. Motor function was rated by the Gross Motor Function Classification System and grouped into 3 categories of severity. We used multiple imputation to account for missing information on motor function and calculated the race-specific prevalence of each cerebral palsy severity level. RESULTS: The prevalence of cerebral palsy was 3.7 per 1000 black children and 3.2 per 1000 white children (prevalence odds ratio [OR] = 1.2 [95% confidence interval = 1.0-1.4]). When stratified by severity of functional limitation, the racial disparity was present only for severe cerebral palsy (black vs. white prevalence OR=1.7 [1.1-2.4]). The excess prevalence of severe cerebral palsy in black children was evident in term and very preterm birth strata. CONCLUSION: Black children in the United States appear to have a higher prevalence of cerebral palsy overall than white children, although the excess prevalence of cerebral palsy in black children is seen only among those with the most severe limitations. Further research is needed to explore reasons for this disparity in functional limitations; potential mechanisms include racial differences in risk factors, access to interventions, and under-identification of mild cerebral palsy in black children.

Disparities in fetal death and first day death: the influence of risk factors in 2 time periods.

OBJECTIVES: We examined how changes in risk factors over time influence fetal, first day, and combined fetal-first day mortality and subsequent racial/ethnic disparities. METHODS: We selected deliveries to US resident non-Hispanic White and Black mothers from the linked live birth-infant death cohort and fetal deaths files (1995-1996; 2001-2002) and calculated changes over time of mortality rates, odds, and relative odds ratios (RORs) overall and among mothers with modifiable risk factors (smoking, diabetes, or hypertensive disorders). RESULTS: Adjusted odds ratios (AORs) for fetal mortality overall (AOR=0.99; 95% confidence interval [CI]=0.96, 1.01) and among Blacks (AOR=0.98; 95% CI=0.93, 1.03) indicated no change over time. Among women with modifiable risk factors, the RORs indicated no change in disparities. The ROR was not significant for fetal mortality (ROR=0.96; 95% CI=0.83, 1.01) among smokers, but there was evidence of some decline. There was evidence of increase in RORs in fetal death among mothers with diabetes and hypertensive disorders, but differences were not significant. CONCLUSIONS: Disparities in fetal, first day, and combined fetal-first day mortality have persisted and reflect discrepancies in care provision or other factors more challenging to measure.

Characterizing Hypertensive Disorders of Pregnancy Among Medicaid Recipients in a Nonexpansion State.

Background: The incidence of hypertensive disorders of pregnancy (HDP) are on the rise in the United States, especially in the South, which has a heavy chronic disease burden and large number of Medicaid nonexpansion states. Sizeable disparities in HDP outcomes exist by race/ethnicity, geography, and health insurance coverage. Our objective is to explore HDP in the Alabama Medicaid maternity population, and the association of maternal sociodemographic, clinical, and care utilization characteristics with HDP diagnosis. Materials and Methods: Data were from Alabama Medicaid delivery claims in 2017. Bivariate analyses were used to examine maternal characteristics by HDP diagnosis. Hierarchical generalized linear models, with observations nested at the county level, were used to assess multivariable relationships between maternal characteristics and HDP diagnosis. Results: Among women with HDP diagnosis, a higher proportion were older, Black, had other comorbidities, and had more perinatal hospitalizations or emergency visits compared with those without HDP diagnosis. There were increased odds of an HDP diagnosis for older women and those with comorbidities. Black women (adjusted odds ratio [aOR] = 1.24, 95% confidence interval [CI]: 1.16-1.33), women insured only during pregnancy by Sixth Omnibus Reconciliation Act Medicaid (aOR = 1.08, 95% CI: 1.02-1.15), and women entering prenatal care (PNC) in the second trimester (aOR = 1.10, 95% CI: 1.03-1.18) had elevated odds of HDP diagnosis compared with their counterparts. Conclusions: Beyond traditional demographic and clinical risk factors, not having preconception insurance coverage or first trimester PNC entry were associated with higher odds of HDP diagnosis. Improving the provision and timing of maternity coverage among Medicaid recipients, particularly in nonexpansion states, may help identify and treat women at risk of HDP and associated adverse perinatal outcomes.