A Fresh Take on the Adverse Event Landscape

Clinical Researcher—February 2019 (Volume 33, Issue 2)


Ann Neuer, MBA




The subject of adverse events (AEs) is often summed up as a series of definitions and reporting requirements, but more recently, there has been an effort to breathe life into this unwieldy topic, bringing greater understanding to the role of AEs in the clinical trial process. The U.S. Food and Drug Administration (FDA) provides guidance on the complex subject of AE reporting,{1} but at the same time, a growing body of literature details an inadequate level of reporting by stakeholders, especially of the more serious AE variety.

Perhaps this dilemma reflects the broad scope of this subject, as there are AEs, serious adverse events (SAEs), suspected adverse reactions, and unexpected AEs, and their proper reporting forms the basis of the critical risk/benefits analysis of investigational therapies. Even the simplest online search of AEs reveals the vastness of this topic, but most publications seem to focus on one aspect only—the industry-wide challenge of collecting complete AE information for already marketed products, known as postmarketing surveillance.

By comparison, the literature on the quality of AE and SAE reporting during ongoing clinical trials is scant, at best, and merits further exploration. In particular, do most investigators understand their reporting responsibilities for various types of AEs? This is a fair question, considering that multi-year data from the Bioresearch Monitoring Program (BIMO), the FDA inspection plan, indicate that “failure to report AEs” is a continual clinical investigator deficiency, and falls under the category of “inadequate subject protection.”{2}

This article takes a look at the changing AE landscape from the clinical trial perspective. Stakeholders share their perceptions as to how clinicians interpret information that may or may not be a type of AE, and why there are reporting challenges. There is discussion on the underreporting of SAEs and AEs by investigators and sponsors, which leads to biased evidence and possibly serious consequences for patients.{3,4} Also presented are recent FDA rules and guidance on new standards for low-risk studies, which do not seem to be making changes to AE reporting. Finally, this article describes the importance of asking the right questions to study subjects in order to determine if an AE has actually occurred.


Figure 1: Definitions of Various Types of Adverse Events

Adverse event (AE): Any untoward medical occurrence associated with the use of a drug in humans, whether or not considered drug related.

Serious adverse event (SAE): In the view of either the investigator or sponsor, an event that results in any of these outcomes: death, a life-threatening AE, inpatient hospitalization, or prolongation of existing hospitalization…. (partial definition)

Suspected adverse reaction: Any AE for which there is a reasonable possibility that the drug caused the AE.

Unexpected adverse event: An event not listed in the investigator brochure or not listed at the specificity or severity that has been observed…. (partial definition)

Source: 21 Code of Federal Regulations 312.32(a)


A Complex Picture

An AE is defined as any untoward medical occurrence associated with the use of a drug in humans, whether or not considered drug related{5} (see Figure 1), and there are a whole host of reasons as to why the reporting of “untoward medical occurrences” during clinical trials can be less than optimal. These range from needing a better understanding of the various types of AEs, to how investigative sites should delineate between AE reporting requirements for the protocol and for their institutional review board (IRB), to differing reporting requirements for pharmaceutical and device trials.

To complicate matters further, it is possible that how AEs are observed is largely subjective. A thoughtful opinion piece was recently published in Clinical Researcher by Robert Jeanfreau, MD, medical director with MedPharma,{3} in which he explores how subjectivity, rather than objectivity, is commonplace among stakeholders and could be a root cause of problematic AE reporting practices.

Specifically, Jeanfreau writes about how the role of bias has been studied extensively in clinical trials, but when it comes to AEs, he claims there has been surprisingly little attention paid to how they are collected and how bias may determine whether AEs are viewed as such and whether they are reported at all. For example, he points out that an investigator may interpret a slight decrease in the hematocrit as an AE, and report it as such, but an increase in the hematocrit of similar magnitude may not be viewed as an AE, and therefore, not reported.

As a clinician with more than a decade of experience in conducting clinical trials, Jeanfreau notes that most scientific endeavors are based on objective techniques of the scientific method, whereas the collecting of AEs is not, and therefore allows for the introduction of bias. “In clinical trials, we don’t take a scientific, objective view of AEs because we are starting from the perspective that if something happens, it is an adverse event, suggesting something negative. That is already a biased view. It would be more objective to observe the event as a ‘change in health,’” Jeanfreau explains. (A new peer-reviewed article related to AEs by Jeanfreau can be found elsewhere in this issue of Clinical Researcher.)

Jeanfreau comments further that subjective perceptions of AEs are sometimes the result of overly simplistic thinking, especially when it comes to evaluating laboratory data for the possibility of an AE. “There is confusion as to how some clinicians and non-clinicians view blood work results. If a lab test is significantly out of range, it may be considered an AE. But oftentimes, the evaluation of lab data is much more complicated than that. For example, even a normal lab result can be an important finding. Over time, if there are four hematocrit readings, but each is lower than the previous one, yet still in the normal range, this is clearly a downward trend, and could be significant. This is different than simply assuming that only abnormal values can qualify as AEs,” Jeanfreau says.

Inadequate Reporting of SAEs and AEs

Making subjective judgments or failing to recognize AEs is only part of the story. There also is the reporting of SAEs, which seems to be a particular challenge.

According to clinical trial regulations for drug trials, investigators are to report SAEs immediately to the sponsor{6}, and in turn, sponsors are to notify FDA and investigators within 15 days of determining that a potential serious risk qualifies for reporting.{7} However, a number of published articles describe continual problems with the reporting of these events.

In particular, SAEs are to be reported to ClinicalTrials.gov by responsible parties—in accordance with the Final Rule of the Food and Drug Administration Amendments Act{8,9}—at various times, both during the trial and after its primary completion date. Too often, however, these results are not published in the medical literature.

Tang, et al. drew this conclusion after conducting a study that evaluated a random convenience sample of 300 trials from ClinicalTrials.gov.{4} All studies were either Phase III or Phase IV trials, and included at least one SAE. Of this sample, 78 (26%) lacked a corresponding publication, and 20 (7%) produced an article that did not match the ClinicalTrials.gov record. Of the 202 remaining studies, 26 (13%) published articles did not mention SAEs, four (2%) reported no SAEs, and 33 (16%) failed to report the total number of SAEs per treatment group.

Similarly, Hughes, et al. found substantial discrepancies between SAEs in Phase II, III, and IV psychiatric trials reported to the former ClinicalStudyResults.org trial registry and the number of corresponding journal articles about those same studies.{10} In retrieving 244 trial summaries for six antidepressant and antipsychotic drugs, researchers found 142 (58.2%) listed an associated article in the summary bibliography, and 72 (29.5%) listed no publication of any kind. Of 1,608 SAEs in drug-treated participants, 694 (43.2%) did not appear in associated articles, and almost 60% of SAEs counted in articles and 41% in trial summaries had no description. Also, most cases of death (62.3%) and suicide (53.3%) were not reported in articles. The authors concluded that SAE reporting is “incomplete, ambiguous, and inconsistent,” meaning that clinical decisions regarding drug use may be based on heavily truncated evidence.

This is but a small sample of the numerous studies documenting the pattern of ongoing problems with sponsors reporting study results to clinical trial registries{11} and to the literature.{12–15}  Additional research is needed to explore why there is a major gap between the detection of SAEs and AEs, and how to include them in publications with adequate detail.

Regulatory Review

The current regulatory landscape for AE and SAE reporting is a mix of existing and newly minted FDA rules and guidances. For ongoing clinical trials, there have long been clear guidelines for stakeholders to report events, as spelled out in several Good Clinical Practice regulations and FDA guidances. The basic requirements for Investigational New Drug (IND) safety reporting appear in 21 Code of Federal Regulations (CFR) 312.32(c)(1){7} and 312.64(b),{6} and describe sponsor and investigator reporting responsibilities. Guidelines for safety reporting in Investigational Device Exemption studies are found in 21 CFR Subpart G 812.150.{16}


Figure 2: Serious Events Unlikely to be Caused by Study Drug/No Safety Report Required

  • SAEs (i.e., mortality or major morbidity) that were likely to have been manifestations of the underlying disease.
  • SAEs that commonly occurred in the study population independent of drug exposure (i.e., strokes or acute myocardial infarctions in an elderly population).
  • SAEs that were study endpoints.

Source: Safety Reporting Requirements for INDs and BA/BE Studies, 2012


For studies conducted under IND applications, in particular, the FDA guidance on “Safety Reporting Requirements for INDs and [Bioavailability/Bioequivalence] Studies”{1} helps sponsors and investigators comply with requirements. The guidance also points out that sponsors often submit unnecessary reports to FDA, which drains agency resources and does little to aid a better understanding of the safety profile of an investigational therapy.

To address this issue, the guidance details how sponsors can determine if there is a reasonable possibility of linking a serious and unexpected event to the study drug, and if not, there is no need to submit an IND safety report. For example, sponsors have frequently reported SAEs that were likely to have been manifestations of the underlying disease, rather than the study drug (Figure 2 lists additional examples).

Recently, the FDA has been releasing additional rules and guidance designed to maintain safeguards that protect the rights, safety, and welfare of subjects. As for whether these new rules impact required AE and SAE reporting, they appear to be unchanged from current requirements. One proposed rule, released in November 2018,{17} is a provision of the 21st Century Cures Act, and if finalized, would permit an IRB to waive or alter certain informed consent elements when clinical research poses no more than minimal risk to subjects.

At the same time, revisions to the Common Rule (“Federal Policy for the Protection of Human Subjects” from the U.S. Department of Health and Human Services) that went into effect in January 2019 intend to promote uniformity and compliance with human subject protections and create a uniform body of regulations across federal departments.{18} Any confusion between the revised Common Rule and the aforementioned proposed rule that would allow for minimal risk clinical research without informed consent will hopefully be addressed during an ongoing harmonization process. As part of this initiative, the FDA released new guidance in October 2018 on how stakeholders can comply with both.{18}

Ask the Right Questions

Of everyone working in the heavily regulated world of clinical trials, those working at the investigative site have a front row seat to the reporting of AEs, SAEs, suspected adverse reactions, and unexpected AEs. It is the study coordinator and the investigator who hear, first hand, from study subjects about potential events that may or may not need to be reported. However, making this determination is often the result of asking the right questions.

Kaitlyn Roberson, clinical research site supervisor at IACT Health, a clinical research firm, explains, “It’s not enough to ask ‘How are you feeling?’ or ‘What’s going on?’ We have to be trained to ask the right questions, like ‘Have you had any new symptoms since your last visit?’ Using this approach, the patient may volunteer that he or she had a cold two weeks ago or maybe started taking a drug to treat it that could interfere with the study medication. At this point, it’s important to ask the right follow-up questions, such as, ‘Did you stop the study drug when you had your cold, and did you re-start it?’ Without the right questions, patients may forget to tell you these details.”

Similarly, Michael Marotta, manager of clinical monitoring services at IMARC Research, a medical device contract research organization, notes, “How a question is asked goes a long way toward capturing what could be an AE. With device trials, we would want to know if the patient had any unscheduled visits, say to a cardiologist, outside the realm of the study. This can be an important point because follow-up for device trials can occur over a number of years, and the further you get from the date of surgery, the greater time there usually is between follow-up visits.”

Marotta also provides insight into why study sites might miss AEs or draft incomplete AE safety reports. He points to the jam-packed schedules that typify many sites where multiple protocols are being undertaken simultaneously for multiple sponsors, with each having different requirements and timelines.

“You can’t think of a trial as occurring in a bubble, which means that sites can have [conflicting] schedules [and] resourcing constraints, and even how the information is being collected can be confusing. And at a granular level, somebody at the site may not be acutely aware of all the requirements for a study. For example, a coordinator could be conducting five different trials, each with a different reporting requirement, so what is appropriate and accurate in one trial may not be appropriate and accurate for another,” Marotta says.

 A Better AE Landscape

As the clinical trials industry works diligently to improve process flow and study conduct, there are signs that these efforts are spilling over into establishing more specific protocol-driven requirements for AE collection and reporting. According to Stephani Hulec, associate director of clinical monitoring services at IMARC Research, changes are happening.

“I’ve seen a definite shift in the industry in the last few years in terms of AE reporting. It’s shifting in a way that some sponsors are writing protocols more focused on the outcomes and what AE information is relevant to those outcomes. The protocols are more specific as to what AEs they require to be reported. This approach more clearly supports the objectives of the study,” Hulec comments.

Taken together, this change, new and existing regulatory guidelines on event reporting, and better training of site personnel represent critical steps in the industry’s attempt to create a more coherent framework for capturing AE-related information. What should follow are more informed evaluations of the risks and benefits of much-needed therapies and devices.


  1. Guidance for Industry and Investigators: Safety Reporting Requirements for INDs and BA/BE Studies. 2012. U.S. Food and Drug Administration. Center for Drug Evaluation and Research. https://www.fda.gov/downloads/Drugs/Guidances/UCM227351.pdf
  2. Bioresearch Monitoring (BIMO) Fiscal Year 2017 Metrics. U.S. Food and Drug Administration. https://www.fda.gov/downloads/ScienceResearch/SpecialTopics/RunningClinicalTrials/UCM604510.pdf
  3. Jeanfreau R. 2017. Opinion: is bias inherent in the current reporting practices for adverse events? Clin Res 31(1):16–9. https://acrpnet.org/2017/02/01/opinion-is-bias-inherent-in-the-current-reporting-practices-for-adverse-events/
  4. Tang E, Ravaud P, Riveros C, et al. 2015. Comparison of serious adverse events posted at ClinicalTrials.gov and published in corresponding journal articles. BMC Medicine 13:189. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4535304/
  5. 21 Code of Federal Regulations32(a). U.S. Food and Drug Administration. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch.cfm?fr=312.32
  6. 21 Code of Federal Regulations64(b). U.S. Food and Drug Administration. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=312.64
  7. 21 Code of Federal Regulations32(c)(1). U.S. Food and Drug Administration. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch.cfm?fr=312.32
  8. Changes from Current Practice Described in Final Rule. 2016. Final Rule for Clinical Trials Registration and Results Information Submission (42 CFR Part 11). U.S. Food and Drug Administration. https://prsinfo.clinicaltrials.gov/FinalRuleChanges-12Dec2016.pdf
  9. Frequently Asked Questions. 2018. ClinicalTrials.gov. https://clinicaltrials.gov/ct2/manage-recs/faq#fr_7
  10. Hughes S, Cohen D, Jaggi R. 2014. Differences in reporting serious adverse events in industry sponsored clinical trial registries and journal articles on antidepressant and antipsychotic drugs: a cross-sectional study. BMJ Open. https://bmjopen.bmj.com/content/bmjopen/4/7/e005535.full.pdf
  11. Enforcing reporting to ClinicalTrials.gov: what’s at stake for research transparency. 2017. Ampersand (The PRIM&R Blog). https://blog.primr.org/enforcing-reporting-to-clinicaltrials-gov/
  12. Golder S, Loke YK, Wright K, et al. 2016. Reporting of adverse events in published and unpublished studies of health care interventions: a systematic review. PLoS Med 13(9). https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002127
  13. Pitrou I. Boutron I, Ahmad N, et al. 2009. Reporting of safety results in published reports of randomized controlled trials. Arch Intern Med 169(19):1756–61. https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/724471
  14. Prayle AP, Hurley MN, Smyth AR. 2012. Compliance with mandatory reporting of clinical trial results on ClinicalTrials.gov: cross sectional study. BMJ https://www.bmj.com/content/bmj/344/bmj.d7373.full.pdf
  15. Anderson ML, Chiswell K, Peterson ED, et al. 2012. Compliance with results reporting at ClinicalTrials.gov. NEJM 372(11):1031–9. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4508873/pdf/nihms677071.pdf
  16. 21 Code of Federal Regulations150. U.S. Food and Drug Administration. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=812.150
  17. Institutional Review Board Waiver or Alteration of Informed Consent for Minimal Risk Clinical Investigations. 2018. Proposed Rule. U.S. Food and Drug Administration. https://www.federalregister.gov/documents/2018/11/15/2018-24822/institutional-review-board-waiver-or-alteration-of-informed-consent-for-minimal-risk-clinical
  18. Impact of Certain Provisions of the Revised Common Rule on FDA-Regulated Clinical Investigation. 2018. Guidance for Sponsors, Investigators, and Institutional Review Boards. U.S. Food and Drug Administration. https://www.fda.gov/downloads/RegulatoryInformation/Guidances/UCM623211.pdf

Ann Neuer, MBA, (aneuer@cinci.rr.com) is President of Medical deScriptions, LLC. Her earlier articles for Clinical Researcher, “Patient Engagement Goes Mobile” and “For Principal Investigators, the One-and-Done Syndrome Persists,” appeared in January 2018 and September 2018, respectively.