Existing approved uses of platform data

Research proposals

Please review these proposals before submitting a data access request. In the interest of maximising the production of evidence to reduce the impact of COVID-19, the platform aims to avoid unnecessary duplication of analyses.

COVID-Intelligence, a Bayesian Network Clinical Decision Support System for COVID-19

Monash University, Australia

With the aim to improve the diagnosis and management of COVID-19, mathematical models have been developed by the research team that describe the COVID-19 disease process, from infection through to the outcome (recovery or death). The models have been built using the knowledge volunteered by a large number of medical experts from various fields, and will now be developed further using data from known and suspected cases of COVID-19. Patient data will be used to test, strengthen and confirm the researchers’ individual patient predictions of COVID-19 disease progression. Ultimately, it is anticipated that the model will be used to understand COVID-19 disease and guide individual patient management.

Read the application

Clinical presentation and outcome of hospitalised Dutch patients with COVID-19

Franciscus Gasthuis and Vlietland, Rotterdam, The Netherlands

The emergence of COVID-19 has had a tremendous impact on global health and international society. In a Dutch multicentre collaboration, the research team describe the clinical characteristics of patients who were admitted to Dutch hospitals with COVID-19 and assess the clinical outcomes of these patients. The objectives of the research are: to describe the clinical features of patients who were admitted to Dutch hospitals with COVID-19, including demographics, admission characteristics (symptoms, signs, and laboratory), treatment-related variables and complications; to assess patient outcomes and determine associations between demographic (including age, gender, ethnicity, comorbidity), admission characteristics (including inclusion O2 saturation, GCS, creatinine, CRP) and patient outcomes.

Read the application

Coagulopathy and thrombosis among COVID-19 patients

University of Utah School of Medicine, USA and Fundação Oswaldo Cruz, Brazil

Due to the novelty of COVID-19 and the conflicting implications of attempting to avoid both serious thrombosis and serious haemorrhage, guidelines on evaluation and management of COVID-19 patients with coagulopathy have substantial variations. This research proposes to characterise patients with bleeding, thrombosis, and coagulopathy with an approach that leverages the strengths of the large database available. Determination of geographic variation in COVID-19 associated coagulopathy observations may provide insight into whether different regions may have different coagulopathy phenotypes. Describing and characterizing the bleeding and coagulopathy seen among COVID-19 patients who are critically ill may help refine risk assessment for the sickest patients. Describing regional variation in practices of anticoagulation may aid in designing prospective clinical trials.

Read the application

Neurological manifestation and outcome for patients admitted to hospital with COVID-19

Johns Hopkins University School of Medicine, USA and University of Cape Town, South Africa

This is a multi-centre international observational study in patients with COVID-19 to report and characterize the prevalence, risk factor, and outcome of neurological manifestations. Recent findings suggested that COVID-19 patients may also develop neurological symptoms by mechanisms not yet elucidated. To date, they have been characterised into three main areas: central nervous system disorders, peripheral nervous system disorder and skeletal muscle symptoms. It is well known that neurological outcome in critically ill patients can be influenced by the development of secondary brain damage, and that COVID-19 patients frequently present hypoxia, as a result of severe respiratory distress, hypotension, and microvascular abnormalities. It can be hypothesized that one of the possible mechanisms involved in neurological manifestations of COVID-19 patients could be the promotion of neuroinflammation and excitotoxicity with increased permeability of the blood brain barrier. The aim of this observational multi-centre international study is to identify the prevalence of neurological manifestations in critically ill confirmed COVID-19 patients and to assess risk factors and outcomes of neurological complications of COVID-19.

Read the application

Acute Kidney Injury in COVID-19

University of Queensland, Australia and Universidad Mayor de San Simon, Cochabamba, Bolivia

Close to half of all patients admitted to hospital with COVID 19 have kidney involvement and a variable proportion, between 0.5 to 36%, will develop acute kidney injury (AKI) during their hospital stay. AKI is especially common among patients who require mechanical ventilation, with much of its burden observed in the intensive care unit setting and around the time of intubation. In this context, development of AKI has been characterised as both a marker of disease severity as well as a negative prognostic indicator. To date, studies looking at kidney disease in COVID 19 patients have been limited to single centre or regional cohorts mostly from China, Europe or USA, which fail to reflect the global experience of kidney disease, particularly in low and middle income countries where the burden of disease may be much higher and the therapeutic resources limited. This analysis has three aims: to characterise patients with COVID-19 who develop AKI; to identify the temporal profile of AKI in patients with COVID-19 throughout the hospital journey; to develop a predictive machine learning model for AKI in COVID-19, using data from predominantly upper middle-income countries, and externally validate this model in a LMIC setting (Latin America).

Read the application

Analysis of the ISARIC International COVID-19 Critically Ill Cohort

British Columbia Children’s Hospital, Canada and Universidad De La Sabana, Columbia

There are few international cohorts of the critically-ill patient population described in the literature and this collaborative project aims to address this gap. The study aims to describe the first and subsequent waves of the COVID-19 pandemic by: summarising the demographic characteristics and clinical features of ~16,000 critically ill patients of any age, admitted to hospital with COVID-19 across high-income and low- and middle-income settings, across temporal phases of the pandemic; describing clinical outcomes (e.g. ICU length of stay, hospital length of stay, ICU mortality, hospital mortality, days of mechanical ventilation) of critically-ill patients with severe COVID-19; characterising the variability in the clinical features and management strategies of these patients; exploring the risk and protective factors associated with mortality for these patients; determining variation over time in various phases of the ICU ‘journey’ including treatments received, length of stay in ICU and duration of mechanical ventilation.

Read the application

Development of a predictive analytics model for need of extracorporeal support in COVID-19

Washington University in St Louis, USA, INOVA Fairfax Hospital, USA and National Cardiovascular Center Harapan Kita, Indonesia

The most resource-intensive and comprehensive support for patients with COVID-19 is extracorporeal membrane oxygenation (ECMO), functioning as an artificial heart and/or lung, removing carbon dioxide, supplementing oxygen to a patient’s blood, and providing life-sustaining oxygen delivery. The decision to place a patient on ECMO is one that takes place after standard therapies such as steroids, mechanical ventilation, and prone positioning among others have failed. During a pandemic it is critical to be cognizant of the significant resources required to place and maintain patients on ECMO, and more than 10 months into the pandemic, there remains difficulty in predicting who will require ECMO. This research plans to construct a model to aid in the prediction of which COVID-19 patients are most likely to necessitate the use of ECMO utilising the ISARIC database. The specific aims of the project are: to predict the necessity for ECMO support utilising variables 24-48 hours prior to ECMO initiation; to develop, validate and test a machine learning model for predicting the need for ECMO at various time windows using a large international database; to test the developed model in a local, holdout dataset, using a simulated real-time approach, and evaluate the decision alert rate.

Read the application

Analysis of non-UK cohort in ISARIC COVID-19 data platform

University of Oxford, UK

An overall descriptive dataset has been identified in collaborator meetings as a priority outcome. There are few international cohorts described in the literature and so this collaborative project hopes to address this. The project aims are: to summarize the demographic characteristics and clinical features of 10,941 patients admitted to hospital with COVID-19 across high-income, middle-income, and low-income settings; to characterise the variability in the clinical features of these patients; to explore the risk factors associated with mortality and ICU admission for these patients. All contributors to the ISARIC database are invited to participate in this analysis through review and input on the statistical analysis plan and resulting publication. The outputs of this work will be disseminated as widely as possible to inform patient care and public health policy.

Read the application

International comparison of age-specific presenting symptoms for patients admitted to hospital with coronavirus disease 2019

University of Oxford, UK

Case definitions for COVID-19 generally require an epidemiological link demonstrating possible exposure to a case. However, as the global pandemic has spread, many countries have sustained community transmission meaning everyone is potentially exposed. For settings where microbiological testing for the causative pathogen (SARS-CoV-2) is not available, and for patients who have mild symptoms and do not seek medical attention, symptomology is the main feature for identifying possible cases. The objectives of this study are: to investigate patterns of symptoms at arrival to hospital for patients admitted to hospital with COVID-19, stratified by age, sex and country; to investigate the utility of commonly used clinical case definitions of COVID-19 for patients of different ages and in different countries.

Read the application

Time variation in the inpatient journey

University of Oxford, UK

This study addresses the following research questions: Does the ‘inpatient journey’ change over time in the course of the COVID-19 pandemic? Do patient’s presenting characteristics change in the course of the COVID-19 pandemic? Do patient’s presenting characteristics explain changes in the ‘inpatient journey’? These will be examined through exploring variation in: the overall ‘inpatient journey’, using both interactive visualization and by taking snapshots (e.g. for the first 100, 1,000, 10,000 patients, and for every 10,000-patient increment in the global dataset); time from symptom onset to hospital admission; time from hospital admission to ICU admission; length of ICU stay; time from hospital admission to oxygen supplementation; time on oxygen supplementation; length of hospital stay.

Read Univ. Oxford's application

Sepsis in COVID-19

Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, China

This is a multi-centre international observational study of patients with COVID-19 based on the COVID-19 Data Platform to report and describe the demographic characteristics, clinical features, prevalence, risk factors, and sepsis prognosis. The study aims to: summarise the demographic and clinical characteristics of patients with sepsis hospitalized for COVID-19; to investigate the incidence of sepsis in non-ICU and ICU patients; to identify the time variation at different ICU stay stages, including treatments received, time in the intensive care unit, and mechanical ventilation time; to describe the complications and clinical outcomes associated with sepsis in patients with COVID-19; to explore the risk factors and protective factors associated with mortality in these patients; to construct clinical prediction models for the risk of sepsis and death in COVID-19 patients and perform external validation.

Read the application