SOP: Signal Detection and Management

This is a ready-to-use SOP for signal detection and management, the workflow that turns individual cases into knowledge about a product’s evolving benefit-risk profile. Replace every <<FILL: ...>> placeholder, route it through document control, and confirm every cited regulation against the current source before you rely on it. This content is educational reference, not legal or regulatory advice.

Document control header

1. Purpose

This procedure defines how <<FILL: COMPANY NAME>> detects, validates, prioritizes, assesses, and dispositions safety signals across its marketed product portfolio, so that every signal has a documented, attributable, and defensible outcome and none is dismissed without a recorded rationale.

2. Scope

This procedure applies to signals arising from qualitative case review, quantitative disproportionality screening, literature, and external sources (health authority communications, partner notifications) for all marketed products of <<FILL: COMPANY NAME>>. It does not cover the underlying case processing that feeds the signal dataset, governed by <<FILL: SOP-ID for ICSR intake>>.

3. Responsibilities

4. Definitions

• Signal: information arising from one or multiple sources, including observation and experiment, that suggests a new potentially causal association, or a new aspect of a known association, between an intervention and an event, judged to warrant verificatory action.
• Disproportionality: a statistical screening method comparing the observed reporting rate of a product-event pair against an expected rate, used to flag pairs for qualitative review, not to establish causation.
• Validation: the step confirming there is sufficient evidence to treat a detected signal as a genuine candidate for further evaluation, as opposed to noise, duplication, or a known and already-labeled association.
• Signal disposition: the documented conclusion of the assessment: no further action, continued monitoring, further study, or a labeling or risk-minimization change.

5. Procedure

5.1 Detection

1. Run the validated quantitative disproportionality screen (for example PRR, ROR, EBGM, or IC, per <<FILL: the statistical method(s) in use>>) on the defined cadence: <<FILL: e.g. monthly for products within 2 years of launch, quarterly thereafter>>.
2. Perform qualitative case review on the same or a more frequent cadence for products under enhanced monitoring, designated medical events, and any product with an active safety topic.
3. Capture every candidate from both routes, along with the source and date of detection, in the signal log (section 8).

5.2 Validation

1. For each detected candidate, assess whether the evidence is sufficient to treat it as a genuine new or changed association, considering case quality, biological plausibility, temporal relationship, and whether it duplicates an already-tracked signal.
2. Record a validated / not validated decision with a named decision-maker and a written rationale. A “not validated” decision is not a dismissal without a rationale; it is a documented judgment that must itself withstand inspection.

5.3 Prioritization

1. Score each validated signal for urgency using seriousness, novelty, strength of association, and public health impact.
2. Assign a prioritization tier and a target timeline for assessment completion consistent with the tier.

5.4 Assessment

1. Conduct a full review of all available evidence: case narratives, literature, non-clinical and mechanistic data, epidemiology, and comparator information where relevant.
2. Document the assessment in a structured format sufficient for the signal review board to reach a decision without needing to reconstruct the underlying data independently.

5.5 Recommendation and disposition

1. The medical reviewer proposes a disposition: no action, continued monitoring, further study (for example a PASS), an RMP or REMS update, or a labeling change.
2. The signal review board reviews and either confirms or returns the recommendation with reasons.
3. Record the final disposition, the decision-maker(s), and the date.

5.6 Exchange of information and action

1. Notify health authorities of the outcome where required by the applicable framework and timeline.
2. Implement any agreed action (label change, RMP/REMS update, communication) through the applicable change-control or regulatory-submission procedure.
3. Close the signal record only once the agreed action is implemented or confirmed as not required.

6. Acceptance criteria

• Every detected signal, from either the quantitative screen or qualitative review, is logged with source and date.
• Every signal has a documented validation decision with a named decision-maker and rationale, whether or not it proceeds.
• Every signal reaching assessment has a structured, evidence-based write-up sufficient for independent review.
• Every disposition, including “no action,” is documented with the decision-maker, the date, and the rationale.
• The screening cadence actually run matches the cadence defined in section 5.1; deviations from the schedule are documented.

7. References

EU GVP Module IX, Signal Management. ICH E2C(R2), Periodic Benefit-Risk Evaluation Report (for how signal disposition feeds the aggregate report). FDA postmarket safety surveillance framework under 21 CFR 314.80 / 600.80.

Confirm the current version and clause numbers of each reference before issue.

8. Record generated: signal log entry

9. Revision history

10. Approvals

Filled specimen

A completed signal log entry for an example product. Illustrative only.

This entry shows the chain an inspector will look for: a statistical flag with its actual numbers, a validation decision attributed to a named physician with a stated rationale, a prioritization tier, and a disposition the board approved, all dated and none skipped, even though the ultimate action was “keep watching” rather than a label change.

Common inspection findings this SOP prevents

• A signal is detected but never logged, because the quantitative screen output was reviewed informally and not captured.
• A “not validated” or “no action” decision has no recorded rationale or decision-maker, so it cannot be reconstructed later.
• The screening cadence in the SOP is not the cadence actually run, with no documented deviation.
• A signal reaches “assessed” status with no structured write-up, only meeting notes referencing it.
• A disposition is implemented (a label change, for example) with no traceable link back to the signal record that recommended it.

How to adapt this SOP

1. Set your document number, owner, and effective date in the header.
2. Name your actual disproportionality methods and screening cadence in section 5.1.
3. Define your organization’s prioritization scoring criteria concretely in section 5.3, not only by name.
4. Point the cross-reference in section 2 to your real ICSR intake and periodic reporting procedures.
5. Confirm every regulation in section 7 against the current published version before issue.