• Kyle Taylor

How can Google Trends help with Lao PDR’s COVID-19 response?




Every day, people worldwide turn to search engines like Google to find answers to their questions about COVID-19. However, more importantly, people are increasingly looking to search engines like Google for health-related information before seeking a health professional, making online search queries a valuable source of information on collective health trends.


This kind of data could have several uses. However, one stands out from the rest: to connect Google searches for COVID-19 symptoms to an uptick in cases, even before an outbreak has been detected.


We wanted to see whether Google searches related to COVID-19 were feasible for obtaining accurate data on the second wave COVID-19 outbreak in Lao PDR and whether accurate predictions could be made regarding future new cases.


To see if patterns emerged, we utilized Google Trends, a website by Google that analyzes the popularity of top search queries in Google across various regions and languages.


We also collected data on daily new cases that have been tracked and reported by Johns Hopkins Coronavirus Resource Center.





We found that the search popularity of keywords related to COVID-19 in Google moved with daily new COVID-19 cases in Lao PDR over time.




We also found that the search popularity increased with daily new COVID-19 cases. To further assess this difference, lag Pearson correlation coefficients were calculated wherein we looked for relationships between daily new COVID-19 cases and Google searches related to COVID-19 in the days prior.



We found strong to very strong correlations between Google searches for COVID-19 and daily new COVID-19 cases in Lao PDR, with R values between 0.49 and 0.69.



We also found that the correlation between Google searches for COVID-19 and daily new COVID-19 cases in Lao PDR were strongest for Google searches five days before the daily new cases were observed.


Based on the correlations, COVID-19 patients could be seeking symptom information on search engines, such as Google, before seeing a health professional and getting tested.


On average, the time from exposure to symptom onset is roughly five to six days. However, studies have also shown that symptoms could appear as soon as three days after exposure to as long as 13 days later.


The correlations between Google searches for COVID-19 and daily new COVID-19 cases in Lao PDR alone cannot reveal the truth about the spread of the virus. Nevertheless, it does suggest that it can be used as one of several datasets when predicting when the next outbreak might occur.


Click the link below to download our working paper.

Using Google Trends for COVID
.pdf
Download PDF • 524KB