In 2017, the Maternal Fetal Medicine Units (MFMU) Network disseminated a Trial of Labor after Cesarean (TOLAC) calculator. An updated (2021) version of the calculator does not include race and ethnicity, but instead focuses on prior medical and OB history, in addition to age and BMI. The objective of this study was to compare the actual vs predicted VBAC rates between the old and new MFMU calculators.

Background

Ultrasound training is a vital component of maternal-fetal medicine fellowships in the United States. Of the 18 months of core clinical training, the American Board of Obstetrics and Gynecology currently requires a minimum of 3 months to be dedicated to ultrasound to be eligible for board certification. However, the experience and degree of hands-on training differ among the fellowship programs and have not been reassessed for nearly a decade.

Objective

To assess regional heterogeneity in the ultrasound training experience during maternal-fetal medicine fellowship in the United States.

Study design

A survey was distributed to postgraduate year (PGY)-6 maternal-fetal medicine fellows registered to attend an annual ultrasound training course before the conference (n=114). For programs with >1 fellow attending (n=39), only 1 of them completed the survey to represent the program. The questions included demographics of the program, ultrasound training structure, the fellows' self-perception of ultrasound capabilities, research, mentorship, and technical aspects of sonography.

Results

Seventy two postgraduate year 6 fellows with a wide geographic distribution as follows completed the survey (96% response rate): 10 (14%) from the West, 16 (22%) from the Midwest, 17 (24%) from the South, and 29 (40%) from the Northeast. Respondents undergoing training in the South were less likely to report feeling comfortable performing nuchal translucency and detailed anatomic surveys than those from other regions (nuchal translucency: P=.046; anatomy: P=.011). Most of the respondents reported feeling comfortable performing growth (78%) and umbilical artery Doppler (58%) and feeling uncomfortable with three-dimensional ultrasound, neurosonography, and fetal echocardiography. Respondents in the Northeast were more likely to report feeling comfortable performing chorionic villus sampling (P=.001). There was no difference among fellowship programs in the presence or absence of ultrasound curriculum, bedside teaching, ultrasound-focused research mentorship, or months of ultrasound training.

Conclusion

Despite the standardization of ultrasound training structure across the United States, there remains regional heterogeneity in fellow self-reported comfort with specific ultrasound techniques and chorionic villus sampling at a midpoint in their fellowship training. The maternal-fetal medicine attending involvement at the bedside did not affect the fellow self-reported comfort with ultrasound surveys. This study highlights the need for further optimization of maternal-fetal medicine fellowship ultrasound training, especially in advanced sonography and diagnostic procedures.

Although prematurity is the single largest cause of death in children under 5 years of age, the current definition of prematurity, based on gestational age, lacks the precision needed for guiding care decisions. Here, we propose a longitudinal risk assessment for adverse neonatal outcomes in newborns based on a deep learning model that uses electronic health records (EHRs) to predict a wide range of outcomes over a period starting shortly before conception and ending months after birth. By linking the EHRs of the Lucile Packard Children's Hospital and the Stanford Healthcare Adult Hospital, we developed a cohort of 22,104 mother-newborn dyads delivered between 2014 and 2018. Maternal and newborn EHRs were extracted and used to train a multi-input multitask deep learning model, featuring a long short-term memory neural network, to predict 24 different neonatal outcomes. An additional cohort of 10,250 mother-newborn dyads delivered at the same Stanford Hospitals from 2019 to September 2020 was used to validate the model. Areas under the receiver operating characteristic curve at delivery exceeded 0.9 for 10 of the 24 neonatal outcomes considered and were between 0.8 and 0.9 for 7 additional outcomes. Moreover, comprehensive association analysis identified multiple known associations between various maternal and neonatal features and specific neonatal outcomes. This study used linked EHRs from more than 30,000 mother-newborn dyads and would serve as a resource for the investigation and prediction of neonatal outcomes. An interactive website is available for independent investigators to leverage this unique dataset: https://maternal-child-health-associations.shinyapps.io/shiny_app/.