DECIDE: A Comparative Effectiveness Trial of Metformin Versus Insulin for the Treatment of Gestational Diabetes

Inclusion Criteria: * Singleton gestation. Twin reduction to singleton, either spontaneously or therapeutically, is eligible if it occurred before 14 weeks gestational age. * Age 18 years or older * Gestational age at randomization between 20 0/7 - 33 6/7 weeks based on project gestational age. * GDM diagnosis less than or equal to 33 6/7 weeks based on project gestational age. * Requires medication for glucose control defined as ≥ 30% elevated glucose values (either fasting or postprandial or both) prior to randomization per determination of the provider or documented in the medical record. * Patient willingness and ability to attend 2-year follow-up visit. Exclusion criteria: * Renal disease (serum creatinine \>1.3 mg/dL) due to the potential impact of metformin on renal function. * Major structural malformation of the fetus. * Known fetal aneuploidy based on invasive testing or positive for aneuploidy on cell-free fetal DNA screening. * Contraindication to metformin or insulin, including: history of lactic acidosis, intractable nausea and vomiting, prior documented allergy and/or anaphylaxis. * For individuals with GDM diagnosed \<20 0/7 weeks, documented A1c ≥\>6.5% within prior 6 months. * Pregestational diabetes documented in the medical record or prior A1c\>= 6.5% * Fasting hyperglycemia \>115 mg/dl for ≥ 50% of fasting glucose values in the past week (due to the high risk of metformin failure with fasting hyperglycemia). * Enrolled in a trial that influences primary study outcomes (composite neonatal outcome at delivery or childhood body mass index at 2 years). * Prenatal care or delivery planned at a location where access to the complete electronic medical record will not be available to research staff. * Language barrier (appropriate translation resources unavailable at the site) * Participation in this trial in a previous pregnancy. Patients who were screened in a previous pregnancy, but not randomized, may be included.