2020/08/01 by Daniel W. Sacks, Nir Menachemi, Sacks, Daniel W. +5
Medicine · #Applications (stat.AP) #COVID-19 Clinical Research Studies #Econometrics (econ.EM) #FOS: Computer and information sciences #FOS: Economics and business #SARS-CoV-2 and COVID-19 Research #SARS-CoV-2 detection and testing
paper · pdf · doi:10.48550/arxiv.2008.00298
openalex publication_date 2020/08/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Measuring the prevalence of active SARS-CoV-2 infections in the general\npopulation is difficult because tests are conducted on a small and non-random\nsegment of the population. However, people admitted to the hospital for\nnon-COVID reasons are tested at very high rates, even though they do not appear\nto be at elevated risk of infection. This sub-population may provide valuable\nevidence on prevalence in the general population. We estimate upper and lower\nbounds on the prevalence of the virus in the general population and the\npopulation of non-COVID hospital patients under weak assumptions on who gets\ntested, using Indiana data on hospital inpatient records linked to SARS-CoV-2\nvirological tests. The non-COVID hospital population is tested fifty times as\noften as the general population, yielding much tighter bounds on prevalence. We\nprovide and test conditions under which this non-COVID hospitalization bound is\nvalid for the general population. The combination of clinical testing data and\nhospital records may contain much more information about the state of the\nepidemic than has been previously appreciated. The bounds we calculate for\nIndiana could be constructed at relatively low cost in many other states.\n