Early Prediction of Hypertensive Disorders of Pregnancy Using Machine Learning and Medical Records from the First and Second Trimesters.

Hypertensive disorders of pregnancy (HDPs) remain a major challenge in maternal health. Early prediction of HDPs is crucial for timely intervention. Most existing predictive machine learning (ML) models rely on costly methods like blood, urine, genetic tests, and ultrasound, often extracting features from data gathered throughout pregnancy, delaying intervention.

Types and timing of trauma exposure across the life course and maternal hypertension.

Background: Exposure to trauma across the life course may be associated with cardio-metabolic dysfunction during pregnancy; however, previous research has been inconsistent, particularly in highly exposed populations.

Objectives: To estimate associations between types and timing (first occurrence) of trauma exposure and hypertension experienced during pregnancy in a safety-net hospital in Atlanta, Georgia, 2011-2022.

Methods: Participants completed a 14-item trauma screener.

Racial/ethnic differences in pre-pregnancy conditions and adverse maternal outcomes in the nuMoM2b cohort: A population-based cohort study.

Objectives: To determine how pre-existing conditions contribute to racial disparities in adverse maternal outcomes and incorporate these conditions into models to improve risk prediction for racial minority subgroups.

Study Design: We used data from the "Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be (nuMoM2b)" observational cohort study. We defined multimorbidity as the co-occurrence of two or more pre-pregnancy conditions.

Slavery, homeownership, and contemporary perinatal outcomes in the southeast: a test of mediation and moderation.

The objective of this study was to estimate the effect of geographic variation in historic slavery on perinatal outcomes (chronic hypertension, hypertensive disorders of pregnancy [HDP], very preterm birth [VPTB], or very low birth weight birth [VLBW]) among Black people living in states where slavery was legal in 1860; and to test mediation by Black homeownership. We linked data on the proportion of enslaved residents from the 1860 Census to natality data on outcomes (2013-2021) using resident county. The percentage of Black residents in a county who owned their home was a potential mediator.

Planning for the forgotten fourth trimester of pregnancy: A parallel group randomized control trial to test a postpartum planning intervention vs. standard prenatal care.

Background: Black and brown birthing people experience persistent disparities in adverse maternal health outcomes, partially due to inadequate perinatal care. The goal of this study is to design and evaluate a patient-centered intervention for obstetric patients with one or more cardiometabolic risk factors for severe maternal morbidity [gestational diabetes, diabetes mellitus, hypertensive disorders of pregnancy (chronic hypertension, preeclampsia, eclampsia, or gestational hypertension), or preconception obesity (BMI > 30)] to promote postpartum visit attendance.

Methods: To address identified unmet needs for postpartum support and barriers to postpartum care, we developed 20 thematic postpartum planning modules, each with corresponding patient educational materials, community resources, care coordination protocols, and clinician support tools (decision aids, electronic medical record prompts and fields).

Childhood maltreatment and trajectories of cardiometabolic health across the reproductive life span among individuals with a first birth during the Coronary Artery Risk Development in Young Adults Study.

Objective: Childhood adversity is associated with poor cardiometabolic health in adulthood; little is known about how this relationship evolves through childbearing years for parous individuals. The goal was to estimate differences in cardiometabolic health indicators before, during and after childbearing years by report of childhood maltreatment in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort study.

Methods: Including 743 individuals nulliparous at baseline (1985-1986) with one or more pregnancies >20 weeks during follow-up (1986-2022), we fit segmented linear regression models to estimate mean differences between individuals reporting or not reporting childhood maltreatment (physical or emotional) in waist circumference, triglycerides, high-density lipoprotein cholesterol, systolic and diastolic blood pressure, fasting glucose, and body mass index (BMI) prior to, during, and following childbearing years using generalized estimating equations, allowing for interaction between maltreatment and time within each segment, and adjusting for total parity, parental education, and race (Black or white, self-reported).

Understanding the role of childhood nurture, abuse, and stability on gestational diabetes in the Coronary Artery Risk Development in Young Adults study (CARDIA).

Background: To estimate associations between facets of the maternal childhood family environment with gestational diabetes (GDM) and to test mediation by pre-pregnancy waist circumference.

Methods: We used data from CARDIA, a cohort of individuals aged 18-30 years at baseline (1985-86), followed over 30 years (2016). We included participants with one or more pregnancies ≥ 20 weeks after baseline, without pre-pregnancy diabetes.

Types and timing of trauma exposure across the life course and maternal hypertension.

Objective: To estimate associations between types and timing (first occurrence and time since) of trauma exposure and hypertension experienced during pregnancy in a safety-net hospital in Atlanta, Georgia.

Methods: Participants completed a 14-item trauma screener. We linked that information to data from the medical record on hypertension (inclusive of chronic hypertension, gestational hypertension, or preeclampsia).

Postpartum healthcare expenditures for commercially insured deliveries with and without severe maternal morbidity.

Background: Although severe maternal morbidity is associated with adverse health outcomes in the year after delivery, patterns of healthcare use beyond the 6-week postpartum period have not been well documented.

Objective: This study aimed to estimate healthcare utilization and expenditures for deliveries with and without severe maternal morbidity in the 12 months following delivery among commercially insured patients.

Study Design: Using data from the 2016 to 2018 IBM Marketscan Commercial Claims and Encounters Research Databases, we identified deliveries to individuals 15 to 49 years of age who were continuously enrolled in noncapitated health plans for 12 months after delivery discharge.

Political representation and perinatal outcomes to Black, White, and Hispanic people in Georgia: a cross-sectional study.

Purpose: Our goal was to estimate differences in perinatal outcomes by racial differences in political representation, a measure of structural racism.

Methods: We gathered data on the racial composition of county-level elected officials for all counties in Georgia (n = 159) in 2022. We subtracted the percent of non-White elected officials from the percent of non-White residents to calculate the "representation difference," with greater positive values indicating a larger disparity.

Social Determinants of Health Needs and Perinatal Risk in Socially Vulnerable Pregnant Patients.

Objectives: To understand perinatal risks associated with social needs in pregnancy Methods. Multivariable log-binomial regression analyses adjusting for age, parity, and insurance were used to evaluate the relationship between any social need (e.g.

Interpretable machine learning to predict adverse perinatal outcomes: examining marginal predictive value of risk factors during pregnancy.

Background: The timely identification of nulliparas at high risk of adverse fetal and neonatal outcomes during pregnancy is crucial for initiating clinical interventions to prevent perinatal complications. Although machine learning methods have been applied to predict preterm birth and other pregnancy complications, many models do not provide explanations of their predictions, limiting the clinical use of the model.

Objective: This study aimed to develop interpretable prediction models for a composite adverse perinatal outcome (stillbirth, neonatal death, estimated Combined Apgar score of <10, or preterm birth) at different points in time during the pregnancy and to evaluate the marginal predictive value of individual predictors in the context of a machine learning model.

A Descriptive Study of Maternal Vaccination Uptake, Attitudes, and Beliefs in Pregnancy Among Persons Delivering at an Urban Safety Net Hospital.

Objective: To estimate uptake of influenza, tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis (Tdap), and COVID-19 vaccines during pregnancy and describe vaccine attitudes and beliefs among predominantly racial and ethnic minority individuals delivering at a publicly funded hospital.

Methods: We collected survey and electronic medical record data for English-speaking postpartum individuals who delivered a live-born infant from July 7, 2022, through August 21, 2022, and agreed to participate in our study. The 58-item survey included questions about general vaccine attitudes and beliefs as well as vaccine-specific questions.

The impact of screening for social risks on OBGYN patients and providers: A systematic review of current evidence and key gaps.

Background: Increasingly, policymakers and professional organizations support screening for social assets and risks during clinical care. Scant evidence exists on how screening impacts patients, providers, or health systems.

Objective: To systematically review published literature for evidence of the clinical utility of screening for social determinants of health in clinical obstetric and gynecologic (OBGYN) care.

Severe Maternal Morbidity and Postpartum Readmission Through 1 Year.

Objective: To conduct a cohort study to estimate risk for readmission through 1 year postpartum and the most common readmission diagnoses for individuals with and without severe maternal morbidity (SMM) at delivery.

Methods: Using national health care claims data from IBM MarketScan Commercial Research Databases (now known as Merative), we identified all delivery hospitalizations for continuously enrolled individuals 15-49 years of age that occurred between January 1, 2016, and December 31, 2018. Severe maternal morbidity at delivery was identified using diagnosis and procedure codes.

Association of Sickle Cell Disease With Severe Maternal Morbidity.

Objective: To evaluate the association between sickle cell disease (SCD) and severe maternal morbidity (SMM) in a contemporary cohort of deliveries by non-Hispanic Black people.

Methods: We retrospectively examined SMM by using electronic health record data on deliveries by non-Hispanic Black patients between 2011 and 2020 at a single tertiary, public institution. Sickle cell disease was identified during the delivery admission by using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes.

Timely treatment of severe hypertension and risk of severe maternal morbidity at an urban hospital.

Background: Hypertensive disorders of pregnancy have been identified as a leading contributor to severe maternal morbidity and mortality. Pregnant persons with hypertensive disorders who develop severe hypertension at delivery admission have been shown to experience higher rates of severe maternal morbidity relative to those without severe hypertension. Current guidelines recommend prompt treatment of severe hypertension given known associated maternal and fetal risks; however, only 1 previous study has described an association between timeliness of antihypertensive therapy and risk of severe maternal morbidity.

A Qualitative Study of Perceptions, Strengths, and Opportunities in Cardiometabolic Risk Management During Pregnancy and Postpartum in a Georgia Safety-Net Hospital, 2021.

Introduction: Despite the strong link between cardiometabolic pregnancy complications and future heart disease, there are documented gaps in engaging those who experience such conditions in recommended postpartum follow-up and preventive care. The goal of our study was to understand how people in a Medicaid-insured population perceive and manage risks during and after pregnancy related to an ongoing cardiometabolic disorder.

Methods: We conducted in-depth qualitative interviews with postpartum participants who had a cardiometabolic conditions during pregnancy (chronic or gestational diabetes, chronic or gestational hypertension, or preeclampsia).

Childhood trauma and postpartum care use, estimating mediation by posttraumatic stress disorder and depressive symptoms.

Purpose: To estimate the effect of childhood trauma on postpartum visit attendance and explore mediation by posttraumatic stress disorder (PTSD) and depressive symptoms.

Methods: We analyzed data from Grady Trauma Project surveys linked to electronic medical records. We measured childhood trauma using the Childhood Trauma Questionnaire and 4-12 week postpartum visit attendance using outpatient records.

Postpartum Psychiatric Outcomes following Severe Maternal Morbidity in an Urban Safety-Net Hospital.

Objective: Severe maternal morbidity (SMM) may be associated with postpartum psychiatric morbidity. However, the direction and strength of this relationship remain unclear. Our goal was to estimate the association between SMM and postpartum inpatient mental health care utilization.

Impact of the COVID-19 Pandemic on Obstetric Interventions at a Public Hospital.

Introduction: In response to the COVID-19 pandemic, health systems quickly implemented changes in care delivery with a goal of balancing patient-focused obstetric care with the need to protect pregnant persons and health care providers from infection. Yet, there is no consensus within the scientific community on the impact these measures have on obstetric outcomes in vulnerable populations. We aimed to assess the impact of the COVID-19 pandemic on rates of obstetric procedures and severe maternal morbidity (SMM) among births at an urban safety net institution.

Neighborhood deprivation and severe maternal morbidity in a medicaid-Insured population in Georgia.

Background: Despite growing acceptance of the role of context in shaping perinatal risk, data on how neighborhood factors may identify high-risk obstetric patients is limited. In this study, we evaluated the effect of neighborhood deprivation and neighborhood racial composition on severe maternal morbidity (SMM) among persons delivered in a large public health system in Atlanta, Georgia.

Methods: We conducted a population cohort study using electronic medical record data on all deliveries at Grady Memorial Hospital during 2011-2020.