Home Monitoring for Fever: An Inexpensive Screening Method to Prevent Household Spread of COVID-19.

Abstract

We compared home temperature measurements from a convenience sample of 1180 individuals who reported being test positive for SARS-CoV-2 to an age, sex, and location matched control group of 1249 individuals who had not tested positive. All individuals monitored their temperature at home using an electronic smartphone thermometer that relayed temperature measurements and symptoms to a centralized cloud based, de-identified data bank. Individuals varied in the number of times they monitored their temperature. When temperature was monitored for over 72 hours fever (≥ 37.6°C or 99.7°F or a change in temperature of ≥ 1°C or 1.8°F) was detected in 73% of test positive individuals, a sensitivity comparable to rapid SARS-CoV-2 antigen tests. When compared to our control group the specificity of fever for COVID-19 was 0.70. However, when fever was combined with complaints of loss of taste and smell, difficulty breathing, fatigue, chills, diarrhea, or stuffy nose the odds ratio of having COVID-19 was sufficiently high as to obviate the need to employ RTPCR or antigen testing to screen for and isolate coronavirus infected cases. Our findings suggest that home temperature monitoring could serve as an inexpensive convenient screen for the onset of COVID-19, encourage earlier isolation of potentially infected individuals, and more effectively reduce the spread of infection in closed spaces.