The high volume of patients admitted during the SARS-CoV-2 pandemic has an independent harmful impact on in-hospital mortality from COVID-19

Abstract

During the Coronavirus disease 2019 (COVID-19) pandemic, advanced health systems have come under pressure by the unprecedented high volume of patients needing urgent care. The impact on mortality of this “patients’ burden” has not been determined. Through retrieval of administrative data from a large referral hospital of Northern Italy, we determined Aalen-Johansen cumulative incidence curves to describe the in-hospital mortality, stratified by fixed covariates. Age- and sex-adjusted Cox models were used to quantify the effect on mortality of variables deemed to reflect the stress on the hospital system, namely the time-dependent number of daily admissions and of total hospitalized patients, and the calendar period. Of the 1225 subjects hospitalized for COVID-19 between February 20 and May 13, 283 died (30-day mortality rate 24%) after a median follow-up of 14 days (interquartile range 5–19). Hospitalizations increased progressively until a peak of 465 subjects on March 26, then declined. The risk of death, adjusted for age and sex, increased for a higher number of daily admissions (adjusted hazard ratio [AHR] per an incremental daily admission of 10 patients: 1.13, 95% Confidence Intervals [CI] 1.05–1.22, p = 0.0014), and for a higher total number of hospitalized patients (AHR per an increase of 50 patients in the total number of hospitalized subjects: 1.11, 95%CI 1.04–1.17, p = 0.0004), while was lower for the calendar period after the peak (AHR 0.56, 95%CI 0.43–0.72, p<0.0001). A validation was conducted on a dataset from another hospital where 500 subjects were hospitalized for COVID-19 in the same period. Figures were consistent in terms of impact of daily admissions, daily census, and calendar period on in-hospital mortality. The pressure of a high volume of severely ill patients suffering from COVID-19 has a measurable independent impact on in-hospital mortality.