Swansea University researchers, using the SAIL Databank, led a major European study published in PLOS One, investigating how socio-economic circumstances influence survival among children born with major congenital anomalies.
Based at Swansea University, SAIL Databank is powered by the Secure eResearch Platform (SeRP) and is funded by Welsh Government through Health and Care Research Wales and the Economic and Social Research Council (ESRC).
The international research, led by Swansea University as part of the EUROlinkCAT collaboration, analysed data from more than 47,000 children with major congenital anomalies across ten congenital anomaly registers in seven European countries. By securely linking health and administrative records within each participating country, researchers were able to examine how childhood mortality varies according to socio-economic status while maintaining strict data privacy and governance standards.
The study found that children with major congenital anomalies from the most socio-economically disadvantaged backgrounds were significantly more likely to die than those from the least deprived backgrounds. While disparities were evident during infancy, they became even more pronounced between the ages of one and ten, highlighting the continuing influence of social and economic factors on long-term health outcomes.
Lead author, Professor Sue Jordan at Swansea University, said, “Differences in mortality between rich and poor were greater after the first year of life, when care depends less on acute services, and more on primary and community care. Differences were consistent in all comparisons only in Wales and Ukraine – the countries with the lowest GDP per capita. We hope these findings will stimulate examination of resources allocated to children with congenital anomalies living in poverty.”
Secure data linkage across Europe
The research brings together experts from Swansea University and Public Health Wales’ Congenital Anomaly Register and Information Service (CARIS) with European institutions and health organisations in England, Northern Ireland, Italy, Finland, Ukraine, Malta, Spain and Denmark.
Rather than pooling sensitive patient-level information into a single database, each participating country securely analysed its own linked datasets using a shared Common Data Model and identical statistical analysis scripts. Only anonymous, aggregated results were transferred to a central repository, where they were combined using meta-analysis.
This collaborative approach enabled consistent analyses across multiple healthcare systems while ensuring that personal health data remained securely governed within each country’s existing legal and ethical frameworks.
SAIL Databank’s contribution
Within Wales, the research was enabled through the SAIL Databank, which provides researchers with secure access to anonymised, linkable population-scale data within a Secure Data Environment.
Using data linked through SAIL, researchers were able to combine information from the Congenital Anomaly Register and Information Service (CARIS), mortality records and measures of socio-economic deprivation to investigate childhood survival over time. The study achieved one of the highest linkage rates among participating countries, with almost all eligible Welsh records successfully linked for analysis.
The Welsh data formed part of a much larger European evidence base, allowing researchers to compare outcomes across different populations while preserving local governance arrangements for sensitive health data.
SAIL Databank’s Senior Data Scientist, Dr Hywel Turner Evans, who directed the analysis on Welsh data sources, said, “Using data from seven countries in Europe, we found that children with major congenital anomalies faced a higher risk of death if they were from more disadvantaged backgrounds. However, this was not seen in Denmark, suggesting that these inequalities are not unavoidable. Countries that reduce the impact of poverty on child health may offer valuable lessons for policymakers to improve lives elsewhere.”
Understanding inequalities in childhood survival
Overall, the study analysed data from 47,134 live-born children with major congenital anomalies for whom socio-economic information was available.
Across the participating countries, children in the most deprived groups had a 47% higher risk of dying during their first year of life than children in the most affluent groups. Between the ages of one and ten, the differences in chances of dying increased further, with mortality approximately twice as high among the most disadvantaged children.
Although the strength of these associations varied between countries, Wales was among the participating regions where socio-economic disadvantage was consistently associated with higher mortality during both infancy and later childhood.
The researchers also found that children born to non-EU nationals experienced higher mortality in the countries where nationality data were available, while no clear association was identified between maternal marital status and childhood mortality.
Supporting international population data research
The authors conclude that secure linkage of routinely collected health and administrative data provides valuable opportunities to better understand the long-term outcomes of children living with congenital anomalies.
By enabling secure access to anonymised, linked population data in Wales, SAIL Databank supported an international collaboration that combined evidence from seven European countries without requiring sensitive data to leave their country of origin. The study demonstrates how Secure Data Environments and common analytical approaches can help answer important public health questions that cannot be addressed by individual datasets alone.
As international collaborations increasingly rely on privacy-preserving approaches to data analysis, studies such as EUROlinkCAT demonstrate the value of trusted research infrastructure in generating robust evidence to inform healthcare policy and improve understanding of health inequalities across Europe.
This research is published in PLOS One and is available at – https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0352025
