ObjectiveTo investigate the effect of different doses of prophylactic dextrose gel on neurocognitive function and health at 6–7 years.DesignEarly school-age follow-up of the pre-hPOD (hypoglycaemia Prevention with Oral Dextrose) study.SettingSchools and communities.PatientsChildren born at ≥35 weeks with ≥1 risk factor for neonatal hypoglycaemia: maternal diabetes, small or large for gestational age, or late preterm.InterventionsFour interventions commencing at 1 hour of age: dextrose gel (40%) 200 mg/kg; 400 mg/kg; 200 mg/kg and 200 mg/kg repeated before three feeds (800 mg/kg); 400 mg/kg and 200 mg/kg before three feeds (1000 mg/kg); compared with equivolume placebo (combined for analysis).Main outcomes measuresToolbox cognitive and motor batteries, as well as tests of motion perception, numeracy and cardiometabolic health, were used. The primary outcome was neurocognitive impairment, defined as a standard score of more than 1 SD below the age-corrected mean on one or more Toolbox tests.FindingsOf 392 eligible children, 309 were assessed for the primary outcome. There were no significant differences in the rate of neurocognitive impairment between those randomised to placebo (56%) and dextrose gel (200 mg/kg 46%: adjusted risk difference (aRD)=−14%, 95% CI −35%, 7%; 400 mg/kg 48%: aRD=−7%, 95% CI −27%, 12%; 800 mg/kg 45%: aRD=−14%, 95% CI −36%, 9%; 1000 mg/kg 50%: aRD=−8%, 95% CI −29%, 13%). Children exposed to any dose of dextrose gel (combined), compared with placebo, had a lower risk of motor impairment (3% vs 14%, aRD=-11%, 95% CI −19%, −3%) and higher mean (SD) cognitive scores (106.0 (15.3) vs 101.1 (15.7), adjusted mean difference=5.4, 95% CI 1.8, 8.9).ConclusionsProphylactic neonatal dextrose gel did not alter neurocognitive impairment at early school age but may have motor and cognitive benefits. Further school-age follow-up studies are needed.