Evolution of COVID-19 mortality over time: results from the Swiss hospital surveillance system (CH-SUR)
Journal Paper/Review - Nov 24, 2021
Roelens Maroussia, Schreiber Peter W, Sommerstein Rami, Troillet Nicolas, Tschudin-Sutter Sarah, Vetter Pauline, Bernhard-Stirnemann Sara, Corti Natascia, Gaudenz Roman, Marschall Jonas, Nussbaumer-Ochsner Yvonne, Senn Laurence, Vuichard-Gysin Danielle, Zimmermann Petra, Zucol Franziska, Iten Anne, Riedel Thomas, Niederer-Loher Anita, Martin Alexis, Friker Brian, Sousa Filipe Maximiano, Thiabaud Amaury, Vidondo Beatriz, Buchter Valentin, Gardiol Céline, Vonlanthen Jasmin, Balmelli Carlo, Battegay Manuel, Christoph Berger, Buettcher Michael, Cusini Alexia, Flury Domenica, Heininger Ulrich, Keiser Olivia
When the periods of time during and after the first wave of the ongoing SARS-CoV-2/COVID-19 pandemic in Europe are compared, the associated COVID-19 mortality seems to have decreased substantially. Various factors could explain this trend, including changes in demographic characteristics of infected persons and the improvement of case management. To date, no study has been performed to investigate the evolution of COVID-19 in-hospital mortality in Switzerland, while also accounting for risk factors.
We investigated the trends in COVID-19-related mortality (in-hospital and in-intermediate/intensive-care) over time in Switzerland, from February 2020 to June 2021, comparing in particular the first and the second wave. We used data from the COVID-19 Hospital-based Surveillance (CH-SUR) database. We performed survival analyses adjusting for well-known risk factors of COVID-19 mortality (age, sex and comorbidities) and accounting for competing risk.
Our analysis included 16,984 patients recorded in CH-SUR, with 2201 reported deaths due to COVID-19 (13.0%). We found that overall in-hospital mortality was lower during the second wave of COVID-19 than in the first wave (hazard ratio [HR] 0.70, 95% confidence interval [CI] 0.63- 0.78; p <0.001), a decrease apparently not explained by changes in demographic characteristics of patients. In contrast, mortality in intermediate and intensive care significantly increased in the second wave compared with the first wave (HR 1.25, 95% CI 1.05-1.49; p = 0.029), with significant changes in the course of hospitalisation between the first and the second wave.
We found that, in Switzerland, COVID-19 mortality decreased among hospitalised persons, whereas it increased among patients admitted to intermediate or intensive care, when comparing the second wave to the first wave. We put our findings in perspective with changes over time in case management, treatment strategy, hospital burden and non-pharmaceutical interventions. Further analyses of the potential effect of virus variants and of vaccination on mortality would be crucial to have a complete overview of COVID-19 mortality trends throughout the different phases of the pandemic.