AUTHOR=Trecarichi Enrico Maria , Olivadese Vincenzo , Davoli Chiara , Rotundo Salvatore , Serapide Francesca , Lionello Rosaria , Tassone Bruno , La Gamba Valentina , Fusco Paolo , Russo Alessandro , Borelli Massimo , Torti Carlo ,  the IDTM UMG COVID-19 Group 

TITLE=Evolution of in-hospital patient characteristics and predictors of death in the COVID-19 pandemic across four waves: are they moving targets with implications for patient care?

JOURNAL=Frontiers in Public Health

VOLUME=Volume 11 - 2023

YEAR=2024

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2023.1280835

DOI=10.3389/fpubh.2023.1280835

ISSN=2296-2565

ABSTRACT=Manuscript contribution to the field: To the best of our knowledge, this is the first study that comparesthrough both descriptive and regression tree analyses-characteristics, clinical outcomes, and predictors of all-cause, in-hospital deaths in COVID-19 patients across four subsequent pandemic waves in Southern Italy. We found that patients with severe COVID-19 across all the pandemic waves, as well as those with both moderate COVID-19 and chronic kidney disease admitted during the most recent waves displayed the highest risk of death. We therefore conclude that: (i) prognostic factors for in-hospital mortality vary across the pandemic waves; (ii) chronic kidney disease is a discriminant comorbidity for estimating the risk of mortality in the current patient population; (iii) prognostic scores that were elaborated in the previous waves should be re-evaluated in the current epidemiological scenario. Overall, our results suggests that knowledge of the changing characteristics of COVID-19 patients might be helpful to better understand how the provision of the disease will be in the future and what clinical/health needs should be approached for the management of this disease. Lastly, despite the pandemic was declared over the last May by WHO, the recent observation of a significant increase of in-hospital deaths and the possibility of a recrudescence of the epidemic make our results more relevant.