Patient-led insulin titration in GDM linked to faster glycemic control and lower macrosomia risk in small RCT

OBJECTIVE: To conduct a randomized controlled trial (RCT) to assess whether patient-led insulin titration (intervention) compared with clinician-led insulin titration (control) resulted in improved glycemic management and pregnancy outcomes for individuals with gestational diabetes mellitus (GDM). METHODS: EMPOWER (Patient Versus Provider-Led Titration of Insulin for Glycemic Control in Gestational Diabetes) was a single-center, nonblinded RCT among individuals with GDM requiring insulin between 20 and 32 weeks of gestation that was conducted from October 19, 2023, to January 10, 2025. Intervention participants self-titrated long-acting insulin, which was started at 10 units nightly and decreased or increased by 2 units per fasting glucose above or below 70 and 95 mg/dL, respectively, every day. Control participants started an insulin dose with weekly titration at the clinician's discretion. The primary outcome was mean fasting glucose in the 36th week or the week before delivery for preterm deliveries. Secondary outcomes included pregnancy outcomes and patient-reported measures. With 80% power, two-sided α of 0.05, and 5% loss-to-follow-up, 56 individuals needed to be randomized to demonstrate at least a 15% difference in mean fasting glucose. Analysis was by intention to treat. RESULTS: Of 89 individuals who were eligible during the study period, 56 consented and were randomized (29 intervention, 27 control). The median duration from starting insulin to delivery was 7.7 weeks. Patient-led insulin titration resulted in a similar mean fasting glucose before delivery compared with clinician-led titration (88.8 vs 90.3 mg/dL; β coefficient -1.50 mg/dL, 95% CI, -5.50 to 2.50) but resulted in more rapid achievement of fasting glucose below 95 mg/dL (mean 1.8 weeks vs 2.5 weeks; hazard ratio 1.48, 95% CI, 1.16 to 1.90). Patient-led titration was associated with a lower risk of macrosomia (6.9% vs 37.0%, relative risk 0.18, 95% CI, 0.04 to 0.84) and large-for-gestational-age (LGA) birth weight (3.3% vs 34.6%, relative risk 0.10, 95% CI, 0.08 to 0.12). Other pregnancy and patient-reported outcomes did not differ between the groups. CONCLUSION: Patient-led insulin titration for GDM resulted in a similar mean fasting glucose compared with clinician-led insulin titration but was associated with more rapid achievement of glycemic control and a lower risk of macrosomia and LGA birth weight. These data support the need for larger patient-centered GDM treatment trials. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov , NCT05922033.