Department of Health and Social Care, Scientific Pandemic Influenza Group on Modelling, Scientific Advisory Group for Emergencies

Government Agency (Health and Adult Social Care)

1087

In support of this national response, this project is undertaking the following areas of work. Using techniques in machine learning and statistical analysis, we are going to identify at-risk groups of COVID-19 in terms of demographics and location. Patterns in the development of symptoms across the UK can be identified, which can then be communicated to decision makers to decide where interventions need to be focused.

Specifically, this project aims to explore the risk factors associated with areas with high developments of symptoms. We will identify environmental factors that may have affected evolution of the epidemic and the symptoms that emerged. For this part, we envisage using Spatial Statistics and Time Series Analysis. Following this, by using individual patient data, we will explore the etiology of the disease, identifying risk factors for each of the symptoms and for the final outcome (improve, intensive-care progression, discharge, death), paying special attention to interactions and mediating Not required IF YOU HAVE TICKED 'NOT REQUIRED' PLEASE SPECIFY THE REASONS: This project is being undertaken in the national interest during a national emergency, with information being provided across governments and health authorities. Research Ethics Not required THE PROJECT USES WILL USE ONLY ANONYMISED DATA, AND THEREFORE RESEARCH ETHICS REVIEW IS NOT REQUIRED Yes 2 of 5 relationships. For this part we envisage using generalized linear models (GLM’s) that will adjust for the longitudinal natural of the observations (mixed logistic regression, for example) and structural equation modelling (SEM). We will use machine learning techniques in our model building strategies. The anticipated outcome of the project is a model of the progression of the disease to identify patterns and environmentally-driven (geographical, economic, etc) clusters of boroughs. The visualizations will consist of heat maps and time series.

21/04/2020

Anonymous

(e) processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller;

(j) processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject.

Not applicable

Not applicable

TRE