In this month’s review of published papers and pre-prints on COVID-19, the Early Career Committee considered dozens of articles made open access this month. They were ranked against core pillars of the HDR UK ethos: research quality, team science, scale, open science, patient and public involvement, and equality, diversity and inclusion. This month’s winning publication was “Real-time spatial health surveillance: mapping the UK COVID-19 epidemic”, co-authored by HDR UK members Richard Fry, Joe Hollinghurst, Helen R Stagg, Daniel A Thompson, Claudio Fronterre, Chris Orton, Ronan A Lyons, David V Ford, Aziz Sheikh, and Peter J Diggle.
To help understand the localised spread of COVID-19, this study adopts sophisticated geo-spatial modelling methods to estimate near-real-time prevalence of infections at small-area resolution using data from the COVID Symptom Study app. While there are limitations with self-reported health outcome data, the maps generated provide the first fine-scale, UK-wide assessment of the geographical distribution of probable COVID-19 infections.
The project has demonstrated the value of a real-time spatio-temporal inferential mapping platform for public health efforts during the emergence and spread of infectious diseases. This study is an exemplar of how a combination of skills (Health Informatics, Statistics and Geography) is needed to provide insights to inform local and national government policy at a UK level throughout the COVID-19 crisis.
Building on the HDR UK ‘One Institute’ principles, the paper has prototyped and delivered data infrastructures and analysis pipelines capable of delivering timely and insightful analytics to all levels of government. This research has been used by the devolved administrations for pandemic planning, for example in identifying local hotspots that previous regional-level mapping may have masked.
Data used in this study is self-reported app data from the COVID Symptom Study app, which holds data on over 4 million individuals, held securely in and extracted from the Secure Anonymised Information Linkage (SAIL) Databank. The SAIL Databank acts as a secure gateway to protect the sensitive privacy. Data extracted from the SAIL Databank is then processed to generate suitable inputs for the geospatial modelling.
Full metadata summary can be found from the HDR Gateway deposit [Health Data Research UK. ZOE Metadata. url: https : / / metadata – catalogue .org/hdruk/#/catalogue/dataModel/06f8c66d-4e91-44dc-a109-1df729b72b61/properties].
The statistical model is a geospatial extension of logistic regression modelling. The system combines real-time data sources and rapid analytical tools and is believed to be the first work that can give predictions at Lower-layer Super Output Area resolution in near-real time. The study was conducted by a strong research team consisting of mainly HDR UK members from multiple medical schools across the UK, representing a collaboration led by BREATHE – The Health Data Research Hub for Respiratory Health. The data is at large-scale and can be accessed by application through the SAIL databank. The self-reported data involves diverse users, including representation from BAME groups.
HDR UK’s Early Career Committee would like to congratulate and commend this team for their contribution to HDR UK’s vision of uniting the UK’s health data to enable discoveries that improve people’s lives.
Linking health data to understand deaths of homeless patients
27 June 2019
A five-year research project led by Dr Robert Aldridge at the UCL Institute of Health Informatics involved the analysis of medical records for nearly 4,000 homeless people, 600 of whom...
Predicting suicide risk using health data at scale
27 June 2019
More than three-quarters of a million people globally take their own lives every year, to devastating personal, social and economic cost. Whilst past research has shown links between suicide and...