Background Despite increases in survival after extremely preterm (EP) birth, it is unclear whether the prevalence of neurodevelopmental morbidity has changed.

Aim To compare the prevalence of disability between national cohorts of EP children born in 1995 and 2006, respectively.

Methods Independent assessors evaluated children born at 25 weeks of gestation or less in England in 1995 (EPICure) and 2006 (EPICure-2). Bayley-III scores were adjusted to produce MDI-equivalent scores for comparison purposes.

Results Of the 260 eligible children in the EPICure cohort, 235 (90%) underwent formal neurodevelopmental assessment at 29–36 months corrected age. In the EPICure-2 cohort, 325/586 (55%) children were assessed at 27–48 months. Use of antenatal steroids, surfactant and effective hypothermia prevention were higher and postnatal steroid use lower in the 2006 cohort compared to 1995 births. Multiple imputation to correct for selective dropout revealed no differences in severe disability (18.9% vs 19.9% respectively) or cerebral palsy (20.1% vs 19.9%).

Conclusions Despite improved survival and reduced early morbidity in EP children between 1995 and 2006, we were unable to detect significant improvements in neurodevelopmental morbidity during early childhood.