| Popis: |
Many hospitals in the US were forced to preemptively cancel elective surgeries in March, 2020 to prepare for a surge in hospitalizations of COVID-19 patients. The resulting lost revenue, patient suffering and the probability that similar pandemics could recur motivates hospitals to find early-warning indicators for future surges. We find that state-level Google searches for COVID-19 symptom keywords "sore throat,'' "fever'' and "cough'' surge about 1-2 weeks before COVID-19-related hospitalizations surge; we adjust for increased symptom awareness during the pandemic by using searches for the technical keywords "dry cough'' and "shortness of breath'' as a measure of symptom awareness rather than a need for symptom relief. We use data from Google Trends and the Centers for Disease Control and Prevention, and give a simple early-warning notification policy based on principles of Statistical Process Control. This may give hospitals an additional tool to help decide scheduling policies during future COVID-19 waves. The approach we use may also be useful for other diseases. |
