COVID-19 Pandemic
Covid-19 Survival Analysis in elderly patients, Rio de Janeiro Brazil Renan MVR Almeida* Renan Moritz VR Almeida VR Almeida VR Almeida Universidade Federal do Rio de Janeiro
Objective: To evaluate the survival of elderly patients (≥60 y.o.) hospitalized for COVID-19 in the state of Rio de Janeiro, Brazil, February 2020 to December 2021.
Methods: This was a retrospective cohort study with COVID-19 hospitalization data provided by OpenDataSUS (a data repository maintained by the country’s Ministry of Health). Individuals aged 60 years or older, hospitalized in the state of Rio de Janeiro, Brazil, were included. The variables analyzed were sex, age group, race/color (white x non-white), municipality of residence (Rio de Janeiro city x others), ICU admission, peregrination (travel between cities for health care), delay in admission, and year of occurrence (2020 x 2021). Survival estimates were obtained using the Kaplan–Meier method, and factors associated with risk of death were evaluated with the Cox proportional hazards model, α=5%. Goodness of fit was assessed by Harrel’s C-index, and the R software (v.4.3) was used for all data analysis.
Results: Of the 1,473,182 hospital admissions during the period, 33,485 eligible cases were selected, and after missing or inconsistent data exclusion, 20,382 could be analyzed, with 11,939 deaths. Harrel’s C-index was 0.563, and the model was statistically significant (p<0.001). A higher risk of death was observed for males (HR=1.05; 95% CI: 1.02–1.09), non-whites (HR=1.10; 95% CI: 1.07–1.15), those admitted to ICU (HR=1.26; 95% CI: 1.21–1.31), those in the older age group (HR=1.32; 95% CI: 1.27–1.37) and those with delayed admission (HR=0.86; 95% CI: 0.83–0.89). No significant associations were observed for municipality of residence, peregrination or year of admission.
Conclusion: The analysis identified clinical and sociodemographic factors associated with the risk of death in elderly patients hospitalized for COVID-19, with emphasis on advanced age and ICU admission. The findings reinforce the importance of using large databases to support planning and response to future public health emergencies.