COVID-19 could disrupt how companies and FDA track drug safety

The way in which the FDA and life sciences industry track adverse events associated with drug products could be disrupted by COVID-19. AgencyIQ explores how this might occur and what it could mean for the industry.


By Kedest Tadesse, MS, RAC and Alexander Gaffney, RAC

As the life sciences industry and FDA work to confront COVID-19, the disease is disrupting the very processes and data the FDA normally uses to ensure drug safety.


After a drug has been approved, the FDA requires that companies keep track of the negative events associated with the use of the drug, such as events that are serious or unexpected. Those data are critical since the real-world use of a drug can often indicate rare or serious problems that may not have shown up in clinical trials used to substantiate the approval of that product.


This process—of tracking a drug’s adverse effects—is known as pharmacovigilance.


But in the age of COVID, this highly regulated, regimented process is increasing subject to the same disruption as the rest of the industry.


Different, complex data


COVID-19 may result in substantial decreases in the types of data being reported.


For example, it may be difficult to determine if an adverse event was caused by the drug being taken, by COVID-19, or by an interaction between the two. According to one case recounted by Murali Doraiswamy, Director of the Neurocognitive Disorders Program at the Duke School of Medicine, a 53-year-old patient taking an immunosuppressant was infected with SARS-CoV-2 and contracted a lung injury.


The potential cause of the adverse event could be the underlying condition, the medication for that condition, the virus infection or the treatment administered to treat the virus. Determining the likely cause of adverse event normally takes place in the form of a follow up report.


Pharmacovigilance events are also likely to be seen at lower rates than normal due to one reason: The number of persons seeking out in-person medical care for non-COVID related illnesses has decreased substantially. Patients are choosing not to visit the doctor’s office or emergency rooms.


For life sciences companies and the FDA, this sudden change in reported data may lead to substantial difficulty determining which data are indicative of a problem when the denominator of data is reduced. For older drugs, there may be a reduced number of reports. For new drugs, it could be difficult to determine a baseline for what “normal” looks like, however.


For other drugs, increased off-label use could result in significant changes to adverse event data. For example, hydrochloroquine sulfate, an anti-malaria drug being used to treat COVID-19 off-label, is known to increase QT prolongation. Furthermore, for patients with pre-existing conditions that are not receiving proper treatment during a pandemic, could spike reports of AE post pandemic.


And then there are the long-term implications of COVID-19. Increased use of telemedicine—both during and after COVID-19—could cause substantial changes to the practice of pharmacovigilance. Some events may be reported more (or less) by patients, or could be augmented by data from wearable devices.


FDA changes


The FDA has also announced changes to its expectations for companies handling pharmacovigilance data. On March 20, 2020, the agency issued a final guidance document recommending how companies could handle adverse event reporting.


One of the FDA’s concerns was that certain companies would be unable to continue business operations as usual due to high rates of employee absenteeism. The guidance called for two general approaches to pharmacovigilance reporting: Normal, and an abbreviated version in which only serious adverse events are reported, and all other events are reported within 6 months of the restoration of the company’s ordinary business operations.


[Read AgencyIQ’s explanation of this guidance document here.]


While the FDA is flexible on adverse reporting for non-serious events, proper data management and tracking for post pandemic reporting are fundamental. The agency expects companies to report non serious AEs that have been collected during the pandemic at a later time. Accurate data collection is critical to ensure accurate signal detection and safety.


For the FDA, there is a significant risk that the pandemic could disrupt the reporting of some safety signals that often result in labeling changes. For example, a 2017 post market safety analysis of all novel therapies approved by the FDA between 2001 and 2010 found that 32% were either withdrawn due to safety concerns, had a black box warning added to the label, or were subject to an FDA safety communication.


The FDA has been approving near-record number of drugs in recent years, indicating that one casualty of COVID-19 may be a reduced number of safety warnings from the FDA.


Data collection


There are also operational challenges like data collection methods for companies to consider, Jim Davis, Executive Vice President at Advera Health Analytics, a pharmacovigilance analytics company, told AgencyIQ in an interview.


The number one challenge of life sciences companies “is being able to transition all of their operations from what has always been a very site-specific operation with specialized technology to a remote environment,” Davis explained. For example, if the systems used by a company for pharmacovigilance aren’t cloud-based, it could be difficult for their staff to access that information from home and make use of it.


There could also be challenges associated with reaching persons who had done the initial adverse event reporting to obtain follow-up information (such as if a medical site is closed). As a result, it’s possible that adverse event reports may become less complete, Davis added.


A lack of follow up reporting could make it difficult to detect or manage “signals” of adverse events associated with drugs. Signals are dependent on how well reports are recorded and coded and to determine causality. For example, the use of a drug may be associated with a person who later died. Determining whether the drug caused the death or was simply associated with the death is sometimes difficult to determine. For example, some drugs are known to have delayed effects, which might cause unintentional drowsiness resulting in an increased risk of automobile accidents.


To contact the authors of this analysis, please contact Kedest Tadesse ( and Alec Gaffney (

Leave a Comment

Your email address will not be published. Required fields are marked *