UCL Great Ormond Street Institute of Child Health


Great Ormond Street Institute of Child Health


Evaluation of educational and healthcare outcomes and their interaction for children with chronic li

Supervisors: Dr Katie Harron, Dr Ania Zylbersztejn, Dr Marianne Samyn

Evaluation of educational and healthcare outcomes and their interaction for children with chronic liver disease in the UK

Each year in the UK 400 children are diagnosed with chronic liver disease (CLD), with a fifth requiring liver transplantation at some stage. Some research evidence suggests that these children may have lower cognitive ability than their healthy peers. This is important, both in terms of developing the self-management skills required for independent management of liver disease in adulthood, and for later education and employment prospects. We do not know how specific liver diseases and their severity affect educational attainment, or how time spent in hospital because of CLD impacts on school outcomes. We expect that earlier detection of behavioural and developmental concerns, followed by timely referral to experienced specialist services, will improve health related quality of life, educational attainment and employment outcomes for patients. There is a lack of data on educational outcomes for children with CLD. 

This study will help establish a national data resource for children with CLD using linked health and education records, which can be used to evaluate healthcare and educational outcomes (including need for learning support), and their interaction. It will generate evidence to inform health services of the educational needs of children with liver disease, and to inform future development of guidance on long-term developmental follow-up. 

Research questions:
RQ1 How do educational outcomes, including special educational needs (SEN) support, of children with liver disease compare to those of the general population?
RQ2 What is the prevalence of school absences in children with liver disease, and to what extent does this explain the association between CLD and educational outcomes?
RQ3 How do educational outcomes of children following liver transplantation compare to those who have not undergone liver transplantation?
RQ4 What is the effect of age at liver transplantation on educational outcomes?

We will use existing linkage between health and education records (the ECHILD database) for all children born in England between 1995 and 2020. We will create phenotypes for CLD based on diagnosis and procedure codes in HES and use NHS Blood transfusion and Transplantation (NHSBT) data to identify a cohort of children up to age 18 with CLD and/or transplantation. Outcomes will include school attainment, SEN and absences. 

Analysis methods: We will describe the proportion of children with each outcome according to specific diagnoses (e.g. biliary atresia). We will use multi-level generalised linear regression models to estimate the relative risk of each outcome for each group, compared to the population of children without CLD, adjusting for relevant confounders (e.g. ethnicity, deprivation, birth characteristics) and clustering of outcomes within schools (RQ1). In order to assess the extent to which school absences explain the association between CLD and attainment, we will also include annual absences in the model (RQ2). To determine the impact of transplantation on outcomes (RQ3), we will use propensity scores to create a matched comparison group for children with CLD (and a subgroup biliary atresia patients) with and without liver transplantation, based on demographic and clinical indicators prior to transplantation. This will allow us to evaluate the difference between two groups of children with similar characteristics, some of whom will have had transplantation. We will extend these models to evaluate whether age at transplantation is associated with outcomes (RQ4) accounting for competing risks (e.g. death).

Months 1-6: Systematic review of cognitive outcomes for children with CLD
Months 7-12: Creation of analysis cohort; data-cleaning; identification of cases; PPI. 
Months 13-18: Analysis and write-up for RQ1-RQ2. 
Months 19-24: Analysis and write-up for RQ3.  
Months 25-36: Analysis and write-up for RQ4; dissemination; PPI.