Adverse Event (AE)

In clinical-trial administration, an adverse event (AE) is any untoward medical occurrence in a subject who has been administered a pharmaceutical or investigational product — whether or not it is considered related to that product. This is the causality-neutral definition set out in ICH E2A, “Clinical Safety Data Management: Definitions and Standards for Expedited Reporting” (finalized 1994), the guideline that underpins safety reporting across all International Council for Harmonisation (ICH) regions.

What makes something an AE

Under ICH E2A, an occurrence qualifies as an AE if it is:

• Medical — a sign, symptom, diagnosis, or abnormal lab/vital-sign finding, not an administrative or protocol-deviation event on its own.
• Untoward — unfavorable or unintended from the subject’s perspective.
• Temporally associated with product administration — it occurred after the subject received (or was exposed to) the investigational product, regardless of whether the product caused it.

The causality-neutral scope is deliberate: a headache, a transient abnormal lab value, or an unrelated fall during a study visit all qualify as AEs, because the definition captures anything unfavorable and unintended that coincides with product administration — not only events later confirmed to be caused by it. Causality is assessed and recorded separately, not screened out at the point of capture.

AE is the broadest term in a cascading set of definitions

ICH E2A defines AE as the base of a narrower, cascading set of terms. Each subsequent term adds a further qualifying condition on top of the AE definition:

• Adverse event (AE) — any untoward medical occurrence; causality not yet assessed or not required.
• Adverse (drug) reaction — an AE for which there is at least a reasonable possibility of a causal relationship to the product.
• Serious adverse event (SAE) — an AE that meets one of a defined set of seriousness criteria (results in death, is life-threatening, requires or prolongs inpatient hospitalization, results in persistent or significant disability/incapacity, is a congenital anomaly/birth defect, or is otherwise judged medically important) — independent of causality or of how severe the event feels to the subject.
• Unexpected — the event’s nature or severity is not consistent with the current Investigator’s Brochure or approved product labeling.
• SUSAR (Suspected Unexpected Serious Adverse Reaction) — an event that is simultaneously serious, unexpected, and has a suspected causal relationship to the product. SUSARs are the subset that triggers expedited regulatory reporting on a strict clock.

A common misreading is to treat “serious” and “severe” as synonyms. They are not: seriousness is a fixed, defined regulatory category (see above); severity (mild/moderate/severe) is a clinical intensity grading, typically captured separately using a scale such as CTCAE. A severe headache is not automatically an SAE; a mild-intensity event that causes hospitalization is.

Worked examples

• AE, not serious, not a reaction: a trial subject reports mild nausea two days after their first dose. It is recorded as an AE; causality is assessed as “unlikely related” based on the timing and known product profile. It does not meet any seriousness criterion, so it is not an SAE and does not trigger expedited reporting — it is captured in routine case report form data and reviewed in aggregate.
• SUSAR: a subject is hospitalized for acute liver injury four weeks into dosing. The investigator assesses the event as “possibly related” to the investigational product, and liver injury is not listed in the current Investigator’s Brochure as an expected effect. This event is serious (hospitalization), unexpected (not in the IB), and suspected of a causal relationship — it is a SUSAR, and it starts the sponsor’s expedited-reporting clock to regulators.

Counter-example

A subject misses a scheduled study visit because of a scheduling conflict unrelated to their health. This is a protocol deviation, not an AE — no untoward medical occurrence took place. Protocol deviations and AEs are tracked through separate processes and should not be conflated in site-level reporting.

Why the distinction matters operationally

Whether an event is classified as an AE, an SAE, or specifically a SUSAR determines what reporting clock applies and to whom. Under US regulation, for example, an investigator must report any SAE to the sponsor immediately regardless of the causality opinion (21 CFR 312.64(b)), while a sponsor’s expedited reporting obligation to the FDA for a fatal or life-threatening SUSAR runs on a 7-calendar-day clock, with complete follow-up due within an additional 8 days, and other SUSARs on a 15-calendar-day clock (21 CFR 312.32). The European Union’s Clinical Trials Regulation ((EU) No 536/2014) applies the same 7-day/15-day structure for SUSAR reporting to EudraVigilance. None of these clocks are triggered by an AE alone — only by the narrower SUSAR category, which is exactly why getting the AE → SAE → SUSAR cascade right at the point of data capture matters.

This entry defines the term itself. For the full mechanics of who reports what, on which timeline, to which body — including the role of periodic aggregate safety reports (DSUR, PBRER) and how pharmacovigilance relates to a trial’s independent safety-oversight body — see the fuller guide: Pharmacovigilance in Clinical Research: AE, SAE, and SUSAR Reporting.

References

• ICH E2A, “Clinical Safety Data Management: Definitions and Standards for Expedited Reporting” (1994) — ich.org efficacy guidelines
• 21 CFR 312.32 and 21 CFR 312.64(b) — US IND safety reporting requirements
• Regulation (EU) No 536/2014 — EU Clinical Trials Regulation, SUSAR reporting to EudraVigilance