GVP Module XVI: Risk Minimisation Measures

Introduction

Risk minimisation is the part of pharmacovigilance concerned with reducing the likelihood or severity of an identified or potential risk associated with a medicinal product. It therefore sits at the point where safety knowledge is translated into measures intended to influence clinical use, monitoring or other conditions of treatment.

Risk minimisation does not mean that every safety concern requires an additional intervention. The appropriate response depends on the nature and seriousness of the risk, the extent to which routine measures are sufficient, the characteristics of the affected population and the evidence available about how the risk can be controlled.

GVP Module XVI provides the principal EU framework for risk minimisation measures. EMA currently lists Module XVI Revision 3 as legally effective from 6 August 2024, together with Addendum I on embryo-fetal risks, effective from 29 August 2025, and Addendum II on methods for evaluating effectiveness, effective from 6 August 2024. citeturn0search0turn0search2

The central question is therefore not simply what measure can be introduced, but what intervention is proportionate to the risk and capable of achieving its intended objective.

1. Risk Minimisation Begins With a Risk-Management Question

A risk minimisation measure should be connected to a defined safety concern.

The organisation first needs to understand what harm is being prevented or reduced, which patients are affected, under what circumstances the risk arises and what factors may make the outcome preventable.

Only then can an appropriate measure be selected. Starting with a preferred communication, educational programme or monitoring requirement and attempting to fit the safety concern to it reverses the logical sequence.

2. Routine and Additional Risk Minimisation

Risk minimisation measures are commonly understood as either routine or additional.

Routine measures are embedded in the normal conditions governing medicinal-product use, including information provided through the product information and other standard regulatory controls. Additional measures are introduced when routine measures alone are not considered sufficient to address a specific safety concern.

The distinction matters because an additional measure requires a stronger justification and a controlled approach to implementation and effectiveness.

3. The Risk Management Plan

The risk management plan provides the principal framework for documenting important safety concerns and the measures and activities intended to manage them.

Risk minimisation should therefore not be developed independently from the RMP. The safety concern, proposed measure, target population, implementation arrangements and effectiveness evaluation should form a coherent chain.

The RMP also provides the context in which new evidence can lead to adaptation of risk minimisation during the product lifecycle.

4. Routine Risk Minimisation

Routine risk minimisation relies on measures that form part of normal medicinal-product regulation and use.

These may include the information in the summary of product characteristics, package leaflet and labelling, together with prescribing status, pack size, legal supply conditions and other applicable regulatory measures.

Routine measures are not passive. Their effectiveness depends on whether the information reaches the relevant users and whether it is sufficiently clear to support appropriate use.

5. Additional Risk Minimisation

Additional risk minimisation measures are used when routine measures are insufficient to manage a specific risk adequately.

Examples can include educational materials, controlled access arrangements, pregnancy-prevention programmes, controlled distribution or specific monitoring arrangements, depending on the risk and regulatory context.

An additional measure should have a defined purpose and target. It should not be introduced merely because a risk is serious; the organisation should have a reason to believe that the intervention can reduce the relevant risk.

6. Selecting the Measure

Selection should consider the mechanism of the risk and the point at which intervention is possible.

If the risk results from inappropriate patient selection, a measure may need to influence prescribing criteria. If it arises from incorrect administration, education or device-related controls may be more relevant. If the risk depends on exposure during a particular physiological state, targeted prevention measures may be appropriate.

The intervention should therefore address the pathway leading to harm rather than simply increase the amount of information available.

7. Proportionality

Risk minimisation should be proportionate to the seriousness and characteristics of the risk.

An intervention that imposes substantial burden on healthcare professionals or patients may be inappropriate if a simpler measure can achieve the same objective. Conversely, a low-intensity intervention may be inadequate for a serious preventable risk.

Proportionality should therefore consider both the burden of the measure and the potential consequences of insufficient control.

8. The Intended Objective

Every important risk minimisation measure should have a defined objective.

The objective might be to ensure that a contraindicated patient is not exposed, that a laboratory parameter is monitored before treatment, that a specific symptom is recognised promptly or that a particular administration error is prevented.

A measurable objective provides the foundation for later evaluation. Without one, effectiveness becomes difficult to distinguish from simple implementation.

9. The Target Population

The target population should be defined sufficiently precisely for the measure to reach the people who need it.

Some risks apply to all users, while others arise only in particular age groups, clinical circumstances, concomitant treatments or stages of treatment.

Overly broad measures can create unnecessary burden and reduce attention to the people at greatest risk. Overly narrow measures can leave important patients outside the control strategy.

10. Healthcare-Professional Role

Healthcare professionals may be central to implementing a risk minimisation measure.

They may need to identify eligible patients, provide counselling, perform monitoring, interpret results, follow prescribing restrictions or report relevant outcomes.

The measure should therefore be designed around the actual clinical workflow. An intervention that requires actions unlikely to occur reliably in practice may be formally comprehensive but operationally ineffective.

11. Patient Role

Patients may also have an important role, particularly where risk depends on recognition of symptoms, adherence to precautions, pregnancy prevention or correct administration.

Patient-facing information should be understandable and focused on the action required. It should not transfer inappropriate clinical responsibility to patients when professional assessment is necessary.

Where patients are expected to perform an action, the feasibility of that action should be considered when designing the measure.

12. Stakeholder Engagement

Risk minimisation can benefit from early engagement with the people who will implement or experience the measure.

Healthcare professionals and patients can identify practical barriers that may not be visible from a purely regulatory or pharmacovigilance perspective.

Engagement should inform the design without replacing the MAH's or regulatory authority's responsibility for the final safety and regulatory assessment.

13. Educational Materials

Educational materials are an example of additional risk minimisation used to provide information beyond the standard product information when a specific safety objective requires it.

Their content should be focused on the risk and the behaviour or decision they are intended to support. More information is not necessarily better information.

Where approval by a competent authority is required, the material should follow the applicable national regulatory process.

14. Controlled Access Measures

Some risks require controls over who can prescribe, dispense or receive a medicinal product, or under what conditions it can be supplied.

Such measures can be appropriate where exposure should occur only after defined criteria have been satisfied.

Because controlled-access systems can impose significant operational burden, their design should be proportionate and their implementation should be monitored carefully.

15. Pregnancy-Prevention Measures

Pregnancy-related risks may require specific measures addressing contraception, pregnancy testing, counselling, treatment eligibility or monitoring.

EMA's current Module XVI Addendum I provides specific guidance for risk minimisation measures concerning embryo-fetal risks and became legally effective on 29 August 2025. It applies to new applications, new risk minimisation measures and new effectiveness studies within its stated scope. citeturn0search19

Such programmes should be designed around the actual mechanism and timing of fetal exposure rather than treated as generic educational interventions.

16. Monitoring as a Risk-Minimisation Measure

Monitoring requirements can form part of risk minimisation when measuring a clinical parameter can identify or prevent harm.

The value of monitoring depends on whether the test is performed at the appropriate time, whether results are interpreted correctly and whether an abnormal result leads to the intended action.

A requirement to perform a test without a defined clinical response may therefore provide less risk control than its existence suggests.

17. Communication and Risk Minimisation

Safety communication and risk minimisation overlap but are not identical.

A communication may simply inform an audience about a safety issue, while a risk-minimisation measure is designed to reduce a defined risk through a specific intervention.

Where communication forms part of risk minimisation, its objective, target population, implementation and effectiveness should be connected to the corresponding risk-management framework.

18. Relationship With Signal Management

A signal may lead to a new or changed safety concern, which may in turn trigger reassessment of existing risk minimisation.

The existence of a signal does not automatically require an additional measure. The evidence must support the need for intervention and the proposed measure must be capable of addressing the relevant risk.

Signal management therefore supplies safety evidence, while risk minimisation translates an established risk-management need into an intervention.

19. Relationship With Regulatory Decision-Making

Risk minimisation measures may be introduced, modified or discontinued through regulatory procedures.

The regulatory decision determines the applicable authorised or required measures, while the MAH is responsible for implementing its obligations within the pharmacovigilance system.

The scientific rationale, regulatory decision and operational implementation should remain traceable to one another.

20. Risk Minimisation as a Lifecycle Process

Risk minimisation should be understood as a lifecycle rather than a one-time intervention:

Safety concern
      ↓
Risk characterisation
      ↓
Need for intervention
      ↓
Measure selection
      ↓
Implementation
      ↓
Evaluation
      ↓
Adaptation
      ↓
Reassessment

This lifecycle is central to modern risk minimisation. A measure that was appropriate when introduced may become unnecessary, insufficient or unnecessarily burdensome as evidence and clinical practice change.

21. Implementing a Risk Minimisation Measure

Implementation converts the approved or selected measure into the activities performed in real clinical settings.

The organisation should identify the relevant population, responsible parties, materials, systems and timing. Implementation arrangements should be sufficiently specific to allow the organisation to determine whether the measure actually reached the intended setting.

A risk-minimisation measure that exists only as an approved document has not yet demonstrated operational value.

22. Distribution Is Not Implementation

Sending educational materials or making information available does not necessarily mean that the measure has been implemented effectively.

Implementation may require healthcare professionals to receive, use and act upon the information, or patients to complete a defined precaution. The distinction becomes particularly important when the measure depends on behaviour.

The organisation should therefore define what constitutes implementation before evaluating performance.

23. Behavioural Objectives

Additional risk minimisation often depends on a change in behaviour.

The relevant behaviour should be stated explicitly. Examples include avoiding a contraindicated combination, performing a test before treatment, counselling a patient, following a preparation procedure or reporting a particular symptom promptly.

A behavioural objective provides a much stronger basis for evaluation than a general statement that awareness should increase.

24. Barriers to Implementation

A measure may fail because the intended behaviour is difficult to perform even when the information itself is correct.

Potential barriers include time pressure, fragmented care, complex workflows, insufficient training, poor access to testing, unclear responsibility or excessive administrative burden.

Understanding these barriers is part of effective risk minimisation because a theoretical intervention may not translate into risk reduction in practice.

25. Quality Management

Risk minimisation measures form part of the pharmacovigilance quality system when they are pharmacovigilance activities under the applicable framework.

Responsibilities, procedures, training, records and quality controls should therefore be appropriate to the measure.

The control system should focus on the critical steps that determine whether the intended risk reduction can occur.

26. Training and Competence

Where implementation requires specific professional actions, relevant personnel need appropriate information and competence.

Training should address the actual task rather than merely demonstrate completion of a generic module. For example, a professional responsible for monitoring may need to understand when to test, how to interpret the result and what action follows.

Training completion alone does not establish that the intended behaviour has occurred.

27. Digital Implementation

Digital systems can support risk minimisation through electronic prescribing, alerts, registries, reminders or access controls.

Digital controls can improve reliability but can also generate alert fatigue, workarounds or inappropriate automation. Their design should therefore consider the clinical environment in which they operate.

Where a digital intervention is critical to risk control, relevant system changes should be subject to appropriate validation and change control.

28. Registries and Controlled Programmes

Some risk-minimisation strategies use registries or controlled programmes to document exposure, monitoring or eligibility.

Such systems should have a defined purpose and data requirements proportionate to that purpose. Data collection should not become an objective in itself.

The organisation should understand how registry information contributes to risk control or effectiveness evaluation.

29. Coordination With the Regulatory Network

Additional risk minimisation can involve the MAH, EMA, national competent authorities and other participants in the EU regulatory network.

Responsibilities for design, approval, distribution, implementation monitoring and effectiveness evaluation should be clear. Where Member State processes differ, the organisation should preserve the common safety objective while meeting applicable national requirements.

The regulatory network's oversight does not remove the MAH's responsibility to operate its own controlled implementation process.

30. Changes to Risk Minimisation

Risk minimisation should be adaptable when new evidence becomes available.

A measure may need to be strengthened when the risk is not adequately controlled, simplified when an unnecessary burden is identified, or discontinued when the underlying concern is no longer relevant and the applicable regulatory requirements permit it.

Changes should be based on evidence and follow the relevant regulatory process.

31. Evaluating Implementation

Implementation evaluation asks whether the measure reached the intended setting and was delivered as planned.

This is distinct from effectiveness. A measure may be implemented perfectly but have little effect because the underlying intervention is ineffective. Conversely, an effective measure may appear weak if implementation was incomplete.

Separating these questions makes evaluation more informative.

32. Evaluating Behavioural Change

Where the measure is intended to change behaviour, evaluation should examine whether the relevant behaviour changed.

For example, a programme designed to encourage a laboratory test should assess whether the testing behaviour changed, not simply whether the educational material was distributed.

Behavioural measures should be interpreted in context because changes may result from other interventions or broader clinical practice changes.

33. Evaluating Clinical Outcomes

The ultimate objective of risk minimisation is usually to reduce the occurrence or severity of the relevant risk.

Outcome evaluation may therefore examine clinical events, exposure patterns or other relevant measures. However, outcome data can be affected by many factors, including changes in exposure and disease incidence.

A reduction in events should not automatically be attributed to the risk-minimisation measure without appropriate consideration of alternative explanations.

34. The Evaluation Framework

GVP Module XVI Revision 3 places greater emphasis on planned and prospective evaluation, including implementation, behavioural change and outcomes. EMA also maintains Addendum II specifically addressing methods for evaluating effectiveness of risk minimisation measures. citeturn0search17turn0search0

A useful conceptual sequence is:

Was the measure implemented?
          ↓
Did the intended behaviour change?
          ↓
Did exposure change appropriately?
          ↓
Did the clinical outcome improve?
          ↓
Can the change reasonably be attributed to the measure?

Not every measure requires every level of evaluation. The appropriate design depends on the objective, risk and available evidence.

35. Evaluation Design

Evaluation methods should be selected prospectively where possible.

Possible approaches include targeted data review, surveys, healthcare-utilisation analyses, observational studies, registries and other appropriate methods. The choice should follow the question being asked rather than the availability of a convenient dataset.

A weak evaluation question can produce a technically sophisticated study that does not demonstrate whether the measure achieved its purpose.

36. Baselines and Comparators

Evaluation is easier when the organisation understands the relevant baseline before or at implementation.

Where appropriate, comparison with a pre-intervention period, reference population or other suitable comparator can help distinguish a meaningful change from background variation.

The comparator should be appropriate to the clinical and epidemiological context.

37. Confounding in Effectiveness Evaluation

Changes observed after implementation may have several causes.

New clinical guidelines, changes in prescribing, product availability, media attention or other risk-minimisation measures can occur at the same time. These factors should be considered when interpreting effectiveness evidence.

The objective is not necessarily to prove causality with absolute certainty, but to make the attribution judgement scientifically credible.

38. When a Measure Is Not Effective

Evidence that a risk-minimisation measure is not achieving its objective should trigger reassessment.

The organisation should determine whether the problem lies in implementation, the intervention itself, the target population, the underlying risk model or the evaluation method.

The response may involve adaptation of the measure, additional intervention, generation of further evidence or regulatory reassessment.

39. When a Measure Creates Unintended Consequences

Risk minimisation can introduce unintended effects.

A restrictive measure may delay access to appropriate treatment, create administrative barriers or shift prescribing to alternatives with different risks. Educational requirements may create burden without improving behaviour.

The overall benefit-risk context should therefore be considered when evaluating whether a measure remains appropriate.

40. Multiple Measures for One Risk

A single safety concern may be addressed through several complementary measures.

For example, product information may be combined with educational materials and monitoring requirements. Evaluation should consider whether the combined strategy is functioning as intended and, where possible, whether individual components contribute meaningfully.

Overlapping interventions can make attribution difficult and should be considered in evaluation design.

41. Multiple Products With the Same Active Substance

The same active substance may be associated with risk-minimisation measures across several products or marketing authorisation holders.

Coordination can be important where inconsistent measures would create confusion for healthcare professionals or patients. EMA's current Module XVI includes specific guidance on coordination of risk-minimisation effectiveness evaluation for medicinal products containing the same active substance. citeturn0search17

The precise regulatory arrangements depend on the products and authorisation procedures involved.

42. Risk Minimisation and the PSUR

Relevant risk-minimisation information can feed into periodic safety evaluation.

The PSUR may contain information about the implementation and effectiveness of risk-minimisation measures, while new aggregate evidence may prompt reassessment of those measures.

This creates another lifecycle interface between routine aggregate evaluation and risk-management activity.

43. Risk Minimisation and Signal Management

Signal management can identify new evidence that changes the risk profile, while risk-minimisation evaluation can generate evidence about whether an intervention is controlling that risk.

The processes therefore interact in both directions.

A new signal may trigger reassessment of an existing measure, while evidence from risk-minimisation activities may alter the interpretation of the underlying safety concern.

44. Risk Minimisation and Safety Communication

Safety communication is often one component of risk minimisation, but not every communication is a risk-minimisation measure.

Where communication is intended to change a defined behaviour or reduce a specific risk, its objective and effectiveness should be integrated into the RMP framework.

This distinction prevents the organisation from counting every safety message as evidence that risk has been controlled.

45. Documentation and Traceability

The organisation should be able to reconstruct why a risk-minimisation measure was selected, what objective it was intended to achieve, how it was implemented and what evidence was subsequently obtained about its performance.

Records may include the relevant safety assessment, RMP, regulatory decision, approved materials, distribution and implementation records, evaluation plans, results and subsequent decisions.

Traceability is particularly important when a measure is modified because the organisation needs to understand what evidence led to the change.

46. Inspection Perspective

An inspector assessing risk minimisation is likely to look beyond whether the RMP contains a measure.

The practical questions include whether the measure was implemented as required, whether responsibilities were clear, whether deviations were identified, whether effectiveness was evaluated appropriately and whether the organisation acted when evidence indicated that the measure was insufficient or unnecessarily burdensome.

The evidence should connect the risk, the measure, implementation and outcome.

47. Illustrative Inspection Scenario: The Measure Exists but Is Not Used

An educational programme is approved and distributed, but records indicate that healthcare professionals rarely use the material in the clinical setting for which it was designed.

The potential weakness is an implementation gap. The existence of approved material does not demonstrate that the intended intervention reached practice.

The organisation should investigate the barrier and determine whether adaptation is required.

48. Illustrative Inspection Scenario: Effectiveness Is Equated With Distribution

A company reports that an educational measure was sent to 95% of the intended recipients and concludes that the risk-minimisation measure was effective.

Distribution is evidence of implementation activity, but it does not demonstrate that the intended behaviour changed or that the risk was reduced.

Where effectiveness evaluation is required, the relevant objective should determine the evidence collected.

49. Illustrative Inspection Scenario: An Ineffective Measure Is Left Unchanged

Evaluation shows that the intended behaviour has not changed, but the organisation continues the same intervention without documented reassessment.

The potential weakness is treating implementation as the endpoint rather than using evaluation to inform the next decision.

Risk minimisation should be adaptive when evidence indicates that the current strategy is not achieving its objective.

50. Illustrative Inspection Scenario: Excessive Burden Is Ignored

A risk-minimisation programme introduces several mandatory steps for healthcare professionals, but there is no assessment of whether the burden affects access to treatment or causes workarounds.

The potential weakness is evaluating only the intended benefit while ignoring unintended consequences.

A proportionate risk-minimisation strategy considers both risk control and the practical effects of the intervention.

51. The QPPV and Risk Minimisation

The QPPV does not need to personally design every risk-minimisation measure. The QPPV's role is to provide appropriate oversight of the pharmacovigilance system and its important safety activities.

For significant risks, QPPV oversight should include sufficient visibility of the risk-management strategy, major changes, important implementation problems and effectiveness conclusions to support appropriate pharmacovigilance governance.

The level of involvement should remain proportionate to the significance of the issue.

52. MAH Responsibility

The MAH remains responsible for operating its pharmacovigilance system and implementing applicable risk-management obligations.

Tasks may be distributed among pharmacovigilance, regulatory, medical, commercial or external functions, but responsibility and oversight should remain clear.

Where implementation is delegated to affiliates or vendors, the MAH should retain access to the information needed to determine whether the measure is functioning as intended.

53. Vendor and Affiliate Oversight

External organisations may distribute materials, operate registries, conduct surveys or support effectiveness studies.

The MAH should define responsibilities, quality controls, data flows, escalation routes and records required for oversight.

The critical question is whether outsourcing preserves control of the risk-minimisation process rather than merely transferring the operational task.

54. Deviations and CAPA

Failures in risk-minimisation implementation should be assessed according to their potential impact on patient safety and regulatory compliance.

A deviation that affects a critical patient-safety control may require a different response from an administrative error with no meaningful effect on risk control.

Where recurring or significant weaknesses are identified, CAPA should address the underlying cause and evaluate whether the measure itself remains appropriate.

55. Change Management

Changes to products, indications, formulations, healthcare practice, patient populations or safety knowledge can affect the suitability of risk-minimisation measures.

The organisation should therefore assess relevant changes for their potential effect on the RMP and associated measures.

A measure designed for an earlier stage of the product lifecycle may need to be revised as evidence and clinical use evolve.

56. Reassessment and Adaptation

Risk minimisation should remain connected to the evolving safety profile.

New evidence may indicate that a risk is better understood, that an existing measure is insufficient, or that a measure is no longer justified. The appropriate response should follow the evidence and applicable regulatory process.

Adaptation is therefore a normal component of effective risk minimisation rather than evidence that the original measure was necessarily inappropriate.

57. A Practical Review Framework

A useful review sequence is:

What risk are we trying to control?
             ↓
Who is exposed and under what circumstances?
             ↓
What behaviour or condition must change?
             ↓
Is the selected measure capable of producing that change?
             ↓
Was the measure implemented as intended?
             ↓
Did behaviour change?
             ↓
Did the clinical risk change?
             ↓
What alternative explanations exist?
             ↓
Should the measure be maintained, adapted or discontinued?

The framework links the scientific rationale to the operational and effectiveness questions that follow.

58. What Effective Risk Minimisation Looks Like

An effective risk-minimisation system does not simply contain a collection of measures.

It connects each measure to a defined risk and objective, identifies the people and behaviours involved, implements the intervention in the real clinical environment and obtains evidence about whether it achieved its purpose.

Where the evidence indicates that the intervention is insufficient or creates disproportionate burden, the system should be capable of adaptation.

59. Final Principle

Risk minimisation is the practical expression of risk management. Its purpose is not to demonstrate that an intervention exists, but to reduce a defined risk through an intervention that is proportionate, feasible and capable of being evaluated.

The strongest risk-minimisation systems therefore maintain a continuous connection between risk characterisation, measure selection, implementation, effectiveness and adaptation.

This is why GVP Module XVI should be read not as a catalogue of possible measures, but as a framework for designing and governing interventions throughout the medicinal-product lifecycle.

Key Takeaways

Routine risk minimisation forms part of normal medicinal-product use and regulatory control, while additional risk minimisation is introduced when routine measures are not sufficient for a specific safety concern.

A measure should have a defined risk, target population and objective. Implementation, behavioural change and clinical outcomes are distinct concepts and should not be conflated when evaluating effectiveness.

Risk minimisation is iterative. New evidence can lead to strengthening, modification, simplification or discontinuation of a measure through the appropriate regulatory process.

References

1. European Medicines Agency. Good Pharmacovigilance Practices (GVP), Module XVI — Risk minimisation measures (Rev. 3).
2. European Medicines Agency. GVP Module XVI Addendum I — Risk minimisation measures for medicinal products with embryo-fetal risks.
3. European Medicines Agency. GVP Module XVI Addendum II — Methods for evaluating effectiveness of risk minimisation measures.
4. European Medicines Agency. Good Pharmacovigilance Practices (GVP), Module V — Risk management systems.
5. European Medicines Agency. Good Pharmacovigilance Practices (GVP), Module XV — Safety communication.
6. Regulation (EC) No 726/2004, as amended.
7. Directive 2001/83/EC, as amended.
8. Commission Implementing Regulation (EU) No 520/2012, as amended.

Regulatory Note

Module XVI Revision 3 is currently listed by EMA with a legal effective date of 6 August 2024. EMA also lists Addendum I on embryo-fetal risks with a legal effective date of 29 August 2025 and Addendum II on effectiveness evaluation with a legal effective date of 6 August 2024. citeturn0search0turn0search19

This article distinguishes regulatory requirements from recommended operational practice. Current legislation, GVP guidance and applicable national procedures should be verified when applying the framework to a specific medicinal product.

Inspection scenarios are illustrative and are not presented as documented regulatory findings.