Ensuring trust in COVID-19 data: A retrospective cohort study

Antwi-Amoabeng, Daniel; Beutler, Bryce D.; Chahal, Gurpreet; Mahboob, Sumaiya; Gullapalli, Nageshwara; Tedja, Rudy; Madhani-Lovely, Farah; Rowan, Chris

There are no standardized methods for collecting and reporting coronavirus disease-2019 (COVID-19) data. We aimed to compare the proportion of patients admitted for COVID-19-related symptoms and those admitted for other reasons who incidentally tested positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Retrospective cohort study Data were sampled twice weekly between March 26 and June 6, 2020 from a “COVID-19 dashboard,” a system-wide administrative database that includes the number of hospitalized patients with a positive SARS-CoV-2 polymerase chain reaction test. Patient charts were subsequently reviewed and the principal reason for hospitalization abstracted. Data collected during a statewide lockdown revealed that 92 hospitalized patients had positive SARS-CoV-2 test results. Among these individuals, 4.3% were hospitalized for reasons other than COVID-19-related symptoms but were incidentally found to be SARS-CoV-2-positive. After the lockdown was suspended, the total inpatient census of SARS-CoV-2-positive patients increased to 128, 20.3% of whom were hospitalized for non-COVID-19-related complaints. In the absence of a statewide lockdown, there was a significant increase in the proportion of patients admitted for non-COVID-19-related complaints who were incidentally found to be SARS-CoV-2-positive. In order to ensure data integrity, coding should distinguish between patients with COVID-19-related symptoms and asymptomatic patients carrying the SARS-CoV-2 virus.