JMIR Publications

ABSTRACT

Abstract: Epidemiologic and syndromic surveillance metrics traditionally used by public health departments may be insufficient for predicting healthcare utilization for coronavirus disease 2019 (COVID-2019). In Montgomery County, Maryland, pulse oximetry (SaO2) measurements obtained by the emergency medical service (EMS) were added to these traditional metrics to enhance the public health picture for decision makers. During a 78-day period, the rolling 7-day average of percent of EMS patients with SaO2 <94% had a stronger correlation to next day hospital bed occupancy (Spearman’s ρ = 0.58 [95% CI: 0.40–0.71]) than either the rolling 7-day average of percent of tests positive (ρ=0.55 [95% CI: 0.37–0.69]) or the rolling 7-day average of percent of emergency department visits for COVID-19-like-illness (ρ=0.49 [95% CI: 0.30–0.64]). Health departments should consider adding EMS data to augment COVID-19 surveillance and thus improve resource allocation.