(835) Health system level intervention to decrease early screening rates for gestational diabetes: associated perinatal outcomes

Friday, January 31, 2025

10:30 AM – 12:30 PM

Submitting Author and Presenting Author(s)

Mara Greenberg, MD (she/her/hers)

Clinical and Research Director, Regional Perinatal Service Center
Kaiser-Permanente Northern California
Oakland, California, United States

Coauthor(s)

YZ

Yeyi Zhu, PhD

Senior Research Scientist
Kaiser Permanente Northern California
Oakland, California, United States
JS

Jun Shan, N/A

Kaiser Permanente Northern California
Oakland, California, United States
AN

Amanda Ngo, MS

Biostatistician
Kaiser Permanente Northern California
Oakland, California, United States
Monique Hedderson, PhD

Associate Director, Women’s and Children’s Health Section. Director of Research Clinics
Kaiser Permanente Northern California
Pleasanton, California, United States
CQ

Charles Quesenberry, PhD

Kaiser Permanente Northern California
Oakland, California, United States
Assiamira Ferrara, MD, PhD (she/her/hers)

Senior Investigator
Kaiser Permanente Northern California
Oakland, California, United States

Objective: To evaluate a system level intervention intended to decrease rates of gestational diabetes (GDM) screening prior to 24 weeks gestation in relation to perinatal outcomes.

Study Design: Retrospective study of all pregnancies without overt diabetes delivered in Kaiser Permanente Northern California 2018-2022. We performed interrupted time series (ITS) analysis to examine changes in rates of GDM screening and perinatal outcomes in relation to a system-wide intervention to decrease screening < 24 weeks. To reflect the level of exposure to the intervention that started in 4/2020, dates of delivery were categorized in 3 time periods: T1 unexposed (1/2018-3/2020), T2 partially exposed (4/2020-12/2020), and T3 fully exposed (1/2021-12/2022). Primary composite outcome included cesarean delivery, preeclampsia, severe maternal morbidity, large for gestational age neonate, shoulder dystocia, preterm birth, NICU admission, and neonatal hypoglycemia. Patient level factors were examined as potential confounders.

Results: Among 221,068 deliveries, early GDM screening rate decreased from 31.1% in T1, to 20.6% in T2 and 4.3% in T3 (Standardized Mean Difference (SMD) T3 vs T1 -0.75). Patient level characteristics did not vary significantly over time, including measures of social drivers of health and comorbid medical conditions. There was no change in the prevalence of GDM (9.7% in both T1 and T2, 9.0% in T3, SMD T3 vs T1 -0.02). There was no change in the prevalence of the composite primary outcome, which occurred in 43.1% in T1, 44.2% in T2 and 45.2% in T3 (SMD T3 vs T1 0.04). Interrupted time series analysis adjusted for covariates showed no change in risk of the composite outcome during T1 (percent change per 4 weeks [95% CI]): 0.06 [-0.01 to 0.13]), T2 (0.02 [-0.21 to 0.25]), nor T3 (-0.02 [-0.26 to 0.23]). No meaningful changes in prevalence of each individual component of the composite outcome were observed.

Conclusion: A substantial decrease in early GDM screening rate did not impact perinatal outcomes in this large diverse cohort of deliveries within an integrated health care system.