Back

Evaluation of post-acute COVID-19 health outcomes (ECHOES) in England: The development of national surveillance system for long- term health outcomes following COVID-19

Allen, H.; Hassell, K.; Rawlinson, C.; Pullen, O.; Campbell, C.; Jodicke, A. M.; Catala, M.; Prats-Uribe, A.; Dabrera, G.; Prieto-Alhambra, D.; Campos-Matos, I.

2024-10-21 epidemiology
10.1101/2024.10.18.24315744 medRxiv
Show abstract

IntroductionElectronic health records can be used to understand the diverse presentation of post-acute and long-term health outcomes following COVID-19 infection. In England, the UK Health Security Agency in collaboration with the University of Oxford have created the ECHOES dataset to monitor how an initial SARS-CoV-2 infection episode is associated with changes in the risk of health outcomes that are recorded in routinely collected health data. MethodsThe ECHOES dataset is as a national level dataset combining national level surveillance, administrative, and healthcare data. Entity resolution and data linkages methods are used to create a cohort of individuals who have tested positive and negative for SARS-CoV-2 in England throughout the COVID-19 pandemic, alongside information on a range of health outcomes including diagnosed clinical conditions and mortality, and risk factor information. ResultsThe dataset contains comprehensive COVID-19 testing data and demographic, socio economic and health related information for 44 million individuals, who tested for SARS-CoV-2 between March 2020 and April 2022, representing 15,720,286 individuals who tested positive and 42,351,016 individuals who tested negative. DiscussionWith the application of epidemiological and statistical methods, this dataset allows a range of clinical outcomes to be investigated, including pre-specified health conditions and mortality. Furthermore, understanding of potential determinants of health outcomes can be gained, including pre-existing health conditions, acute disease characteristics, SARS-CoV-2 vaccination status and genomic variant.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.