tetano
Editor, Senior Moderator
Prev Med Rep
. 2021 Dec;24:101624.
doi: 10.1016/j.pmedr.2021.101624. Epub 2021 Oct 25.
Predicting coronavirus disease (COVID-19) outcomes in the United States early in the epidemic
Eduardo J Simoes[SUP] 1 2 [/SUP], Chester L Schmaltz[SUP] 1 3 [/SUP], Jeannette Jackson-Thompson[SUP] 1 2 3 [/SUP]
Affiliations
Abstract
By 21 October 2020, the coronavirus disease (COVID-19) epidemic in the United States (US) had infected 8.3 million people, resulting in 61,364 laboratory-confirmed hospitalizations and 222,157 deaths. Currently, policymakers are trying to better understand this epidemic, especially the human-to-human transmissibility of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in relation to social, populational, air travel related and environmental exposure factors. Our study used 50 US states' public health surveillance datasets (January 1-April 1, 2020) to measure associations of confirmed COVID-19 cases, hospitalizations and deaths with these variables. Using the resulting associations and multivariate regression (Negative Binomial and Poisson), predicted cases, hospitalizations and deaths were generated for each US state early in the epidemic. Factors associated with a significantly increased risk of COVID-19 disease, hospitalization and death included: population density, enplanement, Black race and increased sun exposure; in addition, COVID-19 disease and hospitalization were also associated with morning humidity. Although predictions of the number of cases, hospitalizations and deaths due to COVID-19 were not accurate for every state, those states with a combination of large number of enplanements, high population density, high proportion of Black residents, high humidity or low sun exposure may expect more rapid than expected growth in the number of COVID-19 events early in the epidemic.
Keywords: Associated factors; COVID-19; NB model; Poisson model; Prediction.
. 2021 Dec;24:101624.
doi: 10.1016/j.pmedr.2021.101624. Epub 2021 Oct 25.
Predicting coronavirus disease (COVID-19) outcomes in the United States early in the epidemic
Eduardo J Simoes[SUP] 1 2 [/SUP], Chester L Schmaltz[SUP] 1 3 [/SUP], Jeannette Jackson-Thompson[SUP] 1 2 3 [/SUP]
Affiliations
- PMID: 34722135
- PMCID: PMC8545716
- DOI: 10.1016/j.pmedr.2021.101624
Abstract
By 21 October 2020, the coronavirus disease (COVID-19) epidemic in the United States (US) had infected 8.3 million people, resulting in 61,364 laboratory-confirmed hospitalizations and 222,157 deaths. Currently, policymakers are trying to better understand this epidemic, especially the human-to-human transmissibility of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in relation to social, populational, air travel related and environmental exposure factors. Our study used 50 US states' public health surveillance datasets (January 1-April 1, 2020) to measure associations of confirmed COVID-19 cases, hospitalizations and deaths with these variables. Using the resulting associations and multivariate regression (Negative Binomial and Poisson), predicted cases, hospitalizations and deaths were generated for each US state early in the epidemic. Factors associated with a significantly increased risk of COVID-19 disease, hospitalization and death included: population density, enplanement, Black race and increased sun exposure; in addition, COVID-19 disease and hospitalization were also associated with morning humidity. Although predictions of the number of cases, hospitalizations and deaths due to COVID-19 were not accurate for every state, those states with a combination of large number of enplanements, high population density, high proportion of Black residents, high humidity or low sun exposure may expect more rapid than expected growth in the number of COVID-19 events early in the epidemic.
Keywords: Associated factors; COVID-19; NB model; Poisson model; Prediction.