GVP Module VI: Seriousness, Expectedness and ICSR Case Assessment

Introduction

Once an ICSR has been established as valid, the organisation must assess the information it contains in a structured and clinically meaningful way.

Among the most important assessments are seriousness, expectedness, and the broader medical evaluation of the case.

These concepts are frequently confused. They answer different questions:

Is the case valid?
        ↓
How serious is the reported event?
        ↓
Is the reaction expected under the applicable reference information?
        ↓
What is the medical and regulatory significance?

A valid case is not necessarily serious. A serious case is not necessarily unexpected. An unexpected reaction is not necessarily causally related to the medicinal product.

Keeping these concepts separate is fundamental to accurate ICSR processing.

1. Validity Comes First

The validity assessment establishes whether the report meets the applicable minimum criteria for an ICSR.

Seriousness should not be used to decide whether a report is valid.

For example, a non-serious adverse reaction can be a valid ICSR, while a report describing a potentially life-threatening event may still require further information before the minimum validity criteria are established.

2. What Is Seriousness?

Seriousness is a regulatory classification based on defined outcome or event criteria.

The assessment is not simply a judgement that an event was clinically severe.

The processor should determine whether one or more applicable seriousness criteria are met based on the information available.

3. Serious Is Not the Same as Severe

This distinction is fundamental.

Severe generally describes the intensity of a reaction or symptom.

Serious is a regulatory classification based on specified criteria.

A severe headache may be non-serious, while a relatively brief event can be serious if it meets an applicable regulatory seriousness criterion.

4. Seriousness Criteria

The applicable framework includes criteria such as:

• death;
• life-threatening condition;
• hospitalisation or prolongation of hospitalisation;
• persistent or significant disability or incapacity;
• congenital anomaly or birth defect;
• and other medically important conditions according to the applicable criteria.

The organisation should use the current regulatory definitions and procedures rather than relying on informal interpretations.

5. Death

Death is a seriousness criterion, but the occurrence of death does not by itself establish that the medicinal product caused the death.

The case should separately capture:

• the outcome;
• the clinical circumstances;
• the suspected reaction;
• and the causality assessment.

6. Life-Threatening

Life-threatening should be applied according to the regulatory meaning of the criterion.

It should not be used merely because the event sounds alarming or potentially dangerous.

The case narrative and available clinical information should support the assessment.

7. Hospitalisation

Hospitalisation can meet a seriousness criterion when the applicable requirements are satisfied.

The organisation should distinguish:

• admission to hospital;
• prolongation of an existing hospital stay;
• emergency evaluation without admission;
• and outpatient treatment.

The actual clinical circumstances should drive the assessment.

8. Disability or Incapacity

Persistent or significant disability or incapacity requires appropriate clinical assessment.

Temporary functional limitation should not automatically be classified as permanent or significant disability without supporting information.

Follow-up may be needed where the eventual outcome is not yet known.

9. Congenital Anomaly or Birth Defect

Pregnancy-related reports require careful assessment of maternal exposure, fetal exposure and outcome.

A pregnancy exposure without an adverse outcome should not automatically be classified as a congenital anomaly or birth defect.

The actual reported outcome should determine the classification.

10. Other Medically Important Conditions

Some events may not meet the more specific seriousness criteria but may nevertheless be medically important.

Medical judgement should be applied consistently according to the applicable regulatory framework and organisational procedure.

The rationale should be defensible and, where appropriate, documented.

11. Expectedness Is a Separate Question

Expectedness asks whether the reported reaction is consistent with the applicable reference safety information.

It is not determined by whether the event is common, clinically plausible or already familiar to the physician.

The relevant reference document and applicable regulatory context must be identified.

12. Seriousness and Expectedness Can Combine in Different Ways

A case can be:

Each combination can have different regulatory implications.

The next chunk will cover reference safety information, listedness versus expectedness, causality boundaries, seriousness reassessment, coding and difficult classification scenarios.

13. Reference Safety Information

Expectedness must be assessed against the appropriate reference safety information for the regulatory context of the case.

For EU post-authorisation pharmacovigilance, the authorised product information is a central reference. The organisation should ensure that the version used for assessment is the applicable version for the product, procedure and relevant reporting context.

The processor should not substitute personal clinical familiarity for the controlled reference document.

14. Expectedness Is Not Frequency

An event can be frequently reported and still require assessment against the reference safety information.

Conversely, an uncommon reaction can be listed in the reference information and therefore be considered expected for the applicable assessment.

Expectedness should not be inferred simply from how often the event occurs in the safety database.

15. Expectedness Is Not Causality

Expectedness does not answer whether the medicinal product caused the reaction.

These are separate questions:

Is the reaction described in the applicable reference information?
                         ↓
                 Expectedness

Is there evidence supporting a causal relationship?
                         ↓
                    Causality

A reaction can be expected but not causally related to the product, or unexpected while still being suspected to be related.

16. Expectedness and Medical Terminology

The reported wording does not always correspond exactly to the terminology used in the reference information.

Medical review may therefore be needed to determine whether the reported reaction is adequately represented by an existing reference concept.

The organisation should avoid both over-matching and under-matching clinical terms.

17. Seriousness Can Change During Follow-Up

Seriousness is not necessarily fixed at initial receipt.

For example, a reaction initially described as non-serious may later be associated with hospitalisation, disability or another seriousness criterion.

When material follow-up information is received, the case should be reassessed and updated according to the applicable process.

18. Outcome Can Affect the Assessment

The eventual outcome can provide important clinical context.

For example, a reaction initially recorded as ongoing may later resolve, recur, result in sequelae or be followed by death.

Outcome should be represented separately from seriousness and causality, while recognising that new outcome information can trigger reassessment of other case elements.

19. Medical Review of Seriousness

Medical review is particularly important for borderline seriousness classifications.

Questions may include:

• Did the event actually result in hospitalisation?
• Was the condition genuinely life-threatening at the time of the event?
• Is the reported disability persistent or significant?
• Does the event meet the medically important criterion under the applicable framework?

The conclusion should be supported by the information available.

20. Coding and Clinical Assessment

Coding should represent the reported clinical information accurately.

The coded term should not be used as a shortcut for determining seriousness.

For example, selecting a severe-sounding term does not automatically establish a seriousness criterion. Conversely, a relatively ordinary-sounding term may represent a serious event when the clinical context demonstrates a qualifying outcome.

21. Seriousness and Multiple Reactions

A single ICSR can contain several reactions with different clinical characteristics.

The organisation should assess the seriousness of the reported reactions and the case as a whole according to the applicable reporting framework.

It should be possible to understand which clinical event led to a seriousness classification when the information permits that distinction.

22. Unexpectedness and Multiple Reactions

Different reactions in the same case may not have identical expectedness assessments.

The assessment should therefore be performed against the applicable reference information for the relevant reaction rather than assigning a single expectedness conclusion mechanically to the entire narrative.

23. Reassessment After Regulatory Changes

Reference safety information can change over the product lifecycle.

The organisation should maintain appropriate controls to ensure that expectedness assessments are based on the correct version for the relevant case and reporting context.

A later change to product information should not automatically be treated as evidence that an earlier case was incorrectly assessed under the reference information applicable at that time.

24. Causality: Keep the Boundary Clear

Causality assessment considers whether the available evidence supports a relationship between the medicinal product and the reported event.

Relevant information can include:

• temporal relationship;
• dechallenge;
• rechallenge;
• alternative causes;
• concomitant medicines;
• medical history;
• and objective clinical findings.

Causality should not be inferred from seriousness or unexpectedness alone.

25. Seriousness, Expectedness and Causality Matrix

The three assessments can be considered independently:

Keeping these questions separate reduces classification errors and makes the case easier to defend during inspection.

26. Difficult Classification Scenario: Severe but Non-Serious

A patient reports an extremely painful headache lasting several hours but receives no hospitalisation, does not experience a qualifying life-threatening condition, disability or other applicable seriousness criterion.

The reaction may be clinically severe while remaining non-serious under the regulatory classification.

The case should reflect the clinical severity and the separate seriousness assessment.

27. Difficult Classification Scenario: Hospitalisation for a Non-Serious-Sounding Symptom

A report initially describes dizziness. Follow-up establishes that the patient was admitted to hospital because of the event.

The seriousness assessment should be reassessed based on the additional information. The apparently mild wording of the reaction should not override the documented clinical outcome.

28. Difficult Classification Scenario: Unexpected but Causality Uncertain

A reaction is not described in the applicable reference information, but the available clinical evidence is insufficient to determine whether the medicinal product caused it.

The organisation should not equate unexpectedness with causality. The two assessments should remain separate.

29. Difficult Classification Scenario: Expected but Clinically Important

A reaction is described in the applicable reference information but results in a clinically significant event.

Expectedness does not remove the need to assess seriousness, causality, outcome or other relevant case elements.

The next chunk will cover complex cases, inspection findings, governance, practical examples, final principles, References + Regulatory Note.

30. Complex Case: One Report, Several Reactions

A single report may contain several adverse reactions, each with different seriousness, expectedness and clinical context.

The assessment should not collapse these distinctions merely for administrative convenience. The case record should preserve the relationship between the reported clinical events and the assessments applied to them.

For example, a patient may report nausea, dizziness and syncope after treatment. Nausea may be non-serious and expected, while syncope may require a separate assessment because of hospitalisation or another seriousness criterion. The available information should determine the assessment rather than the most prominent term in the narrative.

31. Complex Case: Follow-up Changes the Classification

Case assessment is iterative.

A case may initially contain limited information and later acquire clinically important details. Follow-up can establish hospitalisation, a persistent disability, a different outcome, an alternative diagnosis, or information that changes the clinical interpretation.

The organisation should have controls ensuring that material follow-up information triggers appropriate reassessment rather than being appended to the narrative without reconsidering affected case fields.

32. Complex Case: Conflicting Information

Different sources may provide apparently conflicting information about seriousness or outcome.

The organisation should evaluate the available evidence, document the clinical reasoning where necessary and avoid resolving uncertainty through arbitrary selection of the more favourable classification.

Where material uncertainty remains, appropriate medical review and follow-up should be considered.

33. Complex Case: Seriousness Based on Medical Importance

The medically important criterion requires particular care because it is not simply a synonym for “severe”.

A medically important event may justify intervention to prevent one of the recognised serious outcomes even where those outcomes have not actually occurred. The assessment should follow the applicable regulatory definition and the organisation's controlled procedure.

A blanket rule such as “all emergency-room visits are serious” or “only hospital admissions are medically important” is inappropriate.

34. Complex Case: Expectedness After a Product-Information Change

Suppose a reaction was not described in the applicable reference information when a case was assessed, but the product information was subsequently updated to include that reaction.

The later update does not, by itself, mean that the historical assessment was incorrect. The organisation should determine the reference information applicable to the relevant reporting context and maintain appropriate version control.

This is one reason why controlled reference-safety-information governance is important.

35. Practical Processing Sequence

A defensible workflow can be represented as:

Valid ICSR
   ↓
Identify reported reactions/events
   ↓
Assess seriousness
   ↓
Assess expectedness against applicable reference information
   ↓
Assess causality and clinical context
   ↓
Assess outcome and other required data elements
   ↓
Medical review where required
   ↓
Document/update the case
   ↓
Submit or otherwise process according to applicable requirements

The sequence is not intended to imply that every assessment must be performed by a different person or in exactly this chronological order. Rather, it illustrates the logical separation of the concepts.

36. Quality-System Controls

The quality system should make consistent case assessment possible.

Useful controls include:

• controlled access to current reference safety information;
• documented case-processing procedures;
• training on seriousness and expectedness criteria;
• medical-review criteria for difficult cases;
• quality-control checks;
• appropriate audit trails for material changes;
• and escalation pathways for uncertain or disputed classifications.

Controls should address both the initial assessment and reassessment after follow-up.

37. Inspection Perspective

An inspector may examine whether the organisation can demonstrate that case classifications are systematic rather than dependent on individual processor preference.

Potential inspection questions include:

• What procedure defines seriousness assessment?
• How do processors access the applicable reference safety information?
• How is the correct product-information version established?
• When is medical review required?
• How are changes following follow-up identified?
• How are borderline cases handled?
• How does quality control detect inconsistent classifications?

The organisation should be able to demonstrate not merely that a classification exists, but why the classification was reasonable from the information available at the relevant time.

38. Common Failure Modes

Treating “severe” as “serious”

Clinical intensity is incorrectly converted into a regulatory seriousness classification.

Treating unexpectedness as causality

An unexpected reaction is assumed to have been caused by the medicinal product.

Using frequency as expectedness

A frequently reported event is automatically treated as expected without checking the applicable reference information.

Ignoring follow-up reassessment

New hospitalisation, disability or other clinically important information is added to the narrative but the affected assessment fields are not reconsidered.

Using the wrong reference-information version

A current product-information version is applied retrospectively without considering the reference information applicable to the relevant assessment.

Treating one case-level classification as sufficient

Multiple reactions with different clinical characteristics are reduced to a single undifferentiated assessment.

39. Practical Example: A Case That Changes Over Time

A patient reports severe abdominal pain after treatment. At initial receipt, there is no information indicating hospitalisation, disability or another applicable seriousness criterion. The case is therefore assessed using the information available at that point.

During follow-up, the reporter confirms that the patient was admitted to hospital because of the abdominal pain and required treatment.

The case should be reassessed using the new information. The update is not merely a narrative amendment; it may affect the seriousness classification and downstream processing.

40. Practical Example: Expected Does Not Mean Unimportant

A reaction is clearly described in the applicable reference information. The patient nevertheless experiences a clinically significant event requiring substantial medical intervention.

The fact that the reaction is expected does not remove the need for appropriate seriousness, outcome, causality and clinical assessment.

Expectedness is one assessment dimension, not a global determination of the importance of the case.

41. QPPV Considerations

The QPPV should ensure that the PV system provides a controlled and reproducible approach to these assessments.

This includes appropriate procedures, competent personnel, medical oversight where required, quality controls, access to current reference information and mechanisms for detecting material reassessment needs.

The QPPV should also be able to understand significant classification trends and recurring quality issues rather than treating seriousness and expectedness as purely transactional case-processing fields.

42. Key Takeaways

1. Validity, seriousness, expectedness and causality answer different questions.
2. Severe does not automatically mean serious.
3. Expectedness must be assessed against the applicable reference safety information.
4. Unexpectedness does not establish causality.
5. Case assessments may need to change when follow-up information becomes available.
6. Multiple reactions within one case may require distinct assessments.
7. Reference-information version control is essential for defensible expectedness assessment.
8. Medical review is important for difficult or borderline clinical classifications.
9. Inspection readiness requires evidence of a controlled assessment process, not merely completed database fields.
10. The organisation should preserve the reasoning and evidence needed to explain material classifications.

References

1. European Medicines Agency. Good pharmacovigilance practices (GVP): Module VI — Collection, management and submission of reports of suspected adverse reactions.
2. European Medicines Agency. Good pharmacovigilance practices (GVP): Annex I — Definitions.
3. European Parliament and Council. Directive 2001/83/EC on the Community code relating to medicinal products for human use, as amended.
4. European Commission. Commission Implementing Regulation (EU) No 520/2012 on the performance of pharmacovigilance activities provided for in Regulation (EC) No 726/2004 and Directive 2001/83/EC, as amended.
5. European Medicines Agency. Guideline on good pharmacovigilance practices: Product- or Population-Specific Considerations, as applicable to the clinical context.

Regulatory Note

This article is an educational interpretation of the EU pharmacovigilance framework. It does not replace the applicable legislation, current GVP text, product information, regulatory procedures or competent-authority requirements.

GVP modules and associated guidance are revised over time. Before applying a requirement operationally, verify the current EMA version, applicable EU legislation and the reference safety information relevant to the medicinal product and reporting context.