JMIR Publications

ABSTRACT

Background:

The emergence and media coverage of COVID-19 could have affected influenza search patterns, possibly impairing the possibility of using Google Trends (GT) on influenza for surveillance purposes.

Objective:

To investigate if the emergence of COVID-19 associated with modifications in influenza search patterns in the United States.

Methods:

We retrieved GT data for United States searches on Influenza, Coronavirus disease 2019, and shared symptoms between influenza and COVID-19. We assessed the correlation of GT influenza data with GT COVID-19 data for a period of one year after the first diagnosed case of COVID-19 in the United States. We constructed a seasonal autoregressive integrated moving average (SARIMA) model to analyze how influenza data that was predicted using four years of past data differed from actual GT relative search volume, building a similar SARIMA model for GT shared symptoms data. Lastly, we assessed the correlation between GT influenza data and CDC influenza-like-illness data for the past five years.

Results:

We observed a weak, non-significant correlation between GT data on COVID-19 and influenza (ρ=-0.171; p-value=0.226). Influenza search volume for 2020-2021 distinctly deviated from values predicted by SARIMA models – for six weeks, the observed volume of searches was higher than the upper bound of 95% confidence intervals for predicted values. Similar results were observed with data from shared symptoms between influenza and COVID-19. The correlation between GT influenza data and CDC influenza-like-illness data decreased since the emergence of COVID-19 (ρ=0.643 for 2020-2021, down from ρ=0.902 in the previous year).

Conclusions:

Relevant differences were observed between predicted and observed influenza GT data for one year since the onset of COVID-19 in the US. Such differences are possibly due to media coverage, pointing to the limitations of GT as a flu surveillance tool.