Using social media as a public health surveillance tool

Social media. Those two words have drastically changed how we communicate over the last decade.

Information is available instantly, in real-time, with people worldwide providing status updates with pictures and videos with attached timestamps and location information. From public sentiment about national issues to backlash about a local development, from an eye witness account of an event to a review of a popular restaurant, from photos and videos of what people are up to with family and friends to how they're feeling that day – social media media is a wealth of information.

Social media has become an increasingly relevant tool for public health surveillance, as a real-time census, with organizations using social media to detect, monitor and predict epidemic trends to facilitate preparedness and rapid response. This work is especially pertinent in light of the emergence of globalized and growing epidemics for which there is no known cure.

The World Health Organization confirmed its position this year that public health surveillance can provide early warning for impending disaster, record keeping for the purpose of public health monitoring and evaluation and an evidence base for epidemiology, as well as for decision making. When combined with the ever growing use of social media, public health surveillance faces endless opportunities and a myriad of challenges.

Sickweather, a sickness forecasting and mapping tool recently named a Tech Crunch best from 500 Startups, has released a new white paper titled "Social Media as a Public Health Surveillance Tool: Evidence and Prospects." The project reviewed 103 articles on the use of social media for public health surveillance, and provides a handy summary of their research and conclusions.

Public health surveillance opportunities

Several articles studied highlighted the cost-effectiveness, detailed geographical information, timeliness, and sharing of public health information provided by social media public health surveillance in tracking population health sentiments, behaviors, outcomes and emergencies and even forecasting health concerns.

Current means for tracking health epidemics via health information exchanges (HIE) provide real-time clinical data, but are fragmented geographically among so-called data silos and data deserts, and don't capture information at all on the population of those that don't go to the doctor when they get sick, which is unfortunately a growing number of people according to the 2010 Census. Meanwhile, hospital and state surveys help capture additional data from outside the healthcare system, but they are quite expensive and take time to collect and process, as compared to social media that provides localized and characterized information in real-time posts. This enables rapid and focused detection of disease given the much finer temporal and spatial resolution of disease monitoring information provided by social media.

For example, there was a correlation between flu case data from the World Health Organization and tweets that were self-protective in response to epidemics, such as those regarding increased sanitation or avoiding gatherings. These findings are echoed in the research of Sickweather's advisors from Johns Hopkins University in the whitepaper "You Are What You Tweet" (Dredze/Paul), and in the research of our friends at Harvard Medical School who found similar correlations in their paper "Social and News Media Enable Estimation of Epidemiological Patterns Early in the 2010 Haitian Cholera Outbreak" (Chunara/Andrews/Brownstein).

Social media also allows for the sharing of public health information by health organizations and practitioners. For example, Twitter may be used for communicating with the public on surveillance and interventions, and for sharing resources. And, Facebook is an ideal platform for sharing experiences, receiving feedback and asking questions to individuals with whom they share health conditions such as diabetes. Case in point is Facebook's recent partnership with UNICEF in which they worked together to provide advice and resources about Zika to those on Facebook who were concerned about getting the illness and living in high risk areas. This sharing of information on social media enables the democratization of public health knowledge, connecting health experts and the public in the exchange of knowledge.

An opportunity exists in using social media to not only access current health issues, but also to forecast health conditions. Models for social media forecasting are being tested and demonstrating effectiveness. For example, Sickweather's illness forecast is based on a combination of social listening through its patent-pending social media monitoring algorithm; crowdsourcing; population, sales and clinical data; extensive archived data to predict the rate of National influenza activity up to 15 weeks in advance with 91% accuracy.

Validity of surveillance data and ethical concerns

But this isn't without it's challenges. While traditional surveillance means face the challenges of high costs, imprecise resolution and the possibility of bias, social media faces questions regarding validity. Most times, the number of geotagged tweets is few relative to the volume of tweets, which are in turn geo-located to a 'best guess' proximity at the state or metro level. In addition, only a subset of the population is on these platforms (most ages 5-24 and 35-49). There is a need for more scientific research on the validity of surveillance data and text mining, although correlation studies remain positive enough to promote the success and practicality of these methods.

Also, more studies are needed to investigate the validity of social media public surveillance across geographical and cultural contexts, as many of the studies reviewed in Sickweather's white paper were in the United States, and only a few focused on other countries such as Turkey, China, South Korea, United Kingdom, Haiti, Mexico, Australia and Latin America. And although Sickweather's own efforts have been global, they have only been for localizing data in regions where English is spoken; however, we are now beginning the process of localizing in French, Spanish, German, Portuguese and Hindi among the countries where those languages are spoken.

There is much information on social media that interferes with effective surveillance. How do you eliminate the noise and distinguish fact from fake news? It is possible for inaccurate information or information based on vested interests, such as that of industry, to be shared with people seeking health information on the Internet and create an accountability and authenticity challenge. We see these challenges in our own surveillance data, and have developed ways to overcome the majority of them.

There is also need to build infrastructural capacity in health institutions to handle the large volumes of data obtainable from social media. This includes developing disease specific lexicons and more studies on language processing while measuring variation in the effectiveness of machine learning systems across diseases. Ideally, these data could be another spoke in the wheel of the aforementioned HIE's, whereupon all organizations involved in the continuum of care can benefit from them, but that will largely depend on the success of the HIE model across more geographies.

Others challenge the ethical concerns of using social media as a surveillance tool, and the right to use the public's information when consent is not expressly given. Certainly when the information is directly crowdsourced via an app like Sickweather as a 'Waze app for sickness' the expectation is clear that others will benefit from the information being shared. That is not always the case when people are publicly posting their symptoms on social media, although it is implied by the public nature of those posts.

Future and forecasting of public health surveillance

The future of social media based public health surveillance lies in the use of social media to not only identify health outcomes, but as a possible avenue for forecasting population health trends and for delivering interventions. While researchers do not view social media as a stand-alone means for public health surveillance, in light of concerns for the scientific veracity of surveillance procedures, there is strong interest in the use of social media to supplement traditional public health surveillance infrastructure.

Traditional case-based surveillance systems, such as electronic health records, have high veracity but lack granularity and timeliness – and patient generated data such as social media have high velocity variety with the scientific level of validity and reliability. These systems have complementary strengths, and more research, funding and partnerships are needed to explore integrating social media with traditional surveillance systems and building health worker capacity for such levels of analysis.

Future studies will need to move beyond observational analyses to testing the procedures for mining crowd sourced data as well as the validity of social media for forecasting public health trends. This includes further research on natural language processing and parts of speech tagging to reduce noise, increase sensitivity and extract useful information. Also, more structured best practices are needed for geocoding to enable tracking of point of origin and propagation rate of disease symptoms.

If explored, these opportunities will yield more advanced insights on how to optimize the use of social media for public health surveillance. New social media public health surveillance models like Sickweather provide the opportunity to track health conditions in a timely and cost-effective manner, especially for outbreaks for which there is no cure. Social media can allow for the spread of information, for collaboration and for effective detection and monitoring of outcomes.

Graham Dodge, Co-Founder & CEO
Graham is an entrepreneur with diverse experience in design, marketing and business development. In 1998, he was a member of the team that created the first online Crime Map, a visualization of aggregate crime data for the web portal Crime.com. Throughout his career he has worked in varying capacities for such brands as GlaxoSmithKline, Discovery, GEICO, AOL & MTV. He is a board member of PACT: Helping Children with Special Needs, an independent affiliate of the Kennedy Krieger Institute, that provides services for medically fragile kids. He was named a "Social Innovation Rockstar" by Yoxi/BOLD, and featured among Broadband for America's "Faces of Innovation." For more information visit www.Sickweather.com.

The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.

 

Featured Whitepapers

Featured Webinars