Measuring patient-based changes in temperature could improve COVID-19 detection in nursing homes

Among nursing home residents, rather than using a one-size-fits-all threshold temperature, looking at individual changes in body temperature and temperature variability, could lead to better prediction of COVID-19 infection, a recent study has found.

Researchers conducted a retrospective cohort analysis of 6,176 nursing home residents (mean age 71.82 years, 95.74 percent men) who had been tested for COVID-19 because they manifested symptoms. Majority (n=5,262) eventually tested negative.

In the 914 residents who were positive for SARS-CoV-2, researchers observed that body temperatures started rising as early as 7 days before testing, and remained heightened for up to 2 weeks after. Of note, only 47.7 percent of COVID-19-positive participants met the 38^oC temperature threshold.

Statistical modelling showed that using a temperature change of 0.4^oC from baseline as a trigger for testing could identify 47 percent of all residents who tested positive. In contrast, standard protocols, which require a patient to cross the threshold temperature before a test is given, would only identify 39 percent of cases.

Moreover, using the temperature threshold detected the infection on average 3.6 hours later per resident, while applying the temperature-change criteria led to earlier infection detection by an average of 42 hours per resident.

Including temperature variability further improved predictive capacity. For example, a temperature variability of 0.7^oC over 3 days, combined with a change of 0.4^oC from baseline, identified 52 percent of infections at an average of 40 hours earlier.

“This approach may potentially lead to better infection control and reduce the need for sweep testing in nursing homes. These data can be used for creating early detection algorithms that will be significantly more effective when continuous temperature monitoring is available to high-risk residents,” the researchers said.