Goal and design

The consortium aims to better understand and predict which COVID-19 patient should receive which treatment and which type of care. This is done by setting up a central database, providing insights and, where possible, applying advanced data analysis techniques. With these efforts the consortium intends to support physicians in charge of the care of COVID-19 patients.

Two interconnected projects are currently underway:

  • COVID-19 ICU Decision Support
  • COVID-19 Clinical Course

Each project has its own goal, design and methods for data collection, which are explained below.

Because this project is dynamic, the consortium's objectives may be further expanded in due time. Additional research questions can be registered on this page.

Goals

The projects have the following goals:

  • COVID-19 ICU Decision Support
    • Support the intensive care units with frequently updated insights about the optimal treatment strategy of COVID-19 patients in the intensive care unit.
    • When possible, use advanced data analysis and statistical methods, such as machine learning, to analyze which combination of treatment strategies of COVID-19 patients in the ICU is associated with the best outcomes.
  • COVID-19 Clinical Course: predicting the clinical course and optimal treatment for patients with COVID-19.
Design

Both projects are structured as follows:

  • COVID-19 ICU Decision Support: data collection of COVID-19 patients who are being being treated or have been treated in ICUs Dutch hospitals.Demographic information,vital parameters, lab data, and other observational data are collected. Analysis of this data will provide participating hospitals with current insights into the treatment and outcome of COVID-19 patients in intensive care.
  • COVID-19 Clinical Course: participating hospitals collect data from all COVID-19 patients admitted to the nursing ward, other departments or intensive care unit. Information about the diseases’ progression, history, and presentation in the hospital, as well as the course of the hospital stay, is collected retrospectively. After discharge, data on the clinical condition, treatments and complications are also collected. This data is stored encoded in a Castor database and is based on the WHO's COVID CRF.
Data collection

Data collection can be  carried out independently for the two projects. However, in order to reach the  overall goal of the consortium, it is desirable that participating hospitals enroll in both forms of data collection. Where possible, the data from both projects is merged. Please refer to the Participate page to learn more about the data collection and the support offered by the consortium.

Methods

Similar methods are being used in both projects:

  • COVID-19 ICU Decision Support:
    • Frequently updated reports about insights regarding ICU treatment of COVID-19 patients.
    • Analyze which combination of treatment strategies of an individual patient is associated with the best result using machine learning. In addition, this patient group is characterized with descriptive statistics.
  • COVID-19 Clinical Course: predict as accurately as possible how the disease will progress and predict the optimal treatment for each patient using machine learning. In addition, this patient group is characterized with descriptive statistics.
Protocols

The protocols and associated data protection impact analyses assessed by the Medical Ethical Committees are available to participating hospitals upon request.