More people are online today searching and sharing health information. Social media channels such as Twitter and Facebook, commenting on blog posts and participating in online communities such as tweet chats, facilitate this exchange of information. While people are participating and sharing their health experiences with others, it is a given that there will be discussion of treatments and possible adverse events. Regulatory agencies such as FDA in the US and EMA in Europe, monitor and collect data to ensure the safety of the public. Pharmaceutical companies also have the obligation to report to regulatory agencies adverse events. The question is can social media data mining and social media listening prove be useful in early detection of adverse events?
So what is an Adverse Event (also abbreviated AE or AEs)?
An adverse event can be defined as “Undesirable experience that is associated with the use of a medicinal product in a patient”.
In order to report an AE to the FDA, four criteria must be met and those are:
- an identifiable patient;
- an identifiable person reporting the event;
- a specific drug or biological product;
- an adverse event or fatal outcome suspected to be due to the suspect drug or product
What could be some of the opportunities of social media data mining or social media listening for adverse events?
- It could allow for additional information to the already established safety profile of a medicine
- Social media could be more useful for early trends detection or reveal new patterns, which can then be further investigated through research or observational studies
- New technologies, could offer new possibilities for monitoring drug safety
- It could provide for geo-monitoring for drug safety
What could be some of the challenges of social media data mining or social media listening for adverse events?
- Adverse events or potential AEs can be ambiguous, especially due to the word or character limitations (for example Twitter 140 characters)
- People use different language in social media (social media vernacular) such as hashtags, abbreviated words, slang words
- There is a need to account for misspelled words, different languages, retweets or reposting, that is duplication within the same platform and / or across multiple platforms, and fraudulent postings
- Possible bias regarding the demographics of Twitter users, such as age and location
- People may fear being monitored or feel “Big Brother” is watching over your shoulder
What is being done to try and capture adverse events in social media?
Web-RADR (Recognizing Adverse Drug Reactions) Europe is funded by the Innovative medicines Initiative, which in turn is funded by the European Commission and the European Federation of Pharmaceutical Industries and Associations. The initiative also involves drug regulators, pharmaceutical companies and the World Health organization and universities.
Epidemico, a health informatics company and a subsidiary of Booz Allen Hamilton, is working on developing algorithms to pick up adverse events from social media as part of Web-RADR project. The FDA is also collaborating with Epidemico to develop the digital data-mining platform, MedWatcher Social. MedWatcher Social (MWS) is a validated GxP (good for practice) digital listening platform that is designed to monitor social media information about drug safety.
Treato, a company based in Israel, the company offers a reporting tool so that organizations can log adverse drug reactions and it’s website allows patients to search for information that has been collected about other drugs. The goal for Treato is to capture early drug signals and improve patient outcomes.
IMS Health, through it’s social media analytics company Semantelli, has been analyzing adverse events and picking up signals on prescription drugs and over the counter drugs. Also IMS Health has been working on capturing signals on efficacy of drugs.
This sounds like an overwhelming task?
Yes it may well be overwhelming, but the information shared on the internet cannot be ignored today. These networks play an important role in consumer health management for today’s modern patient, caregiver and consumer. However when confronted with regulatory risks and considerations, such as reporting of adverse events, healthcare industry feel the risks may outweigh the benefits in social media listening or engaging.
Social media and pharmacovigilance: A review of the opportunities and challenges. Richard Sloane, prod Osaniou, David lewis, Danushka Bollegala, Simon Maskell and Munir Pirmohamed. Br J Clin Pharmacol:Volume 80, Issue 4, pages 910-920, October 2015.
Systemic review on the prevalence, frequency and comparative value of adverse events data in social media. Su Golder, Gill Norman and Yoon K Loke. Br J Clin Pharmacol: 2015 Oct; 80 (4): 878-88.
Digital drug safety surveillance: monitoring pharmaceutical products in twitter. Freifeld CC, Brownstein JS, Menone CM, Bao W, Filice R, Kass-Hout T, Dasgupta N. Drug Saf. 2014 May;37 (5):343-50.
IMI Web-RADR Webinar: mobile technologies and social medias as new tools in pharmacovigilance.