Disease incidences vs volume of Google search queries
The Internet is being widely used to get health care information (1). This information research can be motivated by an affected health status: ill people might be more liable to searching health information on the web than healthy ones.
Based on this assumption, two recent works focused on words typed in the search engines Google and Yahoo to evaluate if search volumes of influenza-related queries were linked to influenza activity. They showed that this proportion was highly linked to the presence of the epidemic (2, 3).
However, published papers on this field only concerned influenza and the United States or Sweden, for one of them (4).
The Sentinel Network looked into the relationship existing in France between incidences of influenza-like illness (ILI), acute diarrhea and chickenpox and the proportion of queries of various keywords linked to those 3 diseases (5).
A high correlation was discovered between
• the proportion of queries containing the word grippe (influenza), excluding those containing the words vaccine or avian using the minus sign - , and the incidence of ILI,
• the proportion of queries containing the word gastro (gastro) and the incidence of acute diarrhea,
• the proportion of queries containing the word varicelle (chickenpox) and the incidence of chickenpox.
Continuous surveillance of Google data has been implemented by the Sentinel Network, following this study. Each Monday, data are retrieved from the Google database, via the websites Google Trends and Google Insights for Search (6, 7), and the Sentinel Network database (8). Correlations between disease incidences and the corresponding Google query search ratios are calculated. Results are presented on this web page with a graphic and a correlation table.
Based on this assumption, two recent works focused on words typed in the search engines Google and Yahoo to evaluate if search volumes of influenza-related queries were linked to influenza activity. They showed that this proportion was highly linked to the presence of the epidemic (2, 3).
However, published papers on this field only concerned influenza and the United States or Sweden, for one of them (4).
The Sentinel Network looked into the relationship existing in France between incidences of influenza-like illness (ILI), acute diarrhea and chickenpox and the proportion of queries of various keywords linked to those 3 diseases (5).
A high correlation was discovered between
• the proportion of queries containing the word grippe (influenza), excluding those containing the words vaccine or avian using the minus sign - , and the incidence of ILI,
• the proportion of queries containing the word gastro (gastro) and the incidence of acute diarrhea,
• the proportion of queries containing the word varicelle (chickenpox) and the incidence of chickenpox.
Continuous surveillance of Google data has been implemented by the Sentinel Network, following this study. Each Monday, data are retrieved from the Google database, via the websites Google Trends and Google Insights for Search (6, 7), and the Sentinel Network database (8). Correlations between disease incidences and the corresponding Google query search ratios are calculated. Results are presented on this web page with a graphic and a correlation table.
Résultats des comparaisons Google vs Sentinelles
Google data source:
Disease | Query | Correlation | Lag | p | First week | Last week |
---|---|---|---|---|---|---|
Syndromes Grippaux | grippe -aviaire -vaccin -porcine -porc -H1N1 -AH1N1 -A -mexicaine -Mexique -pandemie | 0.74 | 0 | <0.001 | 200402 | 201640 |
Syndromes Grippaux | grippe -aviaire -vaccin | 0.39 | 0 | <0.001 | 200402 | 201640 |
Syndromes Grippaux | grippe | 0.39 | 0 | <0.001 | 200402 | 201640 |
Diarrhée aiguë | gastro | 0.47 | 0 | <0.001 | 200402 | 201640 |
Varicelle | varicelle | 0.61 | 0 | <0.001 | 200402 | 201640 |
Bibliographic references
1. Fox S. Online Health Search. Pew Internet & American Life Project. 2006.
2. Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L. Detecting influenza epidemics using search engine query data. Nature. 2008 Nov 19.
3. Polgreen PM, Chen Y, Pennock DM, Nelson FD. Using internet searches for influenza surveillance. Clin Infect Dis. 2008 Dec 1;47(11):1443-8.
4. Hulth A, Rydevik G, Linde A. Web queries as a source for syndromic surveillance. PLoS ONE. 2009;4(2):e4378.
5. Pelat C, Turbelin C, Bar-Hen A, Flahault A, Valleron A-J. More Diseases Tracked by Using Google Trends. Emerging Infectious Diseases. 2009.
6. Google Insights for Search [database on the Internet]. 2009 [cited 02/27/2009]. Available from: http://www.google.com/insights/search/#.
7. Google Trends [database on the Internet]. 2009 [cited 2009/04/17]. Available from: http://www.google.com/trends.
8. Sentiweb [database on the Internet]. 2009 [cited 2009/04/17]. Available from: www.sentiweb.fr.
2. Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L. Detecting influenza epidemics using search engine query data. Nature. 2008 Nov 19.
3. Polgreen PM, Chen Y, Pennock DM, Nelson FD. Using internet searches for influenza surveillance. Clin Infect Dis. 2008 Dec 1;47(11):1443-8.
4. Hulth A, Rydevik G, Linde A. Web queries as a source for syndromic surveillance. PLoS ONE. 2009;4(2):e4378.
5. Pelat C, Turbelin C, Bar-Hen A, Flahault A, Valleron A-J. More Diseases Tracked by Using Google Trends. Emerging Infectious Diseases. 2009.
6. Google Insights for Search [database on the Internet]. 2009 [cited 02/27/2009]. Available from: http://www.google.com/insights/search/#.
7. Google Trends [database on the Internet]. 2009 [cited 2009/04/17]. Available from: http://www.google.com/trends.
8. Sentiweb [database on the Internet]. 2009 [cited 2009/04/17]. Available from: www.sentiweb.fr.