ICU

Intensive Care Units (ICU)

## What is icumonitoring.ch? icumonitoring.ch site is a platform to follow in near-real time beds/ventilators occupancy in intensive care units (ICU) in Switzerland during the COVID19 epidemic. The platform offers projections of ICU occupancy 3- and 7-days ahead by Regions, Cantons, and hospitals (restricted access). The platform is updated Monday nights, and Thursday nights.

## How does it work? The platform combines forecasts from epidemic (S-E-I-R) models with data of bed occupancy. We estimate the number of cases, death, hospitalizations, ICU hospitalizations, and ventilators required in each Canton/Region. We also estimate in each hospital the number of hospitalization due to COVID19, the number of all-causes ICU hospitalizations, and ventilators required. The methodology used for the epidemic forecasts is described in the Swiss Medical Weekly (Zhao, Tepekule, Criscuolo et al.) and as well as in a Policy Brief from National COVID-19 Science Task Force. The two approaches are works in progress. Finally, the approach used to calculate the maximum inflow of COVID-19 patients is described here. ## Team icumonitoring.ch is maintained by the Health Geography and Policy Group (Prof. Thomas Van Boeckel) at ETH Zürich. The platform was developed by Mr. Nicola Criscuolo (ETH Zürich). Epidemic projections were made by Dr. Burcu Tepekule (BT model, University Hospital Zürich) as well as Cheng Zhao (CZ model, ETH Zürich), based on a framework proposed by Dr. Christian Althaus (University of Bern), and Dr. Riccardo Delli Compagni (RDC Model, ETH Zürich). The data-driven time series forecasts were made by Dr. Monica Golumbeanu and Prof. Melissa Penny (MG model, Swiss Tropical and Public Health Institute). The maximum patient inflow in ICU is estimated using a hospital-level model within-hospital developed by Dr. Peter Ashcroft (ETH Zurich). ## Data sources Data on bed occupancy are obtained from the IOS system and provided by the Coordinated Medical Service of Switzerland. Data on length of stay in ICU, and mortality in ICU are obtained from the RISC-19-ICU register, a collaborative effort with the participation of a majority of the Swiss ICUs maintained by Matthias Hilty, MD, and Pedro Wendel Garcia, at the University Hospital of Zürich. The initiative is supported by the Swiss Society of Intensive Care Medicine. Data on deaths and cases attributable to COVID-19 are used for model inference, and obtained daily from FOPH. ## Funding icumonitoring.ch is supported by the Coordinated Medical Service of Switzerland. ## Contact thomas.vanboeckel@env.ethz.ch