It would seem that the exchange of health information on social media and the Internet is an obvious opportunity to gain insight into emerging disease outbreaks. With new initiatives like Google Flu Trends and HealthMap, there are more ways than ever before to monitor outbreaks. But how useful are these initiatives for public health practitioners who are trying to detect emerging diseases in their own regions?
Now, a new study published in The Milbank Quarterly, a peer-reviewed journal of population health and health policy, focuses on the challenges facing practitioners as they consider ways to integrate social media and Internet data into the detection and management of disease outbreaks. The study, which looks at 20 years of published research about event-based surveillance systems, was led by Edward Velasco, PhD, Senior Scientist, and others at the Robert Koch Institute, a national public health institute, in Berlin. David M. Hartley, PhD, MPH, Associate Professor of Microbiology and Immunology at Georgetown University Medical Center, provided a commentary on the study.
The researchers examined both old and new types of public health surveillance, including 13 surveillance systems that are currently used. They then assessed the challenges facing public health officials as they attempt to integrate the new and old systems.
“This study shows that social media and the Internet are changing event-based surveillance as we know it and hold promise for the future. However, it appears that this promise remains unrealized—and that the exchange of health information on the Internet cannot yet be fully relied upon for this important public health function,” says Howard Markel, MD, PhD, Editor-in-Chief of The Milbank Quarterly and the George E. Wantz Distinguished Professor of the History of Medicine, Center for the History of Medicine, University of Michigan.
Older systems, new information
The two types of public health surveillance systems are indicator-based and event-based surveillance. Indicator-based, the oldest and most common, is widely used by regional, national and international public health agencies. These systems are designed to collect and analyze structured data based on protocols tailored to each disease, including calculating the incidence, seasonality and burden of disease. Their goal is to find increased numbers or clusters that might indicate a threat. There is generally a time lag between the occurrence of an event and the indicator-based surveillance—and these systems lack the ability to detect potential threats more quickly. In addition, these systems are not equipped to detect new or unexpected disease occurrences—because they only collect predefined epidemiological attributes for each disease. That’s why the first cases of severe acute respiratory syndrome coronavirus (SARs-CoV) in Asia, for example, which was a new strain of viral infection, were not detected.
Instead of relying on official reports, event-based surveillance information is obtained directly from witnesses of real-time events or indirectly from a variety of communication channels, including social media and established alert systems, and information channels, including the news media, public health networks and nongovernmental organizations. Because it occurs in “real time,” event-based surveillance can identify events faster than indicator-based surveillance as well as new events not picked up at all by indicator-based surveillance. Health information monitored via the Internet and social media is an important part of this event-based surveillance—and is most often the focus of existing event-based surveillance systems. Research has shown that event-based surveillance identifies trends comparable to those found using established indicator-based surveillance methods. But in practice event-based surveillance systems have not been widely accepted and integrated into mainstream use by national and international health authorities mainly because they have not yet been systematically evaluated within a public health agency.
Factors that might influence integration into official systems
The study identified the 13 event-based surveillance systems that are currently used, including the type of data they collect and how they are used.
They found four limitations associated with integrating new surveillance systems into officially existing systems:
- Information is not always moderated by professionals or interpreted for relevance before it is disseminated to epidemiologists
- There is no standardized system for updates, often resulting in too much information
- Algorithms and statistical baselines are not well developed
- New information about health events is not disseminated in the most efficient way.
Challenges of integrating event-based surveillance
Velasco and coauthors found that no event-based surveillance systems are currently part of national programs for surveillance, although they are used intermittently and as complementary sources of information. The authors surmise that event-based surveillance could improve surveillance activities, but systematic evaluation within a public health agency is needed before this can happen.
Learn more about public health surveillance through sessions and workshops at the 2014 Public Health Informatics Conference:
- Electronic Disease Surveillance Systems in Support of State and Regional Health Departments
- Public Health Reporting Initiative (PHRI), ONC Standards and Interoperability Framework: A Standardized Approach to Electronic Public Health Reporting
- Public and Population Health: Structured Data Capture, Surveillance, and Findings and Implications
- Transitioning to Standards-Based Electronic Reporting and Bi-Directional Exchanges of Public Health Data with Clinicians: Lessons from Five Pilot Projects