tetano
Editor, Senior Moderator
Braz J Med Biol Res
. 2021 Feb 12;54(4):e10766.
doi: 10.1590/1414-431X202010766. eCollection 2021.
Death trends based on autopsy data compared to the beginning of the coronavirus pandemic in Brazil
G A A Gimenez Junior[SUP] 1 [/SUP], P K Zilli[SUP] 2 [/SUP], L F F Silva[SUP] 3 [/SUP], C A Pasqualucci[SUP] 3 [/SUP], A B Campo[SUP] 4 [/SUP], C K Suemoto[SUP] 5 [/SUP]
Affiliations
Abstract
The novel Coronavirus disease (COVID-19) is responsible for thousands of deaths worldwide, especially in Brazil, currently one of the leading countries in number of infections and deaths. The beginning of the COVID-19 epidemic in Brazil is uncertain due to the low number of tests done in the country. The excess number of deaths can suggest the beginning of the pandemic in this context. In this article, we used an autoregressive integrated moving average (ARIMA) model to investigate possible excesses in the number of deaths processed by the S?o Paulo Autopsy Service according to different causes of deaths: all-cause, cardiovascular, and pulmonary causes. We calculated the expected number of deaths using data from 2019 to 2020 (n=17,011), and investigated different seasonal patterns using harmonic dynamic regression with Fourier terms with residuals modeled by an ARIMA method. We did not find any abnormalities in the predicted number of deaths and the real values in the first months of 2020. We found an increase in the number of deaths only by March 20, 2020, right after the first COVID-19 confirmed case in the city of S?o Paulo, which occurred on March 16, 2020.
. 2021 Feb 12;54(4):e10766.
doi: 10.1590/1414-431X202010766. eCollection 2021.
Death trends based on autopsy data compared to the beginning of the coronavirus pandemic in Brazil
G A A Gimenez Junior[SUP] 1 [/SUP], P K Zilli[SUP] 2 [/SUP], L F F Silva[SUP] 3 [/SUP], C A Pasqualucci[SUP] 3 [/SUP], A B Campo[SUP] 4 [/SUP], C K Suemoto[SUP] 5 [/SUP]
Affiliations
- PMID: 33624732
- DOI: 10.1590/1414-431X202010766
Abstract
The novel Coronavirus disease (COVID-19) is responsible for thousands of deaths worldwide, especially in Brazil, currently one of the leading countries in number of infections and deaths. The beginning of the COVID-19 epidemic in Brazil is uncertain due to the low number of tests done in the country. The excess number of deaths can suggest the beginning of the pandemic in this context. In this article, we used an autoregressive integrated moving average (ARIMA) model to investigate possible excesses in the number of deaths processed by the S?o Paulo Autopsy Service according to different causes of deaths: all-cause, cardiovascular, and pulmonary causes. We calculated the expected number of deaths using data from 2019 to 2020 (n=17,011), and investigated different seasonal patterns using harmonic dynamic regression with Fourier terms with residuals modeled by an ARIMA method. We did not find any abnormalities in the predicted number of deaths and the real values in the first months of 2020. We found an increase in the number of deaths only by March 20, 2020, right after the first COVID-19 confirmed case in the city of S?o Paulo, which occurred on March 16, 2020.