Need of care in interpreting Google Trends-based COVID-19 infodemiological study results: potential risk of false-positivity
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
Background
Google Trends (GT) is being used as an epidemiological tool to study coronavirus disease (COVID-19) by identifying keywords in search trends that are predictive for the COVID-19 epidemiological burden. However, many of the earlier GT-based studies include potential statistical fallacies by measuring the correlation between non-stationary time sequences without adjusting for multiple comparisons or the confounding of media coverage, leading to concerns about the increased risk of obtaining false-positive results. In this study, we aimed to apply statistically more favorable methods to validate the earlier GT-based COVID-19 study results.
Methods
We extracted the relative GT search volume for keywords associated with COVID-19 symptoms, and evaluated their Granger-causality to weekly COVID-19 positivity in eight English-speaking countries and Japan. In addition, the impact of media coverage on keywords with significant Granger-causality was further evaluated using Japanese regional data.
Results
Our Granger causality-based approach largely decreased (by up to approximately one-third) the number of keywords identified as having a significant temporal relationship with the COVID-19 trend when compared to those identified by Pearson or Spearman’s rank correlation-based approach. “Sense of smell” and “loss of smell” were the most reliable GT keywords across all the evaluated countries; however, when adjusted with their media coverage, these keyword trends did not Granger-cause the COVID-19 positivity trends (in Japan).
Conclusions
Our results suggest that some of the search keywords reported as candidate predictive measures in earlier GT-based COVID-19 studies may potentially be unreliable; therefore, caution is necessary when interpreting published GT-based study results.
Article activity feed
-
-
-
SciScore for 10.1101/2020.12.30.20249066: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
No key resources detected.
Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:Our study has some limitations. For example, in the VAR model, the effect of each variable is assumed to be fixed throughout the reviewed period, which may not always be true because the public interest and attitude toward COVID-19 could vary over time [21]. This can be suspected by the decreased peak of GT trend for the “COVID” keyword …
SciScore for 10.1101/2020.12.30.20249066: (What is this?)
Please note, not all rigor criteria are appropriate for all manuscripts.
Table 1: Rigor
NIH rigor criteria are not applicable to paper type.Table 2: Resources
No key resources detected.
Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:Our study has some limitations. For example, in the VAR model, the effect of each variable is assumed to be fixed throughout the reviewed period, which may not always be true because the public interest and attitude toward COVID-19 could vary over time [21]. This can be suspected by the decreased peak of GT trend for the “COVID” keyword in the second wave (Fig 2, in Australia, Japan, and the United States). In future studies, state space modeling [22] to incorporate potentially time-varying effects may be useful to overcome the potential weakness of the VAR model, especially when the included period becomes so long. In addition, the keywords’ media coverage was adjusted only in Japanese regional data, which makes the obtained results slightly less generalizable to other countries. The Nikkei telecom we used for media review would not cover all potentially influencing media such as TV talk shows, or social media (e.g., Twitter [23] or Instagram [24]). To conclude, our current results using a more statistically favorable approach suggest that many of the search keywords identified as candidate predictive measures in earlier GT studies have the potential risk of false positives, and that we need to be careful in interpreting the earlier GT-based COVID-19 study results.
Results from TrialIdentifier: No clinical trial numbers were referenced.
Results from Barzooka: We did not find any issues relating to the usage of bar graphs.
Results from JetFighter: We did not find any issues relating to colormaps.
Results from rtransparent:- Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
- Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
- No protocol registration statement was detected.
-
