Randomizing Two Gestational Diabetes Screening Methods in a Diverse HMO

This project randomizes two different screening strategies for diabetes in pregnancy, among a study population of over 17,500 pregnant women and their babies (over 35,000 total) in a large diverse health maintenance organization (HMO), to determine how diagnosis and treatment based on these two strategies in routine clinical care affects complications for the baby and the mother.

Two recent randomized placebo-controlled trials show that gestational diabetes (GDM) treatment (vs. none) improves maternal and perinatal outcomes, based on diagnosis with a 2- step screening strategy. Also, a large multi-center prospective cohort study showed a linear relationship with glucose and maternal and perinatal outcomes, based on screening with a single 75g oral glucose tolerance test (OGTT). Based on this large cohort's findings, the American Diabetes Association recommended that clinical practice adopt the 1-step 75g screening approach for diagnosing GDM. The American College of Obstetrics & Gynecology took the opposite stance, recommending the traditional 2-step screening: because it alone has RCT outcome evidence. What is urgently needed to best inform clinical practice and health policy is not an additional GDM treatment vs. control trial, but a pragmatic randomized controlled trial (RCT) testing the 2 recommended clinical strategies. To pragmatically address this critical research gap, we propose to randomize an estimated 17,626 diverse women to GDM screening (2-step vs. 75g OGTT) as part of their clinical care in the Kaiser Permanente Northwest (KPNW) and Hawaii (KPH) regional health plans. The investigators will use the plans' electronic medical record (EMR) system at the time of GDM screening to randomize the women. Both KPNW and KPH regions universally screen for GDM at 24-28 weeks gestation, as part of clinical care. By randomizing GDM screening in the context of clinical care, the investigators will: Compare GDM prevalences (Aim 1) and differences in maternal and perinatal outcomes between screening strategies (Aim 2). Determine the concordance of the 75g OGTT with GDM diagnosed by 2-step, among a recruited sub-sample of 1,000 pregnant women at KPNW and KPH (Aim 3). The results of this pragmatic RCT are expected to help resolve the current public policy debate on the potential benefits and risks of each strategy in clinical obstetric practice.

Diagnosis of GDM based on laboratory values for each screening approach (1-step or 2-step) as planned in the original protocol.

Includes any of the following: number of neonatal deaths, stillbirths, shoulder dystocia, bone fracture, or nerve palsy

Will evaluate macrosomia, large for gestational age (LGA), small for gestational age (SGA) and average birthweight. LGA remains a primary outcome.

Number of pregnancies for which newborn has a diagnosis of neonatal respiratory distress syndrome; planned in the original protocol.

Number of pregnancies for which newborn has a diagnosis of jaundice and received jaundice treatment; planned in the original protocol.

Number of pregnancies for which newborn has a diagnosis of neonatal hypoglycemia; planned in the original protocol.

Safety outcome. Number of pregnancies in which delivery took place before 37 weeks of gestation; separately, number or pregnancies in which delivery took place before 32 weeks of gestation

Maternal height and weight measurements to evaluate average weight retention post-partum compared to pre-pregnancy

This pragmatic GDM screening RCT evaluates rates of multiple maternal and child GDM-associated outcomes with two standard-of-care screening strategies in a large population of pregnant women at two diverse HMO sites with routine universal screening for GDM. This first primary outcome encompasses the overarching goals of our pragmatic RCT, and other specific primary outcomes will be listed as subsequent primary outcome measures. This outcome was initially registered as an overall primary outcome and was intended (as stated in the study protocol) to include the number of pregnancies diagnosed with GDM based on laboratory values of the two screening approaches; the primary GDM diagnosis has been updated separately as a primary outcome as per protocol.

We will evaluate childhood growth percentiles and trajectories, as well as incidence of childhood overweight and obesity

Based on ICD-10 diagnoses and also measured components including blood pressure, lipid measurements, body mass index, and diabetes

Inclusion Criteria: pregnant adult women in KPNW and KPH Exclusion Criteria: pre-existing diabetes

Hillier TA, Pedula KL, Ogasawara KK, Vesco KK, Oshiro CES, Lubarsky SL, Van Marter J. Further implications from a pragmatic randomized clinical trial of gestational diabetes screening: per-protocol and as-treated estimates. Am J Obstet Gynecol. 2021 Nov;225(5):581-583. doi: 10.1016/j.ajog.2021.08.006. Epub 2021 Aug 9.

Hillier TA, Pedula KL, Ogasawara KK, Vesco KK, Oshiro CES, Lubarsky SL, Van Marter J. A Pragmatic, Randomized Clinical Trial of Gestational Diabetes Screening. N Engl J Med. 2021 Mar 11;384(10):895-904. doi: 10.1056/NEJMoa2026028.