INTRODUCTION: Previous studies have reported widespread unexplained geographical variations in emergency paediatric admissions for epilepsy. OBJECTIVES: Our aim was to investigate the extent to which geographical variation in epilepsy admissions among children and young people (CYP) aged 0-18 years, in England, reflects variation in new epilepsy diagnoses. DESIGN AND SETTING: A retrospective secondary analysis of Hospital Episode Statistics data for emergency admissions between April 2018 and March 2019, and Epilepsy12 audit data for new epilepsy diagnoses in England, between July and November 2018. OUTCOME MEASURES: The ratios of observed to expected epilepsy admissions and new diagnoses were calculated for each hospital Trust, based on their catchment population, and adjusted for age, sex and deprivation. Standardised ratios of observed to expected epilepsy admissions were plotted against standardised ratios of observed to expected new diagnoses of epilepsy at Trust level and the Pearson correlation coefficient was calculated. RESULTS: There were 9246 emergency admissions for CYP to 134 Trusts with a primary diagnosis of epilepsy in England during the study period. 60 Trusts (44.4%) had either significantly lower or higher than expected standardised admission ratios for a primary diagnosis of epilepsy. There were 960 new diagnoses of epilepsy between July and November 2018 for 74 Trusts. 14 Trusts (18.9%) had either lower or higher standardised diagnosis ratios for a new diagnosis of epilepsy. There was no correlation between standardised emergency epilepsy admissions ratios and standardised new epilepsy diagnoses ratios at Trust level (Pearson r -0.06, p 0.63). CONCLUSIONS: Widespread unexplained variation in epilepsy admissions cannot be explained by variation in new epilepsy diagnosis. This raises concerns about the equity and accessibility of epilepsy services. Unit-level benchmarking data on new diagnoses and admissions for epilepsy can inform strategies to improve access to care and health outcomes.