Adverse Event

In clinical research and pharmacovigilance, an adverse event (AE) is any untoward medical occurrence that presents during treatment with a pharmaceutical product but does not necessarily have a causal relationship with that treatment. An adverse drug reaction (ADR) is the subset of adverse events for which a causal relationship with the drug is at least reasonably possible; the ICH defines it as a noxious and unintended response to a medicine used in the normal way for prevention, diagnosis or treatment. Adverse events are recorded systematically in clinical trials, graded for severity, assessed for causality and classified as serious or non-serious, and after approval they are collected through pharmacovigilance systems such as the FDA Adverse Event Reporting System (FAERS).

Trials record every adverse event in every arm, including the placebo group, so that rates can be compared and drug-related harms separated from background illness. Once a drug is approved, rare or delayed harms are tracked through spontaneous reporting systems such as FAERS, which accepts reports from manufacturers, health professionals and the public. These systems detect safety signals but are known to capture only a small fraction of actual reactions.

The distinction between an adverse event and an adverse drug reaction is central to reading safety data. An adverse event is simply something harmful that happened during treatment; it may be caused by the drug, by the underlying disease or by chance. Only after assessing timing, the pattern of illness, test results and sometimes rechallenge can investigators judge whether a reaction is likely drug-related. Comparing event rates between drug and placebo groups in randomized trials is the most reliable way to make that judgment.

Clinical trials are usually too small and too short to detect rare harms, so monitoring continues after approval. FAERS, started in 1969, collects reports from manufacturers, health professionals and the public, and data-mining methods are used to look for statistical signals linking a drug to an event. StatPearls notes that over 1.25 million serious adverse events and nearly 175,000 deaths were reported to FAERS in 2022. A signal prompts further investigation; on its own it does not prove that the drug caused the reports.

Spontaneous reporting also misses most reactions. A systematic review of 37 studies estimated a median under-reporting rate of 94%, and under-reporting remained high even for serious reactions. This matters for unapproved research peptides, which have no structured trial safety data and no manufacturer obligated to collect and report adverse events. A claim that such a peptide has “no reported side effects” usually reflects the absence of a reporting system rather than evidence of safety.

“Every adverse event in a trial was caused by the drug”

An adverse event is recorded regardless of cause. Many occur at similar rates in the placebo group, reflecting the underlying illness or everyday medical events. Causality is assessed separately, and the comparison with placebo is what shows whether a drug increases the risk of a particular event.

“Adverse event, side effect and adverse drug reaction mean the same thing”

An adverse event makes no claim about cause. An adverse drug reaction is an event judged to be related to the drug. “Side effect” is an everyday term that usually refers to known effects of a drug, which may be unwanted but are not always harmful. Mixing these terms can overstate or understate risk.

“If FAERS has many reports, the drug must be dangerous”

FAERS counts are influenced by how widely a drug is used, media attention and reporting habits, and individual reports are not verified for cause. Data-mining can flag a possible association, but establishing that a drug causes a harm requires further study, such as controlled trials or well-designed observational research.

What is a serious adverse event?

A serious adverse event is one with a grave outcome. Under the ICH definition, and US FDA rules for investigational drugs, that means an event that results in death, is life-threatening, requires hospital admission or prolongs a stay, causes lasting or significant disability, or is a birth defect, plus other events doctors judge to be medically important. In clinical trials, serious events must be reported to sponsors and regulators on a faster timeline than other adverse events. “Serious” describes the outcome, which is different from “severe”, a measure of intensity.

What are examples of adverse events?

Anything that happens to a person's health during treatment or a trial counts, whether or not the treatment caused it: symptoms such as nausea, headache or redness at an injection site; abnormal test results such as raised liver enzymes; new illnesses; and injuries such as a fall or a broken bone. Each is recorded first and assessed for a possible link to the treatment afterwards.

How are adverse events graded?

Severity is often graded on a five-point scale. In the US National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE), grade 1 is mild, grade 2 moderate, grade 3 severe or medically significant but not immediately life-threatening, grade 4 life-threatening and grade 5 death related to the event. Grade describes intensity, so a grade 3 event is not automatically a serious adverse event, and a serious event is not always high grade.

What is FAERS?

FAERS is the US FDA Adverse Event Reporting System, a database of adverse event and medication error reports. Reports come from manufacturers, health professionals and the public. Researchers and regulators apply data-mining methods, such as the proportional reporting ratio and reporting odds ratio, to look for signals of possible drug-related harm.

How do doctors decide whether a drug caused a reaction?

Causality assessment considers the timing of the event relative to the drug, whether the pattern of illness fits known effects, laboratory or other investigation results, whether the event improves when the drug is stopped and, occasionally, whether it recurs on re-exposure. Structured tools help standardize this judgment, but certainty is often limited for individual cases.