‘A stitch in time saves nine’: A public health approach to improving children's outcomes
Categories: Research using linked data, Blogs, Datasets, ADR UK Research Fellows, ADR England, Office for National Statistics, Children & young people, Health & wellbeing
30 January 2025
In this blog, ADR UK Research Fellow Dr Amanda Mason-Jones explores her decision to use the Education and Child Health Insights from Linked Data (ECHILD) dataset and shares her ambitions for its application in public health research.
Linked data – especially when it connects de-identified education and health records – is important for policy-relevant research on the health and wellbeing of children. By bridging these domains, researchers can look at the impacts of health on educational outcomes, and vice versa.
ECHILD is one such linked dataset, made available for accredited researchers to access this year. I am using it for my project, Maternal mental health, child health & emergency department utilisation: Impact on children’s educational and development outcomes in England”.
My background in healthcare
Having worked as a nurse and health visitor in the NHS for over a decade, it was clear to me that maternal health and wellbeing had a direct impact on children’s health and development. This might seem obvious, but it felt ‘anecdotal’ at that time. We were using measures such as the Edinburgh Postnatal Depression Scale to assess maternal mental health, and perhaps providing support and referrals for women, but somehow seemed to be missing the intergenerational links to their children’s health, development, and subsequent educational outcomes.
When I later trained in public health and injury epidemiology, I became more acutely aware of these links. Working on childhood injury prevention trials required hours sitting in the emergency department, extracting data from records manually. It also became clear that data collected in a busy hospital environment was not necessarily as neat and tidy as I now, as a researcher, would hope for. Nevertheless, it could provide clear insights into individual journeys through our complex health systems.
Fast forward to an exploration of information collected pre- and post-pregnancy about maternal mental health and emergency department utilisation. Although it was focused on only four geographical locations in England, patterns began to emerge that linked mothers’ emergency department use to their infants’ use, and importantly, how poor maternal mental health was associated with increased visits for both mothers and their infants. We also know that emergency departments are struggling and that costs are increasing the cost of these increases year on year, and that there are also stark inequities in terms of receiving timely and appropriate care.
But what if more could be done?
This is where the ECHILD data resource becomes invaluable. We are now able to link mothers’ de-identified health data to that of their children. And not only this: we can also link in children’s educational data from the National Pupil Database.
Public health is about prevention, and we know (don’t we?) that the proverbial ‘stitch in time saves nine’. So in this project, I want to understand why mental health services might be used, what factors might lead to attending the emergency department and then how this might impact children’s health, development and educational outcomes.
The ECHILD dataset will provide the opportunity for me to really pinpoint where in our systems we can intervene to improve health, development, and educational outcomes – ultimately for the public good.