Common data standards: Seeing the full picture
Data is at its most powerful when it is joined up. This is particularly important in paediatrics as often
children have rare diseases, so it is only by joining up cohorts across the world that it is possible to spot patterns in symptoms, diagnosis and treatment outcomes.
Joining up data was extremely valuable during the COVID-19 pandemic when clinicians and researchers had to rapidly characterise the novel virus to find out how to diagnose and treat people. Especially complications in children which were rare events. This was when 4CE stepped in.
4CE is an international consortium for studies of the COVID-19 pandemic using electronic health record data. This is done in a safe and secure way, where patient data remain within hospital boundaries and only anonymised aggregate data and results are shared centrally for further analysis.
By using the DRE, we were able to extract and analyse the required GOSH patient data locally. The aggregate data from 27 hospitals across six countries were then joined and analysed together. The study analysed over 27,000 laboratory values for 16 different tests which allowed identification of abnormalities that could help characterise severity and inform treatment of COVID-19 in children. The study identified complications associated with COVID-19 infection in children and young people so that clinicians could better plan for their care and inform families about possible outcomes.
Our work to join up paediatric data securely and effectively for research and innovation continues. Find out more in the stories below.
The Observational Medical Outcomes Partnership Common Data Model (OMOP)
We are mapping GOSH’s electronic health record data to The Observational Medical Outcomes Partnership Common Data Model (OMOP) as a data partner in the EHDEN network. This project is in partnership with Aridhia, a company that provides computer software to enable safe, compliant collaboration for digital research.
So far, we have characterised over 7 million electronic health record data items to map to the OMOP model. We are currently automating the translation of our data to these standard categories while ensuring accuracy through a quality assessment process. Structuring our rich data lake in this way will be revolutionary for paediatric research by enabling quicker, collaborative research so we can most effectively improve the lives of patients and families.
"The use of common data models for research means that researchers and clinicians don’t need to spend a lot of time and effort defining data specifications with collaborators and preparing data for each project. They can get started straight away using the data set which increases the speed at which they can get results to their questions.
At GOSH, we may only see a handful of patients with a rare condition so linking up with other centres of excellence in paediatric care can increase our pool of knowledge and the likelihood of making new findings in this underserved area." - Dr Natassa Spiridou, Head of the GOSH Digital Research Environment