Intended for healthcare professionals
Characterization and corroboration of safety signals identified from the US Food and Drug Administration Adverse Event Reporting System
A critical component of drug safety monitoring is post-marketing pharmacovigilance, which through a combination of voluntary and active surveillance efforts, provides early identification of potential adverse drug events (AEs).12 Identifying drug safety issues after regulatory approval is common; the US Food and Drug Administration (FDA) initiates a regulatory safety action for approximately one-third of newly approved drugs, on average of four years after approval.3
In the US, FDA surveillance efforts rely on the FDA Adverse Event Reporting System (FAERS) —a database of spontaneous adverse event reports from drug manufacturers, healthcare personnel and consumers that accounts for more than 2 million AEs annually.4 FAERS has well known limitations including underreporting, redundancies, selection bias and incomplete data.56 More recent surveillance efforts have leveraged aggregated claims or electronic health record data for proactive surveillance, such as FDA’s Sentinel Initiative.7
To better understand the FDA’s pharmacovigilance efforts, in our linked study in The BMJ,8 we characterised safety signals identified from FAERS from 2008 to 2019 and determined how often these signals resulted in FDA regulatory actions. Next, among a subset of signals from 2014 and 2015, we ascertained whether these regulatory actions were corroborated by the studies published in the medical literature or by Sentinel Initiative assessments.
We found that approximately 70% of the 603 potential safety signals reported between 2008 and 2019 were resolved, but just over 30% remain unresolved—the majority of which were from the last three years’ public quarterly reports. Among resolved signals, 22% resulted in no regulatory action while 78% were followed by regulatory safety actions, including drug label changes for three-quarters and drug safety or other public communications from FDA for one-sixth.
Among the subset of 82 signals reported in 2014 and 2015, most were resolved and resulted in regulatory actions. Moreover, the majority were associated with at least one relevant study in the literature. However, 70% of these studies were case reports. Only one-third of resolved signals were associated with one or more relevant studies published in the medical literature that corroborated the FDA’s regulatory action, none with public Sentinel Initiative Assessments. Few of the identified regulatory actions, when resulting in a label change or drug safety communication, cited additional corroborating evidence besides the safety signal identified from FAERS.
Our study points to important considerations for the FDA to ensure timely and robust assessment of potential safety signals from FAERS. First, it highlights the need for deadlines for reviewing safety signals. While the FDA Amendments Act requires FDA to publicly report potential safety signals, there is no strict deadline for completion of safety signal evaluations.9 Encouraging prompt evaluation and resolution of all safety signals within a specific timeframe, along with annual public reporting of progress, may help in the future to further streamline this process.
Further, our findings suggest that robust and larger scale post-market studies are needed to evaluate drug safety, as many regulatory safety actions may only be based on AEs submitted to FAERS or other case studies. This highlights the potential to leverage real world data sources for more robust evaluations, which can in turn be used to prompt trials when needed.
Finally, our study highlights the importance of transparency regarding safety signal evaluations. For instance, FDA recently reported the resolution of several signals which had not been posted due to an administrative error.10 In addition, greater transparency is needed around the evidence used to support regulatory actions to resolve safety signals, including what data were used when making the decision. Such transparency is important as clinicians and patients may change prescribing or use patterns based on subsequent regulatory actions, and would benefit from understanding the strength of the evidence behind these actions.
Disclosures: Meera Dhodapkar currently receives research support through the Richard K Gershon, MD Fund at Yale School of Medicine, and is an Associate Editor for Visual Abstracts for the North American Spine Society Journal. Dr Ross currently receives research support through Yale University from Johnson and Johnson to develop methods of clinical trial data sharing, from the Medical Device Innovation Consortium as part of the National Evaluation System for Health Technology (NEST), from the Food and Drug Administration for the Yale-Mayo Clinic Center for Excellence in Regulatory Science and Innovation (CERSI) program (U01FD005938), from the Agency for Healthcare Research and Quality (R01HS022882), from the National Heart, Lung and Blood Institute of the National Institutes of Health (NIH) (R01HS025164, R01HL144644); in addition, Dr Ross is an expert witness at the request of Relator’s attorneys, the Greene Law Firm, in a qui tam suit alleging violations of the False Claims Act and Anti-Kickback Statute against Biogen Inc. and is the US. Outreach and Associate Research Editor at the BMJ. Drs. Ross and Ramachandran receives research support from Arnold Ventures for the Yale Collaboration for Research Integrity and Transparency (CRIT). Dr Ramachandran also receives additional research support from the Stavros Niarchos Foundation through Yale Law School for a project entitled “Re-envisioning Publicly Funded Biomedical Research and Development.”
Provenance and peer review: commissioned, not peer reviewed.
Respond to this article
Register for alerts
If you are unable to import citations, please contact technical support for your product directly (links go to external sites):
Thank you for your interest in spreading the word about The BMJ.
NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail. We do not capture any email address.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. 
Read related article
See previous polls