· PHARMACOVIGILANCE / GVP

Pharmacovigilance Signal Management

Pharmacovigilance is the science of monitoring the safety of a medicine after it is approved, when it meets a real population far larger and more varied than any trial. Signal management is the disciplined process at its centre: individual adverse-event reports are gathered, a possible new or changed safety concern is recognised as a signal, that signal is validated and assessed against the whole evidence base, and — if it holds — the medicine’s benefit-risk balance and its product information are changed. It is the loop that keeps a marketing authorisation honest for the decades a medicine is on the market.

What an explainer is not

A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.

[ POSITION IN THE FRAMEWORK ]

7 DIMENSIONS · 22 LINKS

Signal management is the pharmacovigilance loop that turns individual case reports into a change to a medicine's benefit-risk balance — worked across the GVP and clinical disciplines and run inside the safety database below.

06 · QUALITY MATURITY — PHARMACOVIGILANCE SIGNAL MANAGEMENT, REACTIVE TO ADAPTIVE

L1
Reactive

Cases are processed to meet reporting clocks but rarely mined; a signal is recognised only when a regulator or the literature raises it first.

L2
Defined

A signal-management SOP and detection method exist, but validation and assessment are inconsistent and prioritisation is ad hoc.

L3
Controlled

Detection, validation, assessment, and action follow the defined lifecycle; non-validated and validated signals are distinguished and prioritised by impact.

L4
Predictive

Statistical disproportionality and case review run together so a validated signal is assessed and acted on before harm accumulates.

L5
Adaptive

Signal knowledge compounds across products; benefit-risk conclusions feed labelling, risk minimisation, and study commitments as a managed loop.

SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →

07 · REGULATORY & EVIDENCE

GOVERNING STANDARDS · 4

Derived from the 4 standards SPEQ maps to this subject, across 3 regulatory bodies: EMA, FDA, ICH.

RECORDS & OBJECTIVE EVIDENCE

  • Individual case safety reports (ICSRs) coded in MedDRA and exchanged in ICH E2B(R3)
  • A signal-detection and validation log distinguishing non-validated from validated signals
  • Signal assessment reports weighing the benefit-risk balance
  • The Pharmacovigilance System Master File (PSMF) describing the system
  • Records of resulting actions — label changes, risk minimisation, safety communications

COMMON INSPECTION FINDINGS

  • Signal detection performed but validation and assessment not documented
  • Validated signals not prioritised or acted on within a defined timeframe
  • Expedited case-reporting deadlines missed
  • Duplicate cases inflating or masking a signal
  • PSMF not reflecting the pharmacovigilance system as actually run
EVERY CHIP IS A DOOR · WALK THE FRAMEWORK FROM ANY SUBJECTHow SPEQ maps the framework →

What a safety signal is — and what it is not

A signal is information suggesting a new potentially causal association, or a new aspect of a known association, between a medicine and an adverse event, that is judged to warrant further investigation. The key words are *potentially causal* and *warrant further investigation*: a signal is a hypothesis, not a conclusion. A single serious report can be a signal; so can a disproportionate pattern across thousands of reports. Neither is proof of harm on its own.

The distinction that trips practitioners is signal versus adverse reaction. An individual adverse reaction is one patient’s event; a signal is a pattern across cases that suggests the medicine may be doing something not yet reflected in its label. Treating every report as a signal drowns the system; treating no report as a signal until harm is undeniable defeats the point. The whole discipline is the structured judgement in between.

The raw material: ICSRs and the reporting system

The unit of pharmacovigilance data is the individual case safety report (ICSR) — a structured record of a suspected adverse reaction in one patient. ICSRs arrive from clinicians, patients, marketing-authorisation holders, and the literature, and they are exchanged between organisations and regulators in a defined electronic format so that a case reported anywhere can be pooled everywhere; ICH E2B(R3) is the standard that makes that exchange machine-readable. ICH E2A established the shared clinical-safety vocabulary — what counts as serious, what counts as expedited — that decides how fast a case must move.

Post-marketing reporting obligations sit in law on the authorisation holder: in the United States, 21 CFR 314.80 governs post-marketing reporting of adverse drug experiences; in the European Union, the GVP framework sets the equivalent expectations. The quality of every downstream signal depends on the completeness and timeliness of these reports, which is why case intake, medical coding (MedDRA), and duplicate detection are treated as controlled processes, not clerical ones.

The signal management lifecycle

The EU good-pharmacovigilance-practice guidance frames signal management as a defined lifecycle, and it is the clearest model of the work. Detection surfaces a **non-validated signal** — a candidate raised from a striking individual case, a series, or statistical disproportionality in a large database. **Validation** is the gate that decides whether there is enough evidence to justify further analysis: is the association plausible, is it already known and adequately labelled, are the data real or an artefact of reporting? Only a signal that passes validation becomes a **validated signal** that moves into assessment — the distinction matters because acting on every raw candidate would swamp the system while dismissing candidates before validation would miss real harms. Validated signals are then prioritised by seriousness and public-health impact.

Assessment then weighs the validated signal against the entire evidence base — clinical trials, epidemiology, mechanism, and the rest of the safety record — to decide whether the benefit-risk balance has actually shifted. The output is a recommendation: no action, routine monitoring, a label change, a new contraindication or warning, a risk-minimisation measure, or in the extreme a suspension. The loop closes only when the action is implemented and its effect is itself monitored.

Accountability: the QPPV, the PSMF, and the benefit-risk clock

Pharmacovigilance is a system with a named owner. In the EU framework a single Qualified Person Responsible for Pharmacovigilance (the QPPV) is personally accountable for the authorisation holder’s entire pharmacovigilance system, and that system is described in a controlled document — the Pharmacovigilance System Master File (PSMF) — that an inspector can pick up and use to reconstruct how safety is actually run. Neither is a formality: an inspection failure here is a finding against the whole safety system, not one case.

What makes pharmacovigilance unforgiving is the clock. Serious cases carry expedited reporting deadlines, periodic safety reports are due on a schedule, and a validated signal that is not acted on is a live patient-safety and regulatory exposure. The consequence layer is real and public — label changes, direct healthcare-professional communications, and withdrawals all begin as a signal that someone detected and refused to normalise.

FREQUENTLY ASKED

What is a safety signal in pharmacovigilance?

A signal is information suggesting a new potentially causal association, or a new aspect of a known association, between a medicine and an adverse event, that warrants further investigation. It is a hypothesis to be validated and assessed — not, on its own, proof of harm.

What is an ICSR?

An individual case safety report (ICSR) is the structured record of a suspected adverse reaction in one patient. ICSRs are the raw material of pharmacovigilance; they are exchanged between organisations and regulators in the ICH E2B(R3) electronic format so cases can be pooled across sources.

What is the difference between a signal and an adverse reaction?

An adverse reaction is one patient’s event. A signal is a pattern across cases suggesting the medicine may be doing something not yet reflected in its approved product information. Signal management is the process of telling the two apart and acting proportionately.

Who is responsible for a pharmacovigilance system?

In the EU framework, a single Qualified Person Responsible for Pharmacovigilance (QPPV) is personally accountable for the marketing-authorisation holder’s pharmacovigilance system, which is documented in the Pharmacovigilance System Master File (PSMF). In the US, post-marketing reporting obligations sit on the holder under 21 CFR 314.80.

What is the difference between a non-validated and a validated signal?

A non-validated signal is a candidate just surfaced by detection — a striking case, a series, or a statistical disproportionality — that has not yet been checked. Validation is the gate: it decides whether there is enough evidence to justify further analysis (is the association plausible, already known and labelled, or an artefact of reporting?). Only a signal that passes validation becomes a validated signal that moves into full assessment. The distinction keeps the system from either acting on every raw candidate or dismissing candidates before they are properly examined.

PROFESSIONAL · INSPECTION PLAYBOOK · SPEQ SYNTHESIS

The inspection-readiness playbook for this topic

CHECKING ACCESS

Checking your Professional access…