Adverse event data from protocol-driven, controlled clinical trials forms the backbone of any rigorous safety assessment on a new drug product. Nevertheless, lawyers defending pharmaceutical agents in product liability litigation must maintain a detailed awareness of spontaneously reported adverse events associated with drug exposures. As part of this effort, a good litigator must develop an understanding of what FDA says about the risks of specific drugs.
Monitoring the safety of drug and medical device products is fundamental to FDA’s role in protecting the public’s health and welfare. While FDA is in the process of launching active surveillance approaches to assess drug safety (e.g., the Sentinel Initiative), passive surveillance remains one of the key pillars of FDA’s efforts to identify adverse event signals of potential importance. In today’s world, the key passive surveillance system is the FDA’s Adverse Event Reporting System or FAERS.
As if accurately and appropriately interpreting passive surveillance trends were not challenging enough for drug and device lawyers, adverse events are now being described well beyond the FAERS database. Specifically, adverse health descriptions are being posted on support group websites, social media (e.g., Facebook, Twitter, Instagram), message boards, and blogs (sometimes called proto-AEs).
Should Internet Posts be Considered Reportable Adverse Events?
Technically, to be a reportable adverse event an applicant (i.e., the drug company) must have the following four pieces of information:
- An identifiable patient
- An identifiable reporter
- A suspect drug or biological product
- An adverse experience
Based on these elements, most of these proto-AEs posted on internet sites would not be considered reportable AEs in a regulatory sense. How can they confirm the identity of the patient or the reporter? And to date, FDA has not provided any clear guidelines on this issue. While FDA has issued a guidance on regulatory requirements for postmarketing submissions of interactive promotional media, they have not developed any clear guidances on reporting of adverse events based on internet posts or posts on social media. How far must a company go to track down the required elements of a proto-AE? We know that drug manufacturers should be monitoring the published scientific literature for unreported AEs. Should manufacturers now be developing SOPs to monitor social media sites for proto-AEs to determine if they are reportable AEs?
Comparing Social Media AEs With Traditional AEs
Despite these open questions and uncertainties, there have been some interesting academic studies of the reliability and accuracy of proto-AEs posted on social media sites. One study performed jointly by FDA, Epidemico, and Harvard Medical School found clear similarities between traditional AEs and proto-AEs posted on Twitter (see figure, below).
Another group reported that social media sources contain different types of information than that reported in FAERS and that these posts are made by different demographic groups. Specifically, posts on social media tended to be made by individuals who were younger and their reviews focused on fewer and less serious AEs.
Finally, a recent systematic review reported that there was a higher frequency of AEs on social media sites and they skewed toward “symptom-related” events that tended to be milder and less severe than events reported in FAERS. The authors concluded that while AEs are identifiable within social media, there is considerable heterogeneity in the frequency and type of events reported. They close by highlighting that the reliability and validity of AEs reported on social media have not been thoroughly evaluated.
Collaborations and Research Tools
Based on this emerging knowledge, in June of 2015, FDA signed a research collaboration with PatientsLikeMe to determine how patient-reported data can give new insights into drug safety. As part of this collaboration, FDA hopes to develop a clearer understanding of the relevance of social media and internet sites to the reporting of AEs and their relevance to the rigorous and valid assessment of drug safety. There are even organizations that have developed specifically designed tools to mine social media for proto-AEs (see this tool developed by Epidemico and this one by AskaPatient).
Potential Opportunities, and challenges
Social media is a fast growing data source for many areas of medicine and science. While a full discussion of the opportunities and risks related to this topic are beyond the scope of this article, I provide below just a few potential areas for consideration by litigators (also check out this post for more discussion on social media monitoring and AEs).
- Posts on social media may uncover new benefits associated with a drug or device
- Assessment of social media AEs may uncover AE trends more rapidly than FAERS
- Assessment of social media AEs may uncover clusters that warrant more detailed investigation for potential alternate causes
- Plaintiff’s social media history can be mined to ascertain facts related to a case
- Language used on social media often differs from conventional medical or everyday spoken language – therefore keyword searching may be tricky
- Confirming veracity of posts may prove difficult
Overall, listening on social media in its present state cannot be used to draw specific conclusions about true adverse drug reactions. Nevertheless it may help identify trends and areas that warrant further investigation.
Don’t let junk science corrupt the facts in your case. Download our eBook, The Litigators Guide to Combating Junk Science – 2nd Edition. Simply fill out the form below.