Phenotyping Neurodisability in Hospital Records in England: A National Birth Cohort Using Linked Administrative Data

Area of Science:

• Pediatric Health
• Neurodevelopmental Disorders
• Health Informatics

Background:

• Children with neurodisability present complex healthcare and educational needs.
• Linked administrative health and education data can enhance inter-service collaboration.
• Identifying neurodisability in hospital records is crucial for understanding these needs.

Purpose of the Study:

• Develop a diagnostic code list to identify neurodisability in hospital admission records.
• Characterize children with hospital-recorded neurodisability and their outcomes.
• Assess the representativeness of the identified neurodisability phenotype.

Main Methods:

• Utilized the Education and Child Health Insights Linked Data (ECHILD) database for a national cohort of English children born 2003-2009.
• Developed an algorithm with expert clinicians to phenotype neurodisability from hospital diagnostic records.
• Analyzed mortality, hospital admissions, and special educational needs (SEN) provision up to age 11.

Main Results:

• 3.6% of children had hospital-recorded neurodisability by age 11; common diagnoses included developmental disorders, autism, epilepsy, and cerebral palsy.
• Children with neurodisability had higher mortality and admission rates, accounting for significant proportions of all hospital admissions.
• 75% of children with hospital-recorded neurodisability received SEN provision, with 39% having more intensive support.

Conclusions:

• A phenotype for hospital-recorded neurodisability was derived, affecting 1 in 28 primary school children in England.
• This group exhibits elevated hospital admission rates and substantial special educational needs (SEN) provision requirements.
• The phenotype can guide research on inequalities and inform service and resource allocation for children with neurodisability.