Predicting infection with COVID-19 disease using logistic regression model in Karak City, Jordan

Abstract

Background: On March 2020, World Health Organization (WHO) labeled coronavirus disease 2019 (COVID-19) as a pandemic. COVID-19 has rapidly increased in Jordan which resulted in the announcement of the emergency state on March 19th, 2020. Despite the variety of research being reported, there is no agreement on the variables that predict COVID-19 infection. This study aimed to test the predictors that probably contributed to the infection with COVID-19 using a binary logistic regression model. Methods: Based on data collected by Google sheet of COVID-19 infected and non-infected persons in Karak city, analysis was applied to predict COVID-19 infection probability using a binary logistic regression model. Results: A total of 386 participants have completed the questionnaire including 323 women and 63 men. Among the participants 295 (76.4%) were aged less than or equal 45 years old, and 91 (23.6%) were aged over 45 years old. Among the 386 participants a total of 275 were infected with COVID-19. The LR chi-square test was used to analyze every demographic characteristic (sex, age, job, smoking, chronic disease, yearly flu injection) in this study to find predictors of the likelihood of COVID-19 infection. The findings indicate that the participants' sex and age are the most important demographic determinants of infection. Cox & Snell R Square (R2 = 0.028) and Nagelkerke R Square (R2 = 0.039) indicators was used to measure model fineness with significant P-value < 0.05. Conclusions: Given a person's age and sex, the final model presented in this study can be used to calculate the probability of infection with COVID-19 in Karak city. This could help aid health-care management and policymakers in properly planning and allocating health-care resources.