By Brittany Wade 

April 4, 2023 | Adverse events (AEs)—any undesirable outcome caused by a drug or medical product—are critical to consider when determining the safety of therapeutic interventions. As such, clinicians and patients rely heavily on AE reporting to ascertain the best course of action for a disease or illness.  

Though little debate exists on the importance of AEs, underreporting and misreporting are common occurrences. “Underreporting has always existed, and it still exists today… from the post-marketing stage onwards. There’s also underreporting in observational studies… and clinical trials,” Su Golder, Ph.D., M.Sc., associate professor and senior research fellow in the Department of Health Sciences at the University of York, tells Deborah Borfitz, Clinical Research News senior writer and the Scope of Things podcast host.  

Golder—who spoke with Clinical Research News on this topic in 2021—points to a lack of knowledge among clinicians and patients about how and where to report as a reason for limited data. She says many healthcare professionals have never formally declared an adverse event throughout their careers. 

Aside from a lack of knowledge, other obstacles prevent consistent AE reporting. In clinical trials, researchers are encouraged to publish as many journal articles as possible, but due to strict word limits, AEs are often eliminated from the finished piece. Since the primary aim of clinical research is to focus on the benefits of a drug rather than the harms, AEs are considered easily expendable. 

AEs can also be found in highly-detailed pharmaceutical documents—clinical study reports (CSRs)—as mandated by the Open Trials initiative. However, because regulatory agencies redirected their focus to the pandemic over the past few years, CSRs have been incredibly challenging to retrieve.  

Organizations like Health Canada and the European Medicines Agency (EMA) maintain strict CSR application stipulations, making accessing data painstaking. Golder recalls requesting Human papillomavirus vaccine CSRs from the EMA. Due to certain limitations on the number of requests allowed per application, Golder and her colleagues in the Cochrane Adverse Effects Methods Group have only recovered half of the HPV AEs, despite submitting requests for over a year. To make matters worse, many CSRs arrive with much of the data redacted. 

Without access to data, researchers cannot meet industry demands. “There’s constant pressure as time goes on for researchers to amalgamate all this data quicker, faster, and more efficiently. You do the research to the best of your ability, but without all of the data, it’s really hard, and you’re under this pressure of time,” explains Golder. 

The current status of AE reporting is not entirely one of doom and gloom, as there have been some significant improvements. For example, the Yellow Card system was created for clinicians and laypeople to communicate AEs directly to the Medicines and Healthcare products Regulatory Agency, the UK’s equivalent to the U.S. Food and Drug Administration. Additionally, guidelines were produced to facilitate clear and detailed reporting, such as the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), the PRISMA harms checklist, and the Consolidated Standards of Reporting Trials (CONSORT) for AEs discovered through clinical research.  

Golder says only half of the published articles comply with CONSORT harms guidelines, so more work must be done to encourage widespread adoption. “It’s still very disappointing to see that harms aren’t included more in the journal articles, particularly as we’re more fortunate now that almost every article is electronic. You have more room in the supplementary material, at least. That’s better than nothing.” 

Adverse Events in the Real World 

With more traditional forms of data harder to come by, researchers are looking for other avenues to understand AEs in specific patient populations. Golder spends a sizable portion of her time reviewing AEs in the aggregate and looking at the differences between those reported from clinical trials and what is observed in the real world.  

“RCTs [randomized controlled trials] should be the best and highest quality data… but they have so many problems,” says Golder. For one, RCTs usually do not include the vulnerable populations likely to report an adverse event—elderly and pregnant patients as well as those with comorbidities—and rarely achieve statistical significance due to relatively small sample sizes. They are also expensive with shorter follow-up periods, preventing observations on how a therapeutic intervention affects patients after prolonged use. 

“We need data, and we need to be transparent,” says Golder. “We need an interplay of the different data sources because they all have strengths and limitations.” She uses social media, unpublished sources, and observational studies to understand the patient’s perspective better. For example, she found that symptoms like muscle pain and persistent cough—that dramatically decrease a patient’s quality of life but are often deemed unimportant compared to death, heart attacks, or stroke—are the primary reasons patients discontinue their medication.  

Golder says, “I’m not suggesting we use social media as a one-stop shop by any means. We need the high-quality data from the RCTs, but [they] can’t provide us with everything we need. If something stops a patient from taking [a drug], then it doesn’t matter how effective [that drug] is. So [patient experience] is the most important thing.”