Objective The aim of the study was to evaluate pregnant women's attitudes toward COVID-19 illness and vaccination and identify factors associated with vaccine acceptability. Study Design This was a cross-sectional survey among pregnant women enrolled in a prospective COVID-19 cohort study in Salt Lake City, UT, Birmingham, AL, and New York, NY, from August 9 to December 10, 2020. Women were eligible if they were 18 to 50 years old and <28 weeks of gestation. Upon enrollment, women completed surveys regarding concerns about COVID-19 illness and likelihood of getting COVID-19 vaccine if one were available during pregnancy. Vaccine acceptability was defined as a response of “very likely” or “somewhat likely” on a 4-point Likert scale. Factors associated with vaccine acceptability were assessed with multivariable logistic regression. Results Of 939 pregnant women eligible for the main cohort study, 915 (97%) consented to participate. Among these 915 women, 39% self-identified as White, 23% Black, 33% Hispanic, and 4% Other. Sixty-two percent received an influenza vaccine last season. Seventy-two percent worried about getting sick with COVID-19. If they were to get sick, 92% worried about harm to their pregnancy and 80% about harm to themselves. Only 41% reported they would get a vaccine. Of women who were unlikely to get vaccinated, the most frequently cited concern was vaccine safety for their pregnancy (82%). Non-Hispanic Black and Hispanic women had lower odds of accepting a vaccine compared with non-Hispanic White women (adjusted odds ratios [aOR] 0.4, 95% CI 0.2–0.6 for both). Receipt of influenza vaccine during the previous season was associated with higher odds of vaccine acceptability (aOR 2.1, 95% CI 1.5–3.0). Conclusion Although most pregnant women worried about COVID-19 illness, <50% were willing to get vaccinated during pregnancy. Racial and ethnic disparities in plans to accept COVID-19 vaccine highlight the need to prioritize strategies to address perceived barriers among groups at high risk for COVID-19. Key Points

Artificial intelligence can use real-world data to create models capable of making predictions and medical diagnosis for diabetes and its complications. The aim of this commentary article is to provide a general perspective and present recent advances on how artificial intelligence can be applied to improve the prediction and diagnosis of six significant complications of diabetes including (1) gestational diabetes, (2) hypoglycemia in the hospital, (3) diabetic retinopathy, (4) diabetic foot ulcers, (5) diabetic peripheral neuropathy, and (6) diabetic nephropathy.

OBJECTIVE: Continuous glucose monitoring (CGM) improves maternal glycemic control and neonatal outcomes in type 1 diabetes pregnancies compared with self-monitoring of blood glucose. However, CGM targets for pregnancy are based on expert opinion. We aimed to evaluate the association between CGM metrics and perinatal outcomes and identify evidence-based targets to reduce morbidity. RESEARCH DESIGN AND METHODS: This was a retrospective cohort study of pregnant patients with type 1 or 2 diabetes who used real-time CGM and delivered at a U.S. tertiary center (2018–2021). Multiple gestations, fetal anomalies, and early pregnancy loss were excluded. Exposures included time in range (TIR; 65–140 mg/dL), time above range (TAR), time below range (TBR), glucose variability, average glucose, and glucose management indicator. The primary outcome was a composite of fetal or neonatal mortality, large or small for gestational age at birth, neonatal intensive care unit admission, hypoglycemia, shoulder dystocia or birth trauma, and hyperbilirubinemia. Logistic regression estimated the association between CGM metrics and outcomes, and optimal TIR was calculated. RESULTS: Of 117 patients, 16 (13.7%) used CGM before pregnancy and 68 (58.1%) had type 1 diabetes. Overall, 98 patients (83.8%) developed the composite neonatal outcome. All CGM metrics, except TBR, were associated with neonatal morbidity. For each 5 percentage-point increase in TIR, there was 28% reduced odds of neonatal morbidity (odds ratio 0.72, 95% CI 0.58–0.89). The statistically optimal TIR was 66–71%. CONCLUSIONS: Nearly all CGM metrics were associated with adverse neonatal morbidity and mortality and may aid management of preexisting diabetes in pregnancy. Our findings support the American Diabetes Association recommendation of 70% TIR.