Pulse Oximetry Improves COVID-19 Screening Sensitivity by Detecting Asymptomatic and Afebrile Infections in a Case Control Study.

Abstract

SARS-CoV-2 has spread quickly despite the implementation of various strategies to limit the spread of the disease. Current screening procedures are unable to detect asymptomatic infections. This study reports the prevalence of abnormal vital signs among incarcerated individuals with asymptomatic infections and proposes an enhanced screening algorithm that may detect asymptomatic infections. A case control study used a retrospective electronic chart review of COVID-19 infected people and matched controls housed in one Federal Bureau of Prisons institution. Data were collected on age, body mass index, medical history, temperature, blood oxygen saturation (SpO2), and symptoms. Fifty-seven laboratory-confirmed COVID-19 infections and 81 confirmed controls were identified. Asymptomatic infections were present in 35/57 people (61%) with a group minimum SpO2 94.4% (standard deviation [SD] 1.4) and a maximum oral temperature of 99.19°F (SD 0.36). An enhanced screening algorithm that uses pulse oximetry yields a sensitivity of 84.2%, 95% confidence interval [72.1-92.5]. Among asymptomatic people, there is a high prevalence of abnormal SpO2 and temperature measurements. A screening algorithm that uses both of these measures can detect asymptomatic infections with a low false positive rate.