Artificial Intelligence-Augmented Clinical Decision Support Systems for Pregnancy Care: Systematic Review.

BACKGROUND: Despite the emerging application of clinical decision support systems (CDSS) in pregnancy care and the proliferation of artificial intelligence (AI) over the last decade, it remains understudied regarding the role of AI in CDSS specialized for pregnancy care. OBJECTIVE: To identify and synthesize AI-augmented CDSS in pregnancy care, CDSS functionality, AI methodologies, and clinical implementation, we reported a systematic review based on empirical studies that examined AI-augmented CDSS in pregnancy care. METHODS: We retrieved studies that examined AI-augmented CDSS in pregnancy care using database queries involved with titles, abstracts, keywords, and MeSH (Medical Subject Headings) terms. Bibliographic records from their inception to 2022 were retrieved from PubMed/MEDLINE (n=206), Embase (n=101), and ACM Digital Library (n=377), followed by eligibility screening and literature review. The eligibility criteria include empirical studies that (1) developed or tested AI methods, (2) developed or tested CDSS or CDSS components, and (3) focused on pregnancy care. Data of studies used for review and appraisal include title, abstract, keywords, MeSH terms, full text, and supplements. Publications with ancillary information or overlapping outcomes were synthesized as one single study. Reviewers independently reviewed and assessed the quality of selected studies. RESULTS: We identified 30 distinct studies of 684 studies from their inception to 2022. Topics of clinical applications covered AI-augmented CDSS from prenatal, early pregnancy, obstetric care, and postpartum care. Topics of CDSS functions include diagnostic support, clinical prediction, therapeutics recommendation, and knowledge base. CONCLUSIONS: Our review acknowledged recent advances in CDSS studies including early diagnosis of prenatal abnormalities, cost-effective surveillance, prenatal ultrasound support, and ontology development. To recommend future directions, we also noted key gaps from existing studies, including (1) decision support in current childbirth deliveries without using observational data from consequential fetal or maternal outcomes in future pregnancies; (2) scarcity of studies in identifying several high-profile biases from CDSS, including social determinants of health highlighted by the American College of Obstetricians and Gynecologists; and (3) chasm between internally validated CDSS models, external validity, and clinical implementation.

Rural-urban and racial differences in cesarean deliveries before and during the COVID-19 pandemic in South Carolina.

PROBLEM: Unnecessary cesarean delivery increases the risk of complications for birthing people and infants. BACKGROUND: Examining the intersectionality of rural and racial disparities in low-risk cesarean delivery is necessary to improve equity in quality obstetrics care. AIM: To evaluate rural and racial/ethnic differences in Nulliparous, Term, Singleton, Vertex (NTSV) and primary cesarean delivery rates before and during the COVID-19 pandemic in South Carolina. METHODS: This retrospective cohort study used birth certificates linked to all-payer hospital discharge data for South Carolina childbirths from 2018 to 2021. Multilevel logistic regressions examined differences in cesarean outcomes by rural/urban hospital location and race/ethnicity of birthing people during pre-pandemic (January 2018-February 2020) and peri-pandemic periods (March 2020-December 2021), adjusting for maternal, infant, and hospital characteristics among two low-risk pregnancy cohorts: 1) Nulliparous, Term, Singleton, Vertex (NTSV, n = 65,974) and 2) those without prior cesarean (primary, n = 167,928). FINDINGS: Black vs. White disparities remained for NTSV cesarean in adjusted models (urban pre-pandemic aOR = 1.34, 95 %CI 1.23-1.46) but were not significantly different for primary cesarean, apart from rural settings peri-pandemic (aOR = 0.87, 95 %CI 0.79-0.96). Hispanic individuals had higher adjusted odds of NTSV cesarean only for rural settings pre-pandemic (aOR = 1.28, 95 %CI 1.05-1.56), but this disparity was not significant during the pandemic (aOR = 1.13, 95 %CI 0.93-1.37). DISCUSSION AND CONCLUSION: Observed rural and racial/ethnic disparities in cesarean delivery outcomes were present before and during the COVID-19 pandemic. Strategies effective in reducing racial disparities in primary cesarean may be useful in also reducing Black vs. White NTSV cesarean disparities.

Employing digital PCR for enhanced detection of perinatal Toxoplasma gondii infection: A cross-sectional surveillance and maternal-infant outcomes study in El Salvador.

Toxoplasma gondii is a parasitic infection that can be transmitted in utero, resulting in fetal chorioretinitis and other long-term neurological outcomes. If diagnosed early, pregnancy-safe chemotherapeutics can prevent vertical transmission. Unfortunately, diagnosis of acute, primary infection among pregnant women remains neglected, particularly in low-and-middle-income countries. Clinically actionable diagnosis is complex due to the commonality of infection during childhood and early adulthood which spawn long-last antibody titers and historically unreliable direct molecular diagnostics. The current study employed a cross-sectional T. gondii perinatal surveillance study using digital PCR, a next generation molecular diagnostic platform, and a maternal-fetal outcomes survey to ascertain the risk of vertical toxoplasmosis transmission in the Western Region of El Salvador. Of 198 enrolled mothers at the time of childbirth, 6.6% had evidence of recent T. gondii infection-85% of these cases were identified using digital PCR. Neonates born to these acutely infected mothers were significantly more likely to meconium aspiration syndrome and mothers were more likely to experience labor and delivery complications. Multivariable logistic regression found higher maternal T. gondii infection odds were associated with the presence of pet cats, the definitive T. gondii host. In closing, this study provides evidence of maternal T. gondii infection, vertical transmission and deleterious fetal outcomes in a vulnerable population near the El Salvador-Guatemala border. Further, this is the first published study to show clinical utility potential of digital PCR for accurate diagnosis of congenital toxoplasmosis cases.

Perinatal dengue and Zika virus cross-sectional seroprevalence and maternal-fetal outcomes among El Salvadoran women presenting for labor-and-delivery.

BACKGROUND: Despite maternal flavivirus infections' linkage to severe maternal and fetal outcomes, surveillance during pregnancy remains limited globally. Further complicating maternal screening for these potentially teratogenic pathogens is the overwhelming subclinical nature of acute infection. This study aimed to understand perinatal and neonatal risk for poor health outcomes associated with flaviviral infection during pregnancy in El Salvador. METHODS: Banked serologic samples and clinical results obtained from women presenting for labor and delivery at a national referent hospital in western El Salvador March to September 2022 were used for this study. 198 samples were screened for dengue and Zika virus IgM, and statistical analyses analyzed demographic and clinical outcome associations with IgM positivity. RESULTS: This serosurvey revealed a high rate of maternal flavivirus infection-24.2% of women presenting for labor and delivery were dengue or Zika virus IgM positive, suggesting potential infection within pregnancy. Specifically, 20.2% were Zika virus IgM positive, 1.5% were dengue virus IgM positive, and 2.5% were both dengue and Zika virus IgM positive. Women whose home had received mosquito abatement assistance within the last year by the ministry of health were 70% less likely to test IgM positive (aOR = 0.30, 95%CI: 0.10, 0.83). Further, statistical geospatial clustering revealed transmission foci in six primary municipalities. Pregnancy complications and poor birth outcomes were noted among the dengue and/or Zika virus maternal infection group, although these outcomes were not statistically different than the seronegative group. None of the resulting neonates born during this study were diagnosed with congenital Zika syndrome. CONCLUSIONS: The high rate of Zika virus detected among pregnant women and the lack of Zika-specific neonatal outcomes monitoring during a non-outbreak year highlights the need for continued surveillance in Central America and among immigrant mothers presenting for childbirth from these countries. As changing climatic conditions continue to expand the range of the disease vector, asymptomatic screening programs could be vital to early identification of outbreaks and clinical management of cases.

Severe maternal morbidity by race and ethnicity before vs. during the COVID-19 pandemic.

PURPOSE: To examine the change in racial disparity in severe maternal morbidity (SMM) during the COVID-19 pandemic and the associations between SARS-CoV-2 infection and SMM. METHODS: This retrospective cohort study used linked databases of all livebirths delivered between 2018 and 2021 in South Carolina (n = 162,576). Exposures were 1) pre-pandemic and pandemic periods (before vs. March 2020 onwards); 2) SARS-CoV-2 infection, severity, and timing of first infection. Log-binomial regression models were used. RESULTS: SMM rate was higher among pandemic childbirths than pre-pandemic period (p = 0.06). The risk of SMM among Hispanics was doubled from pre-pandemic to pandemic periods (adjusted relative risk (aRR)= 2.50, 95% CI: 1.27, 4.94). During pre-pandemic, compared to White women, Black women (aRR=1.37, 95% CI: 1.14-1.64), while Hispanics had lower risk of SMM (aRR=0.42, 95% CI: 0.24-0.73). During the pandemic, the Black-White difference in the risk of SMM persisted (aRR=1.24, 95% CI: 1.00-1.54) and Hispanic-White difference in SMM risk became insignificant (aRR=0.85, 95% CI: 0.54-1.34). SARS-CoV-2 infection, its severity, and the late diagnosis were associated with 1.78-5.06 times higher risk of SMM. CONCLUSIONS: During pandemic, Black-White racial disparity in SMM persisted but the relative pre-pandemic advantage in SMM among Hispanic women over White women disappeared during the pandemic.

Dual Barriers: Examining Digital Access and Travel Burdens to Hospital Maternity Care Access in the United States, 2020.

Policy Points The White House Blueprint for Addressing the Maternal Health Crisis report released in June 2022 highlighted the need to enhance equitable access to maternity care. Nationwide hospital maternity unit closures have worsened the maternal health crisis in underserved communities, leaving many birthing people with few options and with long travel times to reach essential care. Ensuring equitable access to maternity care requires addressing travel burdens to care and inadequate digital access. Our findings reveal socioeconomically disadvantaged communities in the United States face dual barriers to maternity care access, as communities located farthest away from care facilities had the least digital access. CONTEXT: With the increases in nationwide hospital maternity unit closures, there is a greater need for telehealth services for the supervision, evaluation, and management of prenatal and postpartum care. However, challenges in digital access persist. We examined associations between driving time to hospital maternity units and digital access to understand whether augmenting digital access and telehealth services might help mitigate travel burdens to maternity care. METHODS: This cross-sectional study used 2020 American Hospital Association Annual Survey data for hospital maternity unit locations and 2020 American Community Survey five-year ZIP Code Tabulation Area (ZCTA)-level estimates of household digital access to telecommunication technology and broadband. We calculated driving times of the fastest route from population-weighted ZCTA centroids to the nearest hospital maternity unit. Rural-urban stratified generalized median regression models were conducted to examine differences in ZCTA-level proportions of household lacking digital access equipment (any digital device, smartphones, tablet), and lacking broadband subscriptions by spatial accessibility to maternity units. FINDINGS: In 2020, 2,905 (16.6%) urban and 3,394 (39.5%) rural ZCTAs in the United States were located >30 minutes from the nearest hospital maternity units. Regardless of rurality, these communities farther away from a maternity unit had disproportionally lower broadband and device accessibility. Although urban communities have greater digital access to technology and broadband subscriptions compared to rural communities, disparities in the percentage of households with access to digital devices were more pronounced within urban areas, particularly between those with and without close proximity to a hospital maternity unit. Communities where nearest hospital maternity units were >30 minutes away had higher poverty and uninsurance rates than those with <15-minute access. CONCLUSIONS: Socioeconomically disadvantaged communities face significant barriers to maternity care access, both with substantial travel burdens and inadequate digital access. To optimize maternity care access, ongoing efforts (e.g., Affordable Connectivity Program introduced in the 2021 Infrastructure Act), should bridge the gaps in digital access and target communities with substantial travel burdens to care and limited digital access.

Chagas Disease Maternal Seroprevalence and Maternal-Fetal Health Outcomes in a Parturition Cohort in Western El Salvador.

Congenital Chagas disease is a growing concern, prioritized by the World Health Organization for public health action. El Salvador is home to some of the highest Chagas disease (Trypanosoma cruzi infection) burdens in the Americas, yet pregnancy screening remains neglected. This pilot investigation performed a maternal T. cruzi surveillance study in Western El Salvador among women presenting for labor and delivery. From 198 consented and enrolled pregnant women, 6% were T. cruzi positive by serology or molecular diagnosis. Half of the infants born to T. cruzi-positive women were admitted to the NICU for neonatal complications. Geospatial statistical clustering of cases was noted in the municipality of Jujutla. Older women and those knowing an infected relative or close friend were significantly more likely to test positive for T. cruzi infection at the time of parturition. In closing, maternal T. cruzi infections were significantly higher than national HIV or syphilis maternal rates, creating an urgent need to add T. cruzi to mandatory pregnancy screening programs.

Risk for stillbirth among pregnant individuals with SARS-CoV-2 infection varied by gestational age.

BACKGROUND: Despite previous research findings on higher risks of stillbirth among pregnant individuals with SARS-CoV-2 infection, it is unclear whether the gestational timing of viral infection modulates this risk. OBJECTIVE: This study aimed to examine the association between timing of SARS-CoV-2 infection during pregnancy and risk of stillbirth. STUDY DESIGN: This retrospective cohort study used multilevel logistic regression analyses of nationwide electronic health records in the United States. Data were from 75 healthcare systems and institutes across 50 states. A total of 191,403 pregnancies of 190,738 individuals of reproductive age (15-49 years) who had childbirth between March 1, 2020 and May 31, 2021 were identified and included. The main outcome was stillbirth at ≥20 weeks of gestation. Exposures were the timing of SARS-CoV-2 infection: early pregnancy (<20 weeks), midpregnancy (21-27 weeks), the third trimester (28-43 weeks), any time before delivery, and never infected (reference). RESULTS: We identified 2342 (1.3%) pregnancies with COVID-19 in early pregnancy, 2075 (1.2%) in midpregnancy, and 12,697 (6.9%) in the third trimester. After adjusting for maternal and clinical characteristics, increased odds of stillbirth were observed among pregnant individuals with SARS-CoV-2 infection only in early pregnancy (odds ratio, 1.75, 95% confidence interval, 1.25-2.46) and midpregnancy (odds ratio, 2.09; 95% confidence interval, 1.49-2.93), as opposed to pregnant individuals who were never infected. Older age, Black race, hypertension, acute respiratory distress syndrome or acute respiratory failure, and placental abruption were found to be consistently associated with stillbirth across different trimesters. CONCLUSION: Increased risk of stillbirth was associated with COVID-19 only when pregnant individuals were infected during early and midpregnancy, and not at any time before the delivery or during the third trimester, suggesting the potential vulnerability of the fetus to SARS-CoV-2 infection in early pregnancy. Our findings underscore the importance of proactive COVID-19 prevention and timely medical intervention for individuals infected with SARS-CoV-2 during early and midpregnancy.

TORCH Congenital Syndrome Infections in Central America's Northern Triangle.

TORCH pathogens are a group of globally prevalent infectious agents that may cross the placental barrier, causing severe negative sequalae in neonates, including fetal death and lifelong morbidity. TORCH infections are classically defined by Toxoplasma gondii, other infectious causes of concern (e.g., syphilis, Zika virus, malaria, human immunodeficiency virus), rubella virus, cytomegalovirus, and herpes simplex viruses. Neonatal disorders and congenital birth defects are the leading causes of neonatal mortality in Central America's Northern Triangle, yet little is known about TORCH congenital syndrome in this region. This review synthesizes the little that is known regarding the most salient TORCH infections among pregnant women and neonates in Central America's Northern Triangle and highlights gaps in the literature that warrant further research. Due to the limited publicly available information, this review includes both peer-reviewed published literature and university professional degree theses. Further large-scale studies should be conducted to clarify the public health impact these infections in this world region.

Temporal Events Detector for Pregnancy Care (TED-PC): A rule-based algorithm to infer gestational age and delivery date from electronic health records of pregnant women with and without COVID-19.

OBJECTIVE: Identifying the time of SARS-CoV-2 viral infection relative to specific gestational weeks is critical for delineating the role of viral infection timing in adverse pregnancy outcomes. However, this task is difficult when it comes to Electronic Health Records (EHR). In combating the COVID-19 pandemic for maternal health, we sought to develop and validate a clinical information extraction algorithm to detect the time of clinical events relative to gestational weeks. MATERIALS AND METHODS: We used EHR from the National COVID Cohort Collaborative (N3C), in which the EHR are normalized by the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). We performed EHR phenotyping, resulting in 270,897 pregnant women (June 1st, 2018 to May 31st, 2021). We developed a rule-based algorithm and performed a multi-level evaluation to test content validity and clinical validity, and extreme length of gestation (<150 or >300). RESULTS: The algorithm identified 296,194 pregnancies (16,659 COVID-19, 174,744 without COVID-19) in 270,897 pregnant women. For inferring gestational age, 95% cases (n = 40) have moderate-high accuracy (Cohen's Kappa = 0.62); 100% cases (n = 40) have moderate-high granularity of temporal information (Cohen's Kappa = 1). For inferring delivery dates, the accuracy is 100% (Cohen's Kappa = 1). The accuracy of gestational age detection for the extreme length of gestation is 93.3% (Cohen's Kappa = 1). Mothers with COVID-19 showed higher prevalence in obesity or overweight (35.1% vs. 29.5%), diabetes (17.8% vs. 17.0%), chronic obstructive pulmonary disease (0.2% vs. 0.1%), respiratory distress syndrome or acute respiratory failure (1.8% vs. 0.2%). DISCUSSION: We explored the characteristics of pregnant women by different gestational weeks of SARS-CoV-2 infection with our algorithm. TED-PC is the first to infer the exact gestational week linked with every clinical event from EHR and detect the timing of SARS-CoV-2 infection in pregnant women. CONCLUSION: The algorithm shows excellent clinical validity in inferring gestational age and delivery dates, which supports multiple EHR cohorts on N3C studying the impact of COVID-19 on pregnancy.