Abstract

To examine the prevalence of severe maternal morbidity (SMM) and the indicators (i.e., morbidities and procedures) that comprise the SMM composite among adolescents and to compare risks between adolescents and adults.This was a population-based study in California to individuals aged 10-55. We included prenatal, postpartum, and birth hospital discharge records linked to vital records from 2001 to 2020. SMM was defined using the Centers for Disease Control and Prevention index. We estimated adjusted risk ratios (RRs) using modified Poisson regression models.This study included 702,481 adolescent births among 9,529,689 total births. There were 19 cases of SMM per 1,000 adolescent births; prevalence was highest among those aged 10-14 (27.5). Compared to adults aged 25-29, adolescents had higher rates of blood transfusion (11.4 vs. 9.0) and eclampsia (2.3 vs. 0.8). Crude RRs for SMM were 1.18 (95% confidence interval [CI]: 1.15-1.21), 1.22 (95% CI: 1.18-1.26), and 1.74 (95% CI: 1.53-1.99) in ages 18-19, 15-17, and 10-14, respectively. Adjustments for payer, race/ethnicity, parity, and anemia explained excess risk among ages 15-19, but not the youngest adolescents (10-14 years; RR: 1.31; 95% CI: 1.15-1.49). Patterns of SMM risk were similar but less pronounced when excluding blood transfusions. Risks were also elevated among ages 30-34, 35-39, and 40-55.Adolescents are at increased risk for SMM compared to adults, largely explained by social disadvantage, parity, and anemia, but not prepregnancy comorbidities. These findings highlight the need for resources to prevent SMM in adolescents and should be considered during care and public health planning.

View details for DOI 10.1016/j.jadohealth.2025.03.018

View details for PubMedID 40576605

Abstract

OBJECTIVES: Race-ethnicity and place-based variables such as rurality, ZIP code-level poverty, and county maternity care desert status have been associated with severe maternal morbidity and maternal mortality rates in the US. We examined these associations in Louisiana, which has one of the highest maternal mortality rates in the US.STUDY DESIGN: This was a birth cohort study.METHODS: We used state-wide inpatient birth hospitalization discharge data in Louisiana between 2016 and 2021. Mixed-effects logistic regression models with individuals nested within ZIP codes were used to estimate adjusted odds ratios (aOR) for non-transfusion severe maternal morbidity (nt-SMM, defined using the CDC index) according to race-ethnicity and place-based variables (overall and stratified by race-ethnicity).RESULTS: Among 326,597 birth hospitalizations, 2486 (0.77%) involved nt-SMM. Non-Hispanic Black and Hispanic individuals had increased risk of nt-SMM compared to Non-Hispanic White individuals, after adjustment for sociodemographic factors, a comorbidity index, and place-placed variables (aORs 1.36, 95%CI 1.23-1.51 and 1.39, 95%CI 1.21-1.59, respectively). Residence in a maternity care desert county or rural ZIP code did not increase risk; however, residence in a ZIP code in the highest quartile of poverty was associated with increased risk (aOR 1.26, 95%CI 1.04-1.51). When stratified by race and ethnicity, an increased risk remained for Non-Hispanic Black and Hispanic individuals residing in the highest-poverty ZIP codes (aORs 1.33, 95% CI 1.00-1.78 and 1.32, 95% CI 1.05-1.65, respectively), and a potential increased risk associated with living in a maternity care desert emerged for Black individuals (aOR 1.33, 95% CI 1.00-1.76), but confidence intervals included 1.00.CONCLUSIONS: In Louisiana, to reduce the rate of nt-SMM, social factors must be addressed, especially for non-Hispanic Black individuals living in areas with the highest levels of poverty and in maternity care deserts, as they had the highest risks in this population.

View details for DOI 10.1016/j.puhe.2025.105824

View details for PubMedID 40543228

Abstract

Perinatal mental disorders (PMD) affect birthing persons during pregnancy and postpartum. While racial and ethnic differences are documented, how structural racism influences these differences remains understudied.To estimate associations between the index of concentration at the extremes (ICE) -a proxy for anti-Black structural racism- and hospital-reported PMD, we analyzed data from a population-based cohort of all California live hospital births, 1997 to 2018 (N = 10,155, 036). PMD outcomes were identified from hospital discharge records throughout the perinatal period. Black-white ICE race-income was calculated and categorized into tertiles from most structurally deprived to privileged at the census tract level. Race and ethnicity-stratified mixed effects log-binomial models estimated the risk of hospital-reported PMD, adjusting for maternal age, education, and insurance, and accounting for clustering by census tract.In fully adjusted models, Black (aRR = 0.75, 95 % CI = 0.70, 0.81), Asian or Pacific Islander (aRR = 0.78, 95 % CI = 0.72, 0.84), and Hispanic (aRR = 0.65, 95 % CI = 0.62, 0.68) birthing persons living in the most structurally deprived neighborhoods had a reduced risk of hospital-reported PMD. Conversely, white (aRR = 1.09, 95 % CI = 1.05, 1.13) birthing persons living in the most structurally deprived neighborhoods had an increased risk of hospital-reported PMD.Findings reveal a complex association between racialized economic segregation and hospital-reported PMD. Living in structurally deprived neighborhoods might reflect underdiagnosis for minoritized populations or confer protection, while white counterparts can more readily access mental healthcare elsewhere. Further research may help inform place-based interventions aimed at improving perinatal mental health outcomes.

View details for DOI 10.1016/j.socscimed.2025.118296

View details for PubMedID 40544755

Abstract

Research on new-onset hypertensive disorders of pregnancy (HDP) and long-term maternal cardiovascular disease (CVD) death has focused on mothers of small-for-gestational-age infants rather than large-for-gestational-age infants.We further explored this focus by investigating CVD death in mothers with HDP by gestational age at delivery across the full spectrum of infant birth size.We used data from the Medical Birth Registry of Norway, the Norwegian National Population Register, and the Norwegian Cause of Death Registry, with information on mothers giving birth 1967-2020. This data was used to predict CVD death in the decades following pregnancy.We found the lowest CVD mortality among mothers with no HDP, term delivery, and a first infant with birthweight above average. These women constituted our reference group in the analyses. We found the highest risk of CVD death among mothers with preterm HDP and infants with above average birthweight for gestational age (HR 6.87, 95% CI 4.98, 9.48), not with infants below average birthweight for gestational age (HR 3.06, 95% CI 2.37, 3.93).There is an interactive association between HDP and large infant birthweight in preterm first births. The high risk associated with the particular combination of HDP, preterm birth, and high infant birthweight for gestational age warrants further research to understand its causal underpinnings.

View details for DOI 10.1111/ppe.70033

View details for PubMedID 40471648

Abstract

Orofacial clefts (OFC) are a common birth defect with few known risk factors (e.g., smoking). Maternal diet is associated with birth defects, though the relationship with OFC is less clear as studies typically only investigate single nutrients. We assessed the association between periconceptional maternal diet and OFC in the United States, using the National Birth Defects Prevention Study (1997-2011), including 3,649 cases (2,480 cleft lip with/without palate [CL/P] and 1,169 cleft palate [CP]) and 10,584 controls (infants without a birth defect). Using latent class analysis, we derived dietary patterns based on relative consumption of self-reported food items via food frequency data. We used logistic regression to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CIs) for the effect of dietary patterns on OFC. Four patterns were identified: Prudent (reference), Western, Low-Calorie Western, and Mexican. The Western diet had increased odds of CL/P (aOR: 1.3, CI: 1.2-1.5) and CP (aOR: 1.2, CI: 1.1-1.4). Low-Calorie Western (CL/P aOR: 1.2, CI: 1.0-1.4; CP aOR: 1.0, CI: 0.9-1.2) and Mexican diets (CL/P aOR: 1.1, CI: 0.9-1.3; CP aOR: 0.8, CI: 0.6-1.1) had a weaker or null association. Findings underscore the importance of further investigation into the effect of periconceptional diet on OFC occurrence.

View details for DOI 10.1093/aje/kwaf102

View details for PubMedID 40391741

Abstract

BACKGROUND: In the US, there is substantial variability in low-risk cesarean birth rate by hospitals and race/ethnicity. The contribution of inequitable hospital quality to disparities in low-risk cesarean births is uncertain. We examine the contribution of birth hospital to racial/ethnic disparities in low-risk cesarean births.METHODS: We used vital records linked with maternal birth hospitalization data (California, 2007-18). We examined self-reported race/ethnicity and low-risk cesarean birth, i.e., nulliparous, term, singleton, and vertex (NTSV) births. Poisson regression models with a mixed effect for hospital and bootstrapped errors were used to compare racial/ethnic differences in cesarean prevalence, adjusted for maternal and hospital characteristics. We used G-computation to assess how the prevalence of cesarean section by racial/ethnic group would change if all births occurred at the same distribution of hospitals as births to White individuals.RESULTS: Among 1,594,277 NTSV births at 212 hospitals, 26.9% were cesarean. After adjustment for hospital characteristics, risk ratios for cesarean birth ranged from 1.05 for foreign-born Hispanic (95% CI 1.02-1.09) to 1.28 for Black (95% CI 1.22-1.33) individuals, relative to White individuals. In the G-computation substitution, cesarean prevalence among NTSV births was reduced for some race/ethnicities and increased for others, ranging from 87 excess events (0.3% increase) in Black populations to 6473 avoided events (5.6% decrease) among US-born Hispanic populations.CONCLUSIONS: Racial/ethnic disparities in cesarean prevalence among low-risk births in California are not explained by individual-level maternal or hospital characteristics.

View details for DOI 10.1007/s40615-025-02464-z

View details for PubMedID 40327292

View details for Web of Science ID 001507428200065

View details for Web of Science ID 001507428200043

Abstract

Existing patient-reported outcome measures (PROMs) evaluating outpatient postpartum recovery lack content validity and were mostly not designed for this population. A Delphi process was performed, aiming to develop a patient-reported outcome measure for outpatient postpartum recovery and then evaluate it in a multicenter cohort study.Development of the Stanford Obstetric Recovery Checklist (STORK) involved 3 phases: (1) postpartum recovery questions were identified in published reviews; (2) after institutional review board approval, 16 multidisciplinary experts and patient stakeholders participated in 3 Delphi rounds (January 11 to April 12, 2021) to select items, resulting in the development of STORK (47 items; total score range, 0-188, with 0 indicating the worst recovery and 188 indicating the best recovery); and (3) cognitive debriefing interviews were conducted with 10 postpartum individuals to finalize STORK items. Individuals then completed STORK during their inpatient stay and at 2, 6, and 12 weeks post partum in a prospective, 3-center, US longitudinal cohort study conducted from June 13, 2022, to February 28, 2023. Recruitment occurred until 300 six-week STORK surveys were completed. STORK was evaluated at 6 weeks for validity (ability to measure recovery), reliability, and responsiveness. Validity included (1) structural validity (exploratory factor analysis using root mean square residual [RMSR]; <0.08 indicates a good fit); (2) convergent validity (correlation with global health visual analog scale score [GHVAS; scale, 0-100] and EuroQoL Five-Dimensions Three-Levels [EQ-5D-3L]); (3) discriminant validity (mean difference in STORK scores with GHVAS <70 vs ≥70); and (4) confirmatory telephone interviews with postpartum individuals scoring the highest and lowest 10th percentiles of STORK scores. Reliability (consistency of STORK scores) was evaluated using Cronbach α, interitem correlation, split-half reliability, and floor and ceiling effects. Responsiveness (ability of STORK to detect changes in recovery over time) was evaluated using percentage change in score from baseline to 12 weeks.A total of 525 individuals were recruited after all delivery modes (response rate, 62% [324 of 525] at 6 weeks); 498 (mean [SD] age, 33.3 [4.9] years) completed baseline inpatient postpartum surveys. STORK demonstrated validity: (1) a 4-factor model was the best fit (RMSR = 0.05); (2) correlation with GHVAS scores was ρ = 0.52 (95% CI, 0.43-0.61), and correlation with EQ-5D-3L scores was ρ = -0.67 (95% CI, -0.76 to -0.63); (3) STORK was able to discriminate between patients reporting good and poor recovery (good recovery: median STORK score, 151 [IQR, 136-163] vs poor recovery: median STORK score, 129 [IQR, 107-148]; P < .001); and (4) the highest and lowest scores corresponded to subjective assessments. STORK demonstrated reliability (Cronbach α = 0.92; interitem correlation r = 0.20; and split-half reliability ρ = 0.98). It also demonstrated responsiveness: percentage increases in overall STORK scores from baseline to week 12 were 19% after spontaneous vaginal delivery, 31% after operative vaginal delivery, 27% after scheduled cesarean delivery, and 20% after nonscheduled cesarean delivery (P < .001).In this cohort study of US individuals, STORK was found to be a valid, reliable, and responsive measure of outpatient postpartum recovery. Future clinical trials are needed to determine its clinical utility.

View details for DOI 10.1001/jamanetworkopen.2025.5713

View details for PubMedID 40244582

View details for DOI 10.1016/j.whi.2025.02.005

View details for PubMedID 40133145