Signal Tracking and Documentation

Signal management is a scientific decision process carried out over time. A safety hypothesis may begin as a single case, a statistical alert, a literature finding or a regulatory communication; it may then be validated, prioritised, assessed, acted upon or closed. Documentation is what allows another qualified reviewer to reconstruct that journey and understand not only what was decided but why.

Purpose and Regulatory Framework

EU signal management is governed by Directive 2001/83/EC, Regulation (EC) No 726/2004, Commission Implementing Regulation (EU) No 520/2012 as amended, and GVP Module IX — Signal management. The wider quality-system and inspection framework is described in GVP Module I and GVP Module III.

The regulatory objective is not a particular spreadsheet, signal database schema or template pack. It is an effective process whose important decisions are documented, traceable and capable of being reviewed. GVP Module IX requires signal-management activities and decisions to be appropriately documented. The specific form of the documentation may differ between organisations.

As of September 2026, GVP Module IX Rev. 1 remains published while the underlying legal framework has been amended by Commission Implementing Regulation (EU) 2025/1466. EMA has stated that Module IX will be revised for alignment. Documentation systems should therefore be designed to accommodate regulatory change rather than encode obsolete procedural assumptions permanently.

What Traceability Means in Signal Management

Traceability is the ability to follow a signal record in both directions.

Starting with the original observation, a reviewer should be able to identify:

• where the information came from;
• why it entered the signal process;
• how validation and prioritisation decisions were made;
• what evidence was assessed;
• who had decision authority;
• what conclusion was reached;
• which actions followed; and
• whether those actions were completed.

Starting with a downstream action, such as an RMP update or product-information change, the reviewer should also be able to trace backwards to the signal assessment and evidence that justified it.

This bidirectional relationship is more important than the visual appearance of the tracker.

The Signal Record as a Connected Evidence Chain

A mature signal record is not one document. It is a connected set of controlled records that together answer the scientific and governance questions arising during the lifecycle.

A useful conceptual sequence is:

source information → signal entry → validation → prioritisation → assessment → decision → action → implementation → closure or monitoring.

Each stage should add the information needed for the next stage without requiring the reader to rediscover the reasoning from scratch.

Core Information That Should Remain Traceable

The exact data model is organisation-specific, but the following information is commonly necessary for reliable reconstruction:

• unique signal identifier;
• medicinal product or active substance;
• event or safety issue being evaluated;
• source and date of detection;
• status within the signal lifecycle;
• important review and decision dates;
• evidence sources considered;
• validation and prioritisation rationale;
• assessment document and version;
• decision authority or governance forum where applicable;
• agreed actions, owners and due dates;
• regulatory interactions;
• downstream RMP, PSUR/PBRER or product-information consequences;
• closure rationale; and
• links to supporting records.

Not every signal needs the same amount of documentation. The record should be proportionate to the significance and complexity of the issue while remaining sufficient to reconstruct the decision.

Documentation Is Not the Same as Data Accumulation

More documentation does not automatically create better traceability. A signal file can contain hundreds of attachments yet still fail to explain why the signal was validated or why no regulatory action was taken.

High-quality documentation distinguishes between:

1. source evidence — cases, literature, studies, analytical outputs;
2. scientific interpretation — what the evidence means and its limitations;
3. decision — the conclusion reached;
4. governance — who had authority and how important disagreements were resolved; and
5. implementation evidence — what happened after the decision.

The record should make these relationships explicit.

Documentation Across the Signal Lifecycle

The content required at each stage follows from the question being answered at that stage.

Detection and intake

Detection records should establish the origin of the observation and why it entered the signal process. Depending on the source, relevant evidence may include a case-series trigger, literature citation, study result, regulatory communication or analytical output. The record should preserve enough information to understand the detection context without implying that a statistical alert is itself a validated signal.

Validation

Validation documentation should explain whether the available information contains sufficient evidence to justify further analysis. A bare status such as “validated” or “not validated” is insufficient if the rationale cannot be reconstructed.

The record should ordinarily identify:

• the evidence reviewed;
• the question considered;
• important supporting and contradictory information;
• the validation conclusion; and
• any follow-up needed before the next step.

The companion article [[signal-validation]] addresses this decision in greater depth.

Prioritisation

Prioritisation records should show why one signal received greater urgency or resource than another. Factors may include seriousness, severity, public-health impact, exposure, vulnerable populations, strength of evidence, preventability and potential regulatory consequence.

A numerical scoring matrix can be used, but GVP does not require one. If a score is used, the organisation should retain the reasoning behind the score rather than allowing the number to become a substitute for scientific judgement.

Assessment

Assessment documentation is generally the richest part of the signal record. It should define the safety hypothesis, evidence set, methods, limitations, alternative explanations, integrated interpretation and conclusion.

The assessment should be capable of standing as a scientific argument. A reviewer should understand why particular evidence was weighted more heavily than other evidence and why the final conclusion followed from that interpretation.

Decision and recommendation

Where a decision is made through a committee or another governance route, the documentation should record the decision and sufficient rationale. It is not necessary to reproduce the full meeting discussion, but the record should not reduce a complex scientific decision to “approved” without explanation.

Action implementation

A signal is not fully managed when the assessment is signed. If the conclusion requires an RMP update, product-information variation, PASS, additional monitoring, risk minimisation or regulatory communication, those actions should remain traceable until implementation.

This is an important distinction between scientific closure and operational completion.

Signal Tracking Systems

A signal-tracking system provides the operational index to the signal lifecycle. It may be a dedicated application, a validated workflow platform or another controlled solution appropriate to the organisation's scale and risk.

Its essential purpose is to answer questions such as:

• What signals are currently open?
• What stage is each signal in?
• Which important decisions are pending?
• Which actions are overdue?
• Which signals have affected regulatory documents?
• Can the supporting assessment and decision records be retrieved?

The system should support the process rather than force scientifically different situations into inappropriate fixed workflows.

Status Design

Signal statuses should be defined clearly enough that two trained users interpret them consistently. Excessive numbers of statuses can create ambiguity; too few can conceal meaningful differences.

A simple model might distinguish:

• detected or under validation;
• validated and awaiting assessment;
• under assessment;
• decision made / actions in progress;
• closed; and
• reopened.

This is illustrative only. Organisations may need different states, but the meaning of each state and the transition rules should be controlled.

Action Tracking

Action tracking is often separated from the scientific signal register, but the connection should remain visible. For each material action, the organisation should be able to identify:

• the originating signal and decision;
• action description;
• accountable owner;
• due date where applicable;
• current status;
• completion evidence; and
• any subsequent effectiveness or follow-up requirement.

A product-information submission that has been initiated but not approved and implemented should not be represented as though the safety action were complete.

Version Control and Audit Trails

Signal documentation changes over time. Assessments are revised, new evidence arrives and decisions may be reconsidered. The record should therefore preserve meaningful version history and auditability.

The objective is not to prevent legitimate correction. It is to preserve enough history to understand what changed, when it changed and why.

Electronic audit trails can be particularly valuable where the tracking platform itself contains regulated decision fields. However, not every explanatory document requires the same technical control. Controls should be proportionate to the record's role in the pharmacovigilance system.

Linking Signals to Other Pharmacovigilance Processes

Signal records should not become isolated from the documents they influence. Important interfaces include:

• PSUR/PBRER: signal status, cumulative evaluation and benefit-risk consequences;
• RMP: addition, reclassification or removal of safety concerns and changes to pharmacovigilance or risk-minimisation activities;
• product information: safety-related changes to the SmPC, PIL or labelling;
• PASS and studies: questions requiring additional evidence;
• risk minimisation: implementation and effectiveness evaluation;
• regulatory correspondence: requests, recommendations and commitments.

Cross-references do not need to duplicate content. Their purpose is to show that one scientific conclusion has been propagated consistently through the relevant parts of the system.

Reopening a Closed Signal

Closure does not mean the subject can never be reconsidered. New cases, a study result, class evidence or regulatory action may justify reopening the hypothesis.

The new record should preserve the relationship to the previous assessment so that the organisation can explain what new information changed the earlier conclusion. Treating the reopened issue as if no prior assessment existed wastes evidence and weakens traceability.

Outsourced Signal Activities

Vendors may perform detection, literature review, analytics, case retrieval or even draft assessments. The MAH should nevertheless retain access to the evidence needed to understand and defend its decisions.

A vendor tracker that provides only status summaries may be insufficient if the MAH cannot retrieve the underlying rationale. Oversight should therefore address both operational performance and record accessibility.

Practical Implementation

A practical documentation model should be designed from the decisions that must remain reconstructable. The following sequence is recommended operational practice, not an EMA-prescribed template:

1. assign a stable signal identifier;
2. preserve the originating source and detection context;
3. record the validation decision and rationale;
4. document priority and escalation where relevant;
5. link the scientific assessment and evidence set;
6. record the decision authority and conclusion;
7. create traceable downstream actions;
8. retain implementation evidence; and
9. close only when the scientific and operational status are accurately represented.

This approach is usually more robust than creating a large mandatory form containing fields that are irrelevant to many signals.

Potential Failure Modes

The following are illustrative failure modes, not reported inspection findings.

Inspection Considerations

An inspector may select a small number of signals and trace them through the entire system. Possible questions include:

• Where did this signal originate?
• Why was it validated or rejected?
• Why did it receive this priority?
• Which evidence was included in the assessment?
• How were contradictory findings handled?
• Who had authority to make the decision?
• What downstream actions followed?
• How do you know those actions were implemented?
• What changed when the signal was reopened or reassessed?
• Can the same conclusion be found consistently in the signal record, RMP, PSUR/PBRER and product information where relevant?

The strongest evidence is a coherent chain of records, not an “inspection pack” assembled shortly before the inspection.

Practical Documentation Review Checklist

The following checklist is recommended operational practice.

1. Does every signal have a stable identifier?
2. Can the source and date of detection be reconstructed?
3. Is the validation rationale explicit?
4. Is prioritisation supported by scientific and public-health reasoning?
5. Are important evidence sources linked or retrievable?
6. Does the assessment distinguish evidence from interpretation?
7. Are limitations and uncertainty visible?
8. Is decision authority clear?
9. Are governance records consistent with the scientific assessment?
10. Are downstream actions linked to the originating signal?
11. Is implementation evidence retained before closure?
12. Can a closed signal be reopened without losing its history?
13. Are vendor records accessible to the MAH?
14. Are system migrations and changes protecting historical traceability?
15. Does the amount of documentation remain proportionate to the significance of the issue?

Key Takeaways

Signal tracking and documentation provide the evidentiary backbone of signal management. Their purpose is to make the lifecycle reconstructable from the original safety information through scientific interpretation, decision and implementation.

GVP does not prescribe a universal signal database schema, mandatory committee record or fixed template pack. Organisations should design controls that reliably preserve the information needed to understand and defend decisions.

Documentation quality depends on relationships, not volume. Source evidence, interpretation, governance decisions and operational implementation should be linked but clearly distinguished.

A signal should not be considered operationally complete merely because the scientific assessment is finished. Required RMP, labelling, study, risk-minimisation or regulatory actions should remain traceable to implementation.

The best test of documentation is whether an independent qualified reviewer can reconstruct what happened, why it happened and what changed as a result.

References

1. European Medicines Agency. Guideline on good pharmacovigilance practices (GVP) Module IX — Signal management (Rev. 1). EMA/827661/2011 Rev. 1.
2. European Medicines Agency. Questions and answers on signal management. EMA/261758/2013 Rev. 5, updated January 2026.
3. European Medicines Agency. Signal management. Current EMA procedural information. https://www.ema.europa.eu/en/human-regulatory-overview/post-authorisation/pharmacovigilance-post-authorisation/signal-management
4. European Medicines Agency. Guideline on good pharmacovigilance practices (GVP) Module I — Pharmacovigilance systems and their quality systems. EMA/541760/2011.
5. European Medicines Agency. Guideline on good pharmacovigilance practices (GVP) Module III — Pharmacovigilance inspections. EMA/119871/2012 Rev. 1.
6. European Union. Commission Implementing Regulation (EU) No 520/2012, as amended by Commission Implementing Regulation (EU) 2025/1466.
7. European Union. Directive 2001/83/EC, as amended.
8. European Union. Regulation (EC) No 726/2004, as amended.

Regulatory Note

This article distinguishes regulatory requirements for an effective, documented signal-management process from organisation-specific records, database fields, templates, sign-offs and governance structures. As of 8 September 2026, GVP Module IX Rev. 1 remains published while EMA prepares revisions following Commission Implementing Regulation (EU) 2025/1466. Current EMA procedural material should be checked before designing or materially revising a live signal-management process.