twitter in the age of pandemics: infodemiology and infoveillance
DESCRIPTION
presented at Medicine 2.0'10 in MaastrichtTRANSCRIPT
Gunther Eysenbach MD MPH
Gunther Eysenbach MD MPHDirector, Consumer Health & Public Health Informatics Lab
Associate Professor Department of Health Policy, Management and Evaluation, University of Toronto;
Senior Scientist, Centre for Global eHealth Innovation,Division of Medical Decision Making and Health Care Research; Toronto General Research Institute of the UHN, Toronto General Hospital, Canada
Pandemics in the Age of Twitter: A Case Study of Infodemiology and Infoveillance as New Methods for Knowledge Translation Research and Syndromic Surveillance
Medicine 2.0MaastrichtNov 2010
Economists have something public health practitioners don’t have: Real-time indices to track behavior and emotions
The premise
“The Internet has made measurable what was previously immeasurable: The
distribution of health information in a population, tracking (in real time) health
information trends over time, and identifying gaps between information
supply and demand. “
Eysenbach G. Infodemiology. Proc AMIA Fall Symp 2006
Research Goals
Developing innovative tools & methods to measure/track health-related attitudes, knowledge, emotions, public attention, behavior in real time for public healthusing textual data from the Internet & Social Media
Investigate how the public is using social media during a pandemic, and how social media can be used to engage the public
Gunther Eysenbach MD, MPH, www.medicine20congress.comImage Source:
http://web2.wsj2.com/
Studying information patterns in the era of user-generated information (Web 2.0) enables us to measure user attitudes, behavior, awareness, knowledge, attention, information needs etc.
Infoveillance
• Predicting/tracking outbreaks and other public-health relevant events,
• Tracking changes in behavior, attitudes, knowledge (e.g. as a result of public health messages or interventions)
• Situational awareness regarding current concerns, issues, questions, emotions, of the public
Eysenbach G. Infodemiology and InfoveillanceJ Med Internet Res 2009: e11
http://www.jmir.org/2009/1/e11
The science of distribution and determinants of disease in populations
Epidemiology,Polls, Focus groups
Public Health ProfessionalsPolicy Makers
Public Health InterventionsPolicy Decisions
Population Behaviour, Attitudes, Health Status
Traditional Knowledge Translation Circle
PR / Media Campaigns
The science of distribution and determinants of disease in populations
Epidemiology,Polls, Focus groups
Public Health ProfessionalsPolicy Makers
Public Health InterventionsPolicy Decisions
Population Behaviour, Attitudes, Health Status
Information &Communicationpatterns
Web 1.0: Webpages, News
Web 2.0: User generated content, social media
Searches, Navigation, Clicks
Traditional Knowledge Translation Circle
PR / Media Campaigns
“Infodemiology”the epidemiology of information
Analyzing information & communication patterns (on the web)
The science of distribution and determinants of disease in populations
Epidemiology,Polls, Focus groups
Public Health ProfessionalsPolicy Makers
Public Health InterventionsPolicy Decisions
Population Behaviour, Attitudes, Health Status
Information &Communicationpatterns
Web 1.0: Webpages, News
Web 2.0: User generated content, social media
Searches, Navigation, Clicks
Traditional Knowledge Translation Circle
PR / Media Campaigns
Infoveillance
Metrics
InfovigilAggregator/Datamining/Vizualisation
InfovigilVision: an open source infoveillance prototype
Centre for Global eHealth Innovation, Toronto
Public,Clinicians,Epidemiologists
Websites
FilterKeywords / Concepts of Interest
OnlineQuestionnaires
Swine Flu / H1N1 Tweets Analytics Project
• between May 1st, 2009 and April 1st, 2010, we archived over 3 million tweets containing the keywords or hashtags (#) “H1N1”, “swine flu”, and “swineflu”.
• Also archived content of cited URLs using webcitation.org
What are people talking about in tweets?
Qualitative analysis of H1N1/Swine Flu tweets
23 %
53 %
14 %
8 %
1 %
2 %
Chew C, Eysenbach G. Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak. PLoS ONE, 2009 November 29th;5(11): e14118. http://dx.plos.org/10.1371/journal.pone.0014118.
Absolute number of tweets(Blue: swine flu, red: h1n1)
spikes mainly due to major news events e.g • [A] WHO declares pandemic, • [P] Obama declares national emergency• [B] Harry Potter actor Rupert Grint has Swine Flu
Media Resonance Analysis
Relative usage of “H1N1” terminology over “Swine Flu”H1N1:SwineFlu Ratio
• The relative proportion of tweets using “H1N1” increased from 8.8% to 40.5% in an almost linear fashion (R2= .788; p < .001), indicating a gradual adoption of the WHO-recommended H1N1 terminology as opposed to “Swine Flu”
• also social media campaigns show some effect ([G] #oink campaign of farmers)
“Happiness / Humor / Mood Index”:Smileys : Frowneys Ratio
Question IndexNumber of tweets with ? : Total Tweets
Prayer IndexNumber of tweets with “pray” : Total Tweets
H1N1 Hospitalizations / Deaths
Personal Experiences
Chew C, Eysenbach G. Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak. PLoS ONE, 2009 November 29th;5(11): e14118. http://dx.plos.org/10.1371/journal.pone.0014118.
Number of tweets with “personal experiences” correlates to H1N1 incidence
Chew & Eysenbach. Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak. PloS One 2010 (in press)
Vaccine / Vaccination Mentionings
Chew C, Eysenbach G. Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak. PLoS ONE, 2009 November 29th;5(11): e14118. http://dx.plos.org/10.1371/journal.pone.0014118.
Sentiment AnalysisH1N1 Vaccine Sentiment over Time
10
20
30
40
50
60
18-May-09 15-Jun-09 13-Jul-09 10-Aug-09 7-Sep-09 5-Oct-09 2-Nov-09 30-Nov-09 28-Dec-09
% of Sample
ANTI
PRO
negative emotion3%
paranoia/distrust
physiological safety/harm/harm to
children24%
vaccine and pandemic downplay/dissuasion
16%
dissatisfaction roll-out
negative intention5%
Anti-Vaccination Themes
Qualitative content analysis of n=689 anti-vaccination tweets
18 May - 28 Dec 2009
Conclusions• Infoveillance: New methodology, offers wealth of
quantitative + qualitative data, complementary to traditional survey methods, more timely and inexpensive
• Twitter is a rich source of opinions and experiences, which can be used for near real-time content and sentiment analysis, knowledge translation research, and potentially as a syndromic surveillance tool, allowing health authorities to become aware of and respond to real or perceived concerns/issues raised by the public
• Social media appeared underused by Canadian public health authorities during the H1N1 pandemic
“In the era of the 24-hour news cycle, the traditional once-a-day press conference
featuring talking heads with a bunch of fancy titles has to be revamped and supplemented
with Twitter posts, YouTube videos and the like.
The public needs to be engaged in conversations and debate about issues of public
health, they don’t need to be lectured to.”-Andre Picard
Picard A (2010) Lessons of H1N1: Preach less, reveal more. Globe and Mail. Available: http://www.webcitation.org/5qYZly99e.
Principal Investigator:Gunther Eysenbach MD MPH
Director, Consumer Health & Public Health Informatics LabCentre for Global eHealth Innovation
• Thanks to CIHR & Reviewers
• Cynthia Chew (MSc Student): Coding & Qualitative Analysis of Tweets
• Latifa Mnyusiwalla (MHI Student): Vaccination Sentiment Analysis
• Marina Sokolova PhD, CHEO Ottawa: Natural Language Processing
• Phil Cairns: Developer
Acknowledgements