The opportunities and challenges of using administrative data for international comparisons of child health
Dr. Ania Zylbersztejn (UCL Institute of Child Health, UK) is a population data scientist in child health research in England. During her research exchange at the Edwin S.H. Leong Centre for Healthy Children in May 2024, she examined how administrative data can be used for international comparisons of health outcomes for children with chronic conditions, and when such comparisons can be useful.
International comparisons of health outcomes can help policy makers identify opportunities for improvement relative to countries with better outcomes and different policies. My interest in international comparisons started with my PhD, which aimed to explain why child mortality is almost twice as high in England compared with Sweden. We found that these differences were largely due to more babies being born prematurely, with low birth weight or with birth defects in England compared to Sweden. Policies focusing on maternal health and wellbeing before and during pregnancy to improve birth outcomes are therefore likely to have most impact on reducing child mortality in England.
Ontario, Canada and England are generally comparable high-income jurisdictions, both with publicly funded universal health care systems and established re-use of administrative health records for research, an excellent opportunity for an international study. My current research focuses on children with rare or complex health conditions, and their families, therefore this research placement aimed to set the groundwork for future cross-country comparison of outcomes for children with complex chronic conditions.
For a fair comparison it is important to assess the comparability of the datasets used, to ensure that observed differences are not due to data artefacts (e.g. differences in what “counts” as a hospital admission in different countries), and to understand differences in healthcare provision or healthcare use (e.g. the thresholds for getting admitted to hospital might differ). I met with many researchers at the Edwin S.H. Leong Centre for Healthy Children and The Hospital for Sick Children to discuss these methodological challenges, different ways in which children with chronic conditions can be indicated in administrative records and developing a common data model (a tool for standardizing data from different settings to ensure common structure, format, and definitions). Although my placement has finished, we continue to work on a summary of best practices for future international studies of child health, so watch this space!
During my placement I also had the opportunity to share insights from work we do at UCL using ECHILD, a novel national database linking health and education records for all children in England. Our work includes the HOPE research program, which examines inequalities in and the impact of additional educational support in schools (referred to as special educational needs provision in England). We also study trends in health outcomes for vulnerable adolescents, who receive different levels of statutory support from education and social care, as they become young adults. Finally, we are planning to use ECHILD to obtain comparison cohorts for an independent process and impact evaluation of The Generation Study, a research study run by Genomics England and NHS England to explore the possibility of using genomic sequencing to expand current newborn screening programmes. You can view my presentation here.