Infection with SARS-CoV-2 is becoming the leading cause of death in most countries during the 2020 pandemic. The objective of this study is to assess the association between COVID-19 and cause-specific death. The design is retrospective cohort study. We included data from inpatients diagnosed with COVID-19 between March 18 and April 21, 2020, who died during their hospital stay. Demographic, clinical and management data were collected. Causes of death were ascertained by review of medical records. The sample included 128 individuals. The median age was 84 (IQR 75-89), 57% were men. In 109 patients, the death was caused by SARS-CoV-2 infection, whereas in 19 (14.8%, 95 CI 10-22%), the infection acted only as a precipitating factor to decompensate other pathologies. This second group of patients was older (88y vs 82, p < 0.001). In age-adjusted analysis, they had a greater likelihood of heart failure (OR 3.61 95% CI 1.15-11.32), dependency in activities of daily living (OR 12.07 95% CI 1.40-103.86), frailty (OR 8.73 95% CI 1.37-55.46). The presence of X-ray infiltrates Electronic supplementary material The online version of this article (

No abstract

No abstract

Abstract Background: Infection with SARS-CoV-2 is becoming the leading cause of death in most countries during the 2020 pandemic.Objective: to assess the association between COVID-19 and cause-specific death.Design: Retrospective cohort study.Setting and Participants: We included data from inpatients diagnosed with COVID-19 between March 18 and April 21, 2020, who died during their hospital stay. Demographic, clinical and management data were collected. Causes of death were ascertained by review of medical records.Results: The sample included 100 individuals. The median age was 85 (IQR 15), 59% were men. Fifteen patients (15%) died from complications unrelated to COVID-19. In univariate analysis, these patients compared to those who died from COVID-19 complications had a greater likelihood of dementia (47% vs 20%) and dependency in activities of daily living (93% vs 48%), higher Charlson comorbidity index scores (6 vs 5), a lower likelihood of fever at diagnosis (47% vs 77%), and higher lymphocyte counts (1,000 vs 660). In multivariate analysis, patients were older (OR 1.10, 95% CI 1.00–1.23) and more often had heart failure (OR 5.58, 95% CI 1.09–28.66). The presence of X-ray infiltrates was uncommon (OR 0.03, 95% CI 0.01–0.17). A higher percentage of patient deaths from causes unrelated to COVID-19 complications occurred during the two first weeks of the pandemic.Conclusions: Fifteen percent of patients with COVID-19 died from causes unrelated to COVID-19 complications. Most of these patients had more comorbidities and were frail and elderly. These findings can partially explain the excess mortality in older people.