Pharmacovigilance in Clinical Trials: The Classify-Then-Clock Workflow for AE, SAE, and SUSAR Reporting

At a glance

• Trial pharmacovigilance is not a “why safety matters” topic. It is a triage workflow: every event is classified by seriousness, causality, and expectedness, and that classification is what fires a reporting clock.
• The fork that matters: an adverse event (AE) that is serious, unexpected, and has a reasonable causal link to the investigational product becomes a SUSAR, and only SUSARs drive expedited sponsor reporting to authorities.
• Two clocks govern expedited reporting. Fatal or life-threatening unexpected reactions go in 7 calendar days; all other serious unexpected reactions go in 15 calendar days. Both start from the sponsor’s first knowledge.
• The duty split is fixed: investigators report serious adverse events immediately to the sponsor with a causality assessment; the sponsor makes the expectedness call and owns the expedited clock to regulators.
• Where teams lose: a late causality call, a “non-serious” event upgraded to serious after the fact (which retroactively starts the clock), and SAE-to-database reconciliation that breaks before lock.

Most pharmacovigilance (PV) explainers tell you that PV is the science of monitoring drug safety and that it is important. That framing is useless at 4:55 PM when a site emails you a hospitalization and you have to decide, today, whether a clock has already started. This guide treats trial PV the way it actually runs on a study: as a classify-then-clock workflow. You collect an event, you classify it on three axes, and the classification you land on is the thing that tells you who must report what, to whom, and within how many days. Get the classification right and the clocks follow. Get it wrong and you miss an expedited deadline.

Trial pharmacovigilance vs post-market PV: same science, different clock

Post-marketing PV runs on spontaneous reports, periodic safety update reports, and signal detection across an approved label. Trial PV runs on a defined cohort, a protocol, an Investigator’s Brochure (IB), and a set of expedited reporting obligations imposed on the sponsor and investigator while the product is still investigational. The science of causality is shared, but the obligations and the clock are not. This article is about the trial side.

The unit of reference in a trial is the IB (or, where applicable, current product information). ICH E6(R3) §3.13 states that the Investigator’s Brochure, or the current scientific information where applicable, forms the basis of safety assessment and reporting for the clinical trial. That matters because “expected” and “unexpected” are defined against that document, not against general medical knowledge.

The PV workflow end to end

The operational sequence is: collect, assess seriousness, assess causality, assess expectedness, code and report, reconcile, then trend and signal. The first four steps are the classification. The classification decides everything downstream.

1. Collect. Site staff capture the event. ICH E6(R3) §2.7.2(a) requires that adverse events and abnormal test results needed for safety evaluations be reported to the sponsor per the protocol’s requirements and timelines.
2. Assess seriousness. Does the event meet a seriousness criterion (below)? Seriousness is a fixed regulatory definition, not a severity judgment. A mild rash can be non-serious; a “moderate” event that causes hospitalization is serious.
3. Assess causality. Is there a reasonable possibility the product caused it? This is the gate between an adverse event and a suspected adverse reaction.
4. Assess expectedness. Is the event already described, at this specificity and severity, in the IB?
5. Code and report. MedDRA coding, narrative, and the expedited report itself if the event qualifies.
6. Reconcile. Match every SAE in the safety database against the clinical (EDC) database before lock.
7. Trend and signal. Aggregate review of emerging safety information.

The triage that drives everything: AE vs SAE vs SUSAR

ICH E2A draws the definitions the whole workflow runs on. An adverse event is any untoward medical occurrence in a subject administered a pharmaceutical product, which does not necessarily have a causal relationship with the treatment (ICH E2A, Definitions). An event becomes serious when, per ICH E2A’s seriousness definition, it results in death, is life-threatening, requires or prolongs hospitalization, causes persistent or significant disability or incapacity, or is a congenital anomaly or birth defect; important medical events may also be treated as serious on medical judgment. Note that ICH E2A is explicit that “life-threatening” means the patient was at risk of death at the time of the event, not that the event hypothetically could have become fatal if more severe.

Causality is the next axis. ICH E2A states that a causality assessment is required for clinical investigation cases, and that any case judged by either the reporting health care professional or the sponsor to have a reasonable suspected causal relationship qualifies as an adverse drug reaction. The US framing is close but not identical: FDA’s IND safety rule defines a suspected adverse reaction as any adverse event for which there is a reasonable possibility that the drug caused the event, and clarifies that “reasonable possibility” means there is evidence to suggest a causal relationship (21 CFR 312.32(a)). There is a real difference worth flagging: under 21 CFR 312.32 the sponsor makes the causality judgment for IND reporting purposes, whereas ICH E2A recommends the judgment rest on either the investigator’s or the sponsor’s opinion. Both regimes are in scope here, and they do not phrase the causality owner identically; state the tension, do not paper over it.

Expectedness is the third axis. ICH E2A defines an unexpected adverse drug reaction as one whose nature or severity is not consistent with the applicable product information, and adds that an event more specific or more severe than described in the Investigator’s Brochure is “unexpected.” FDA’s rule aligns: under 21 CFR 312.32(a) an event is “unexpected” if it is not listed in the IB, or not listed at the observed specificity or severity.

A SUSAR is the intersection of all three: serious, unexpected, and a suspected (reasonable-possibility) causal relationship. ICH E6(R3) §3.13.2(b) requires the sponsor, in accordance with applicable regulatory requirements and with ICH E2A, to expedite reporting to regulatory authorities of all suspected, unexpected, and serious adverse reactions, that is, SUSARs. The SUSAR is the only output that triggers the expedited clock.

Classification decision table

Question	If yes	If no	Who acts
Untoward medical occurrence in a subject on product?	It is an AE; record it	Not an AE	Site / investigator
Meets a seriousness criterion (death, life-threatening, hospitalization, disability, congenital anomaly, or important medical event)?	It is an SAE; report immediately to sponsor	Stays a non-serious AE; routine protocol handling	Investigator decides; reports to sponsor
Reasonable possibility the product caused it?	It is a (suspected) adverse reaction	Serious but unrelated; not a SUSAR	Investigator assesses; sponsor judges for IND reporting
Unexpected vs the IB / reference safety information?	If also serious + related, it is a SUSAR	Serious, related, but expected; not expedited as a SUSAR	Sponsor assesses expectedness

ICH E2A is blunt about the negative cases: expedited reporting of reactions that are serious but expected is ordinarily inappropriate, and serious events considered not related to study product are not subject to expedited reporting whether expected or not.

The reporting clocks: 7-day vs 15-day, and who owes them

Once an event is a SUSAR, two clocks apply, and they are owned by the sponsor.

ICH E2A sets the expedited timeframes. For fatal or life-threatening unexpected ADRs, regulatory agencies must be notified as soon as possible but no later than 7 calendar days after first knowledge by the sponsor that the case qualifies, followed by a complete report within 8 additional calendar days. For all other serious, unexpected ADRs, the report must be filed as soon as possible but no later than 15 calendar days after first knowledge by the sponsor.

FDA’s IND rule imposes the same two clocks on US trials. Under 21 CFR 312.32(c), the sponsor must notify FDA and all participating investigators of a serious and unexpected suspected adverse reaction no later than 15 calendar days after the sponsor determines the information qualifies. For an unexpected fatal or life-threatening suspected adverse reaction, the report to FDA is due no later than 7 calendar days after the sponsor’s initial receipt of the information; FDA notes that if the 7-day report is complete, no separate 15-day submission is required.

EU CT-3 sets matching timeframes for EU trials and adds the recipients. The sponsor must ensure that fatal or life-threatening SUSARs are recorded and reported as soon as possible to the competent authorities in all Member States concerned and to the Ethics Committee, in any case no later than 7 days after the sponsor’s knowledge, with follow-up within an additional 8 days. All other SUSARs must be reported within a maximum of 15 days of first knowledge by the sponsor. EU CT-3 also specifies the channel: SUSARs are reported to the national competent authority, increasingly through the EudraVigilance Clinical Trials Module (EVCTM), and the sponsor must also inform all investigators.

Expedited reporting timeline table

Event type	Clock	Reporter	Recipient	Source
Fatal / life-threatening, unexpected, related (SUSAR)	7 calendar days + 8 days follow-up	Sponsor	Regulatory authority(ies)	ICH E2A
Fatal / life-threatening, unexpected suspected adverse reaction	7 calendar days (then complete report)	Sponsor	FDA + all participating investigators	21 CFR 312.32(c)
Fatal / life-threatening SUSAR	7 days + 8 days follow-up	Sponsor	Competent authorities (all concerned MS) + Ethics Committee	EU CT-3
Other serious, unexpected, related (SUSAR)	15 calendar days	Sponsor	Regulatory authority(ies)	ICH E2A
Other serious and unexpected suspected adverse reaction	15 calendar days	Sponsor	FDA + all participating investigators	21 CFR 312.32(c)
Other SUSAR	15 days	Sponsor	Competent authorities concerned + Ethics Committee (via EVCTM)	EU CT-3
Any SAE (upstream of the above)	Immediately	Investigator	Sponsor	ICH E6(R3) §2.7.2(b)

The sponsor-vs-investigator duty split

This is where most misclassification damage compounds, because the two parties own different parts of the workflow.

The investigator’s job is to surface and characterize. ICH E6(R3) §2.7.2(b) requires that all serious adverse events be reported immediately to the sponsor after the investigator reasonably becomes aware of them, and that the investigator include an assessment of causality. EU CT-3 carries the same duty from Directive 2001/20/EC Article 16(1): the investigator shall report all serious adverse events immediately to the sponsor, except those the protocol or IB identifies as not requiring immediate reporting. Under FDA’s regime, 21 CFR 312.64 requires investigators to provide a causality assessment for each serious adverse event reported to the sponsor.

The sponsor’s job is to aggregate, judge expectedness, and run the expedited clock. ICH E6(R3) §3.13.1 makes the sponsor responsible for aggregating and reviewing relevant safety information in a timely manner, and §3.13.2(c) requires that safety reporting to authorities be undertaken by assessing expectedness against the applicable product information, such as the reference safety information in the IB. FDA reinforces the division: the sponsor, not the investigator, makes the causality judgment for IND safety reporting purposes under 21 CFR 312.32, even though the investigator supplies a causality assessment with each SAE. The practical rule: the investigator owns “is this serious, and what do I think caused it,” reported immediately; the sponsor owns “is this unexpected, does it qualify as a SUSAR, and which clock applies.”

PV across the phases

The classification logic does not change across Phase I to Phase IV, but the reference frame does. In first-in-human Phase I, the IB is thin, so a far larger share of serious events read as unexpected, and more events become SUSARs. As the IB fills out across later phases, more events become “expected” against an evolving reference safety information document, and EU CT-3 is explicit that for SUSAR reporting the version of the reference safety information at the moment the SUSAR occurred applies, so a change to the IB changes which events must be reported as SUSARs going forward. The clocks and the duty split are identical at every phase; what shifts is how many events clear the expectedness gate.

Where teams get it wrong

Late causality calls. Because a SUSAR requires a “reasonable possibility” causal link, an event with no documented causality assessment cannot be correctly classified. ICH E2A requires a causality assessment for clinical investigation cases, and FDA’s rule under 21 CFR 312.64 obliges investigators to supply one for each SAE. When the site leaves causality blank, the sponsor must still evaluate, and the clock does not wait for the missing field.

Missed SUSAR upgrades. An event first logged as non-serious, or serious-but-expected, can be reclassified later as the picture develops. The clock under ICH E2A and 21 CFR 312.32(c) runs from the sponsor’s first knowledge that the case qualifies for expedited reporting, so a late upgrade can mean the deadline effectively started days ago. Treat every reclassification as a potential clock event, not a paperwork tidy-up.

Blinding handled wrong for reporting. Reporting a SUSAR generally requires knowing the treatment. ICH E2A recommends that, when a serious reaction is judged reportable on an expedited basis, the blind be broken only for that specific patient by the sponsor. ICH E6(R3) §2.11 requires the investigator to document and explain to the sponsor any premature unblinding, including unblinding due to an SAE, and §3.15.2 expects the sponsor to maintain a mechanism that protects the blinding of the trial where a participant’s assignment is unblinded for safety reporting. Unblind the one patient for the report; do not compromise the study blind.

Broken SAE reconciliation. Every SAE in the safety database must match the clinical database before lock. This is not itself a single-section regulatory mandate, but it is the operational consequence of the investigator’s duty to ensure the accuracy and completeness of reported data; ICH E6(R3) §2.12.5 requires the investigator to ensure the accuracy, completeness, legibility, and timeliness of data reported to the sponsor, including SAE reports. A discrepancy found after lock is far more expensive than one found in routine reconciliation.

Site PV checklist for coordinators and investigators

• On every adverse event, run the three axes in order: serious? related? unexpected vs the current IB? Do not skip to “unexpected” before settling seriousness.
• Report every SAE to the sponsor immediately, with your causality assessment attached (ICH E6(R3) §2.7.2(b); EU CT-3 Article 16(1)).
• Never downgrade a hospitalization or other seriousness criterion on the basis of mild severity; seriousness is a fixed criterion, not a severity rating.
• Flag any event that you suspect may be reclassified, and tell the sponsor in writing, because the expedited clock keys off the sponsor’s first knowledge that the case qualifies.
• For emergency unblinding tied to an SAE, follow the protocol’s procedure and document and explain the unblinding to the sponsor (ICH E6(R3) §2.11).
• Keep your SAE source records reconciliation-ready so the safety and clinical databases match before lock.
• Confirm the protocol’s list of SAEs exempt from immediate reporting (e.g., protocol-defined endpoints) before assuming an event is expedited or not.

Trial pharmacovigilance is not about whether safety matters. It is about classifying each event correctly and respecting the clock that classification starts. The regulations here describe what sponsors and investigators are required to do; running the workflow well is what lets a team meet those obligations, and the sponsor remains responsible for the safety evaluation throughout. Related reading on this site covers IND safety reporting and SUSAR mechanics in depth, protocol deviation triage and timelines, and risk-based quality management, each of which intersects this workflow.

Sources

• ICH E6(R3) Good Clinical Practice — ICH, version r3 — https://www.ich.org/page/efficacy-guidelines
• ICH E2A Clinical Safety Data Management: Definitions and Standards for Expedited Reporting — ICH, version r1 — https://database.ich.org/sites/default/files/E2A_Guideline.pdf
• FDA Guidance for Industry and Investigators: Safety Reporting Requirements for INDs and BA/BE Studies (21 CFR 312.32) — FDA, version 2012 — https://www.fda.gov/regulatory-information/search-fda-guidance-documents/safety-reporting-requirements-inds-and-babe-studies
• EudraLex Vol 10 CT-3: Detailed guidance on adverse event/reaction reports from clinical trials (2011/C 172/01) — European Commission, version 2011 — https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52011XC0611(01)