Pseudo-Likelihood Based Logistic Regression for Estimating COVID-19 Infection and Case Fatality Rates by Gender, Race, and Age in California

• DOI: 10.1101/2020.06.29.20141978
• Publication Date: 2020-07-01T16:52:25Z
• Authors: Di Xiong; Lu Zhang; Gregory L. Watson; Phillip Sundin; Teresa Bufford; Joseph A. Zoller; John Shamshoian; Marc A. Suchard; Christina M. Ramirez
• Keywords: Case fatality rate; Pandemic

Abstract In emerging epidemics, early estimates of key epidemiological characteristics the disease are critical for guiding public policy. particular, identifying high risk population subgroups aids policymakers and health officials in combatting epidemic. This has been challenging during coronavirus 2019 (COVID-19) pandemic, because governmental agencies typically release aggregate COVID-19 data as marginal summary statistics patient demographics. These may identify disparities outcomes between broad subgroups, but do not provide comparisons more granular defined by combinations multiple We introduce a method that overcomes limitations aggregated yields infection case fatality rates — quantities policy related to control prevention across demographic characteristics. Our approach uses pseudo-likelihood based logistic regression combine with population-level survey estimate illustrate our on California test-based gender, age, race ethnicity. analysis indicates California, males have higher age race/ethnicity groups, gender gap widening increasing age. Although elderly infected at an elevated mortality, increase monotonically LatinX African Americans than other groups. The highest 5 American male, Multi-race Asian female, Indian or Alaska Native indicating especially vulnerable subpopulation.