Drug–Drug Interactions and Pharmacovigilance: Recognising, Assessing and Characterising Safety Risks

Purpose and Scope

Drug–drug interactions are a recurring source of preventable or potentially preventable adverse drug reactions, but their pharmacovigilance assessment is more complicated than identifying two medicines that were taken together. Concomitant use is common, particularly in patients with multiple conditions, and the presence of two medicines in a case does not by itself establish that an interaction occurred or contributed to the reported event.

The pharmacovigilance task is therefore to connect the clinical event with the medicines and exposures that may have produced it. That requires consideration of pharmacokinetic and pharmacodynamic mechanisms, dose and exposure, timing, patient characteristics, alternative explanations and the quality of the available evidence. The same interaction may also have different clinical consequences depending on the therapeutic index of the affected medicine and the susceptibility of the individual patient.

This article examines drug–drug interactions from a pharmacovigilance perspective. It focuses on recognition, assessment and safety characterisation rather than providing a comprehensive clinical pharmacology reference for interaction management. The distinction matters because the information required to manage an interaction in an individual patient is not always the same as the evidence required to establish and characterise an interaction as a population-level safety concern.

The article also distinguishes three related but different questions:

1. Did an interaction occur?
2. Did the interaction contribute to the adverse event?
3. Does the evidence indicate a broader safety risk requiring pharmacovigilance action?

These questions are related, but they should not be collapsed into a single judgement. A plausible interaction mechanism may support the assessment of an individual case without being sufficient to establish a general safety signal. Conversely, a recognised interaction may be clinically important even when an individual case contains insufficient information to determine exactly how much it contributed to the event.

Why Drug–Drug Interactions Matter in Pharmacovigilance

An interaction can change the safety profile of a medicine without changing the intrinsic toxicity of either medicine when administered alone. One medicine may increase or decrease the exposure to another, or two medicines may produce an adverse effect through complementary or additive pharmacological actions. The resulting event may therefore be attributed superficially to the medicine that was most recently started, even though the clinically important factor was an interaction between exposures.

This is particularly relevant when the affected medicine has a narrow therapeutic margin, when exposure changes are large, or when the resulting pharmacodynamic effect is clinically consequential. Examples include excessive anticoagulation, additive central nervous system depression, hypotension from combined pharmacological effects, or toxicity resulting from increased concentrations caused by inhibition of metabolism or transport. These examples illustrate mechanisms rather than establishing that every occurrence of such an event is interaction-mediated.

From a pharmacovigilance perspective, interactions also create an important distinction between product-specific risk and context-dependent risk. A medicinal product may have an established interaction with another medicine, yet the risk may arise only under particular conditions of dose, exposure, timing, organ function or patient susceptibility. Effective safety assessment therefore requires more than identifying the interacting products; it requires understanding the circumstances under which the interaction becomes clinically relevant.

The Conceptual Basis of a Drug–Drug Interaction

A drug–drug interaction occurs when the effects of one medicinal product are altered by another medicinal product in a manner attributable to their combined use. In pharmacology, interactions are commonly classified as pharmacokinetic or pharmacodynamic, although the distinction does not always capture every clinically relevant mechanism.

A pharmacokinetic interaction changes the concentration or exposure of a medicine by affecting processes such as absorption, distribution, metabolism or elimination. For example, inhibition of a metabolic enzyme can increase exposure to a substrate medicine, while induction can reduce exposure. Transporter inhibition or induction can similarly alter concentrations, and changes in renal function or renal tubular transport can affect the elimination of susceptible medicines.

A pharmacodynamic interaction occurs when medicines influence the same physiological pathway or clinical effect without necessarily changing the concentration of either medicine. Effects may be additive, synergistic or antagonistic. Two medicines may, for example, both lower blood pressure, prolong cardiac repolarisation, impair haemostasis or depress the central nervous system. The resulting clinical event may reflect their combined pharmacological effects even when plasma concentrations are unchanged.

Some cases involve both mechanisms. A medicine may increase the exposure to another medicine while also producing an overlapping pharmacodynamic effect. The pharmacovigilance assessment should therefore avoid treating pharmacokinetic and pharmacodynamic categories as mutually exclusive explanations.

Concomitant Use Is Not the Same as an Interaction

The first interpretive safeguard is to distinguish concomitant exposure from interaction. Two medicines appearing in an individual case establishes that they were used together, if the exposure information is reliable, but it does not establish that one altered the effect of the other.

This distinction is particularly important in spontaneous reporting. Patients commonly receive several medicines because of the same underlying illness, and the adverse event may be caused by the primary disease, one of the medicines independently, another exposure, or a combination of factors. The interaction hypothesis should therefore be supported by a clinically plausible mechanism and by the temporal and clinical evidence available in the case.

A useful conceptual sequence is:

concomitant exposure → plausible interaction mechanism → relevant change in exposure or effect → compatible clinical event → assessment of contribution

Not every case will contain evidence for every step. The absence of a documented mechanism does not automatically exclude an interaction, particularly when information is incomplete, but the strength of the conclusion should reflect the evidence that is actually available.

The Pharmacovigilance Perspective

The EU pharmacovigilance framework explicitly accommodates interactions within the individual case safety-reporting process. GVP Module VI addresses reporting of drug–drug, drug–food, drug–device and drug–alcohol interactions and provides for the interacting medicines to be characterised accordingly in the safety report. citeturn0search18 The current EMA GVP framework identifies Module VI as the module governing collection, management and submission of suspected adverse reactions, while the wider quality-system framework requires pharmacovigilance processes to be designed and controlled so that they support the objectives of pharmacovigilance. citeturn0search7turn0search17

This regulatory context is important because interaction information is not merely an optional clinical annotation. Where an interaction is suspected, the relevant medicines and the resulting adverse reaction need to be represented accurately in the safety information so that the case can be interpreted appropriately and, where warranted, contribute to aggregate safety assessment.

The remainder of the article therefore moves from the pharmacological mechanisms of interaction to the practical assessment of individual cases and then to the higher-level question of when interaction evidence becomes a pharmacovigilance signal or supports regulatory action.

Pharmacological Mechanisms and Clinical Context

Once an interaction is suspected, the assessment becomes more useful when the mechanism is translated into the clinical exposure and effect actually observed. A pharmacological interaction is not itself an adverse reaction; it is a mechanism that may change the probability, severity or character of an adverse reaction. The safety assessment therefore needs to connect mechanism with the patient's exposure and clinical course.

Pharmacokinetic Interactions

Pharmacokinetic interactions alter the concentration or exposure of one or more medicines. The relevant processes include absorption, distribution, metabolism and elimination. In practice, enzyme inhibition or induction, transporter effects and changes in renal elimination are among the mechanisms that can materially alter exposure.

The direction of the exposure change matters, but magnitude alone is not sufficient. A two-fold increase in exposure may have little clinical consequence for one medicine and substantial consequences for another, depending on the exposure-response relationship, therapeutic index and patient susceptibility. Similarly, an interaction that reduces exposure may be clinically important because of treatment failure rather than toxicity.

For pharmacovigilance assessment, the relevant question is therefore not simply whether an interaction is pharmacologically possible. It is whether the mechanism could reasonably produce the observed change in exposure under the circumstances of the case and whether that change is compatible with the reported event.

Pharmacodynamic Interactions

Pharmacodynamic interactions occur when medicines modify a common physiological or pharmacological effect without requiring a change in concentration. Additive effects are often clinically more important than the terminology suggests. Two medicines with individually acceptable effects on blood pressure, haemostasis, cardiac repolarisation or central nervous system function may produce a clinically important combined effect.

Synergistic effects can also occur, although the term should not be used merely because two medicines were present. The evidence should support an interaction beyond the independent effects of the individual exposures. Conversely, antagonistic interactions may reduce therapeutic effect rather than produce an adverse reaction and can still have pharmacovigilance relevance when treatment failure has safety consequences.

Exposure, Dose and Therapeutic Window

Interaction assessment should consider the actual exposure circumstances. Dose, dosing interval, duration, adherence, formulation, route of administration and changes in treatment can all influence whether an interaction becomes clinically relevant. Organ function and other patient characteristics may further modify exposure.

The therapeutic window is particularly important. When the affected medicine has a narrow therapeutic index, a relatively modest change in exposure may have disproportionate clinical consequences. Where the therapeutic margin is wide, the same pharmacokinetic change may have little observable effect. This is one reason why an interaction cannot be assessed solely from a pharmacology database or a generic interaction warning.

Timing and Temporality

Temporal information is central to an interaction assessment. The interacting medicine must generally have been present at a time when its proposed effect could influence the exposure or pharmacological action of the affected medicine. The expected onset and offset of inhibition, induction or pharmacodynamic effects should therefore be considered rather than relying only on the dates of treatment initiation.

For example, an interaction caused by enzyme inhibition may become relevant relatively quickly, whereas enzyme induction may require time to develop and may persist after the inducer is stopped. The expected time course is therefore part of the causal evidence. A temporal relationship that is incompatible with the proposed mechanism weakens the interaction hypothesis, although incomplete exposure information may limit the conclusion.

Assessing an Interaction-Related Individual Case

An individual case should be assessed as a clinical evidence problem rather than as a database coding exercise. A useful sequence is to establish the exposure history, identify the proposed mechanism, characterise the adverse event, examine the temporal relationship, evaluate alternative explanations and determine the likely contribution of the interaction.

The assessment should distinguish between evidence that the medicines were taken together and evidence that their combined use altered the patient's risk. This distinction becomes particularly important when several medicines were introduced simultaneously or when the underlying disease itself can produce the reported event.

The Exposure History

The reviewer should establish, as far as the case permits, which medicines were taken, at what doses, for how long, and whether treatment changed before the event. Over-the-counter medicines, supplements, foods and other relevant exposures may also matter. A case that records only the two products named in a spontaneous report may omit the exposure that actually explains the event.

The timing of discontinuation is also relevant. If the suspected interaction depends on continued enzyme inhibition, for example, the effect may not disappear immediately when the interacting medicine is stopped. Conversely, an event that begins after the interaction mechanism should have resolved may require another explanation.

The Clinical Event

The reported event should be characterised sufficiently to determine whether it is compatible with the proposed interaction. Laboratory findings, vital signs, diagnostic investigations, severity, treatment and clinical outcome can provide evidence that strengthens or weakens the hypothesis.

A pharmacodynamic interaction should generally produce an event consistent with the combined pharmacological effects. A pharmacokinetic interaction should produce an event compatible with the consequences of the altered exposure. The more specific the clinical phenotype, the more informative the mechanistic assessment may become.

Alternative Explanations

Alternative causes are particularly important because patients receiving multiple medicines often have multiple potential explanations for an adverse event. The underlying disease, comorbidities, other medicines, alcohol or recreational exposure, infection, dietary factors and laboratory abnormalities may all be relevant depending on the event.

The purpose is not to eliminate every conceivable alternative. Rather, the reviewer should identify realistic competing explanations and consider whether the interaction hypothesis explains the available evidence better than those alternatives. A well-established interaction mechanism does not automatically make it the cause of every event occurring during concomitant treatment.

Dechallenge and Rechallenge

Dechallenge and rechallenge can provide additional evidence, but neither should be treated as an automatic proof of interaction. Improvement after stopping one medicine may reflect withdrawal of the interacting medicine, withdrawal of the affected medicine, treatment of the event, or the natural course of the condition. Rechallenge can be informative when it reproduces a compatible event under controlled circumstances, but deliberate rechallenge is generally a clinical decision rather than a routine pharmacovigilance test.

The interpretation should therefore consider which exposure was changed, whether the proposed mechanism predicts the observed timing, and whether other factors changed simultaneously. This is the same evidence discipline required for general causality assessment, but interaction cases add the question of whether the combined exposure changed the risk.

Documenting the Interaction Hypothesis

The case record should allow a later reviewer to understand why an interaction was suspected and how it was assessed. Where an interaction is suspected, the interacting medicinal products and the resulting adverse reaction should be represented consistently with the applicable ICSR data standards. GVP Module VI specifically addresses coding and characterisation of suspected drug interactions in individual case reports. citeturn0search19

Documentation should distinguish what was reported by the primary source from what was subsequently inferred during medical review. A statement that an interaction was "possible" should not be transformed into a confirmed mechanism unless the evidence supports that stronger conclusion.

The result of the individual case assessment is therefore not necessarily a binary answer. It may be more appropriate to conclude that an interaction is established, plausible, unsupported, unlikely, or indeterminate, depending on the available evidence and the organisation's applicable assessment framework. The terminology used should remain consistent with the relevant procedures and should not imply a level of certainty that the evidence cannot support.

The individual case is only the first level of assessment. Once multiple reports or other evidence point toward a recurring interaction pattern, the question changes from whether the interaction plausibly contributed to a particular event to whether the total evidence supports a broader safety concern. That transition leads to signal assessment.

From Individual Cases to a Pharmacovigilance Signal

An interaction hypothesis becomes a broader pharmacovigilance question when similar cases, mechanistic evidence or other data suggest that the association may extend beyond an isolated patient. The presence of multiple reports does not by itself establish causality, because the same interaction may be suspected repeatedly simply because the medicines are commonly co-prescribed. Signal assessment therefore requires integration of the interaction mechanism with the clinical pattern, exposure and alternative explanations.

Case Series and Pattern Recognition

A case series can reveal features that are difficult to appreciate from an individual report. These may include a consistent latency, a characteristic clinical phenotype, recurrence after changes in exposure, concentration measurements, a dose relationship, or repeated occurrence in patients receiving the same combination.

The strength of a case series depends on the quality and independence of its cases. Duplicate reports, reports generated from the same underlying patient, stimulated reporting and selective publication can distort the apparent frequency or consistency of an interaction. Conversely, apparently heterogeneous cases may represent a genuine interaction if the mechanism and clinical consequences are coherent.

The reviewer should therefore ask whether the cases form a meaningful pattern rather than simply counting how many contain the same pair of medicines.

Disproportionality and Exposure

Spontaneous-reporting databases can be useful for detecting unusual reporting patterns, but disproportionality does not measure incidence and does not by itself establish an interaction. The interpretation can be especially difficult for common drug combinations, because the medicines may be frequently used together for reasons unrelated to the adverse event.

Exposure information is consequently valuable. If the number of patients exposed to a combination can be estimated, an apparent increase in reporting can be considered in a more informative context. Even then, observational exposure data can be affected by confounding, missing information and differences in prescribing patterns.

The appropriate conclusion is therefore proportionate to the evidence. A statistical signal may justify further investigation without demonstrating that the interaction caused the observed events.

Mechanistic and External Evidence

Mechanistic evidence can substantially strengthen an interaction hypothesis when it is consistent with the observed clinical pattern. Relevant evidence may include pharmacokinetic studies, pharmacodynamic studies, concentration measurements, experimental work, clinical pharmacology literature and established knowledge of enzyme or transporter pathways.

External evidence should be interpreted in context. An interaction demonstrated under experimental conditions may not occur to the same extent in routine clinical use, while an interaction that is theoretically plausible may have little clinical relevance at therapeutic doses. The pharmacovigilance assessment should therefore connect the external evidence to the actual exposure conditions and outcomes represented in the safety data.

Signal Validation and Characterisation

Within the EU pharmacovigilance framework, signal management is a structured process rather than a declaration that an association has been proved. GVP Module IX describes scientific and quality aspects of signal management and applies to organisations involved in EU signal management. citeturn0search20

For an interaction signal, validation should establish whether the available evidence supports a reasonable basis for further evaluation. Characterisation then considers the clinical importance, affected population, seriousness, consistency, mechanism, strength of evidence, uncertainty and potential regulatory implications.

A useful evidence chain is:

interaction hypothesis → case evidence → pattern recognition → mechanistic evidence → external evidence → clinical relevance → regulatory assessment

Each step answers a different question. A strong mechanism can support plausibility, but it does not replace clinical evidence. A cluster of cases can identify a pattern, but it does not establish incidence. A statistical association can prioritise investigation, but it does not prove causality.

Regulatory Assessment of Interaction Risks

When an interaction is sufficiently supported, the regulatory question is how the risk should be reflected in the benefit–risk assessment and product information. The appropriate action depends on the nature and magnitude of the risk, the affected population, the availability of alternatives and the degree to which the risk can be managed.

Potential measures can include changes to contraindications, warnings and precautions, dosing recommendations, monitoring requirements or other risk-minimisation measures. The appropriate regulatory wording should reflect the evidence and should distinguish established interaction information from emerging or uncertain evidence.

An interaction may also affect the interpretation of other pharmacovigilance data. For example, an apparent increase in an adverse event after a new medicine is introduced may be explained partly by changes in concomitant treatment. Conversely, a new interaction may increase the occurrence of an established adverse event without representing a new intrinsic toxicity of either medicine. These possibilities should be considered during aggregate safety evaluation.

Interaction Information in Individual Case Safety Reports

GVP Module VI contains specific requirements and guidance for reports of suspected drug interactions. For drug/drug interactions, the suspected interacting medicines are represented in the ICSR data and characterised as interacting in the applicable E2B fields. citeturn0search19 This is more than a technical coding detail: accurate representation allows interaction-related cases to remain identifiable during subsequent case review and aggregate assessment.

The same regulatory framework distinguishes drug/drug interactions from interactions involving food, devices, alcohol or other non-drug compounds. The information structure differs according to the nature of the interaction and the applicable reporting standard. The case narrative remains important where structured fields cannot capture the complete clinical context.

Interaction Risk Across the Product Lifecycle

Interaction risks can emerge or change during development and after authorisation. New medicines may be introduced into clinical practice, prescribing patterns may change, additional indications may broaden exposure to particular patient populations, and new evidence may identify previously unrecognised mechanisms.

Lifecycle surveillance should therefore consider whether the known interaction profile remains appropriate for the populations and combinations encountered in practice. A risk that is adequately controlled in one population may become more important when the medicine is used in patients with different comorbidities, organ impairment or concomitant treatment patterns.

This is one reason interaction assessment should not be treated as a static entry in a prescribing database. Pharmacovigilance is concerned with whether the totality of evidence continues to support the existing characterisation of the risk.

Integrating Interaction Evidence With Benefit–Risk Assessment

The final regulatory interpretation should consider both the magnitude of the interaction-related risk and the therapeutic benefit of the affected medicines. A clinically important interaction does not automatically mean that a medicine's overall benefit–risk balance is unfavourable. The relevant question is whether the risk can be characterised and managed sufficiently within the authorised conditions of use.

This requires an explicit distinction between three levels of conclusion:

• mechanistic plausibility — the pharmacology provides a credible explanation;
• clinical evidence — cases or studies demonstrate an association under relevant exposure conditions;
• regulatory significance — the evidence is sufficient to justify a change in product information, monitoring, risk minimisation or another regulatory action.

Moving from one level to the next requires additional evidence and judgement. Keeping the levels separate prevents both overinterpretation of weak interaction signals and under-recognition of clinically important risks.

The next section applies this framework to practical scenarios and considers how organisations can govern interaction surveillance, document decisions and demonstrate control during inspection.

Practical Application

The principles become clearer when interaction assessment is applied to cases in which the evidence differs in strength. The following scenarios are illustrative rather than reports of actual inspection findings or individual patients. Their purpose is to show how an experienced pharmacovigilance reviewer separates concomitant exposure, interaction evidence and regulatory significance.

Scenario 1: Increased Exposure With a Compatible Toxicity

A patient receiving a medicine with a narrow therapeutic margin begins a second medicine known to inhibit the principal metabolic pathway of the first. Shortly afterwards, the patient develops a toxicity consistent with excessive exposure. The timing is compatible with the inhibitory mechanism, and the event improves after the interacting medicine is withdrawn while treatment of the toxicity is provided.

This case contains several mutually reinforcing elements: a plausible mechanism, compatible timing, a susceptible medicine, a compatible clinical event and a plausible change in exposure. It would nevertheless be inappropriate to describe the interaction as proven solely because the sequence is persuasive. The strength of the conclusion should reflect the actual evidence, including whether concentrations or other objective measures are available and whether competing explanations have been assessed.

Scenario 2: Two Medicines With an Additive Pharmacodynamic Effect

A patient receives two medicines that independently lower blood pressure and subsequently develops symptomatic hypotension. No material pharmacokinetic interaction is known. The event may nevertheless be interaction-related because the combined pharmacodynamic effects can exceed the patient's tolerance.

The assessment should consider baseline blood pressure, doses, timing, other antihypertensive treatment, dehydration, acute illness and other plausible causes. The absence of a pharmacokinetic mechanism does not exclude a pharmacodynamic interaction, but the presence of two medicines with overlapping effects does not by itself establish one.

Scenario 3: A Plausible Interaction With Weak Clinical Evidence

A spontaneous report describes an adverse event during treatment with two medicines for which an interaction is theoretically possible. However, the report lacks dose information, treatment dates are uncertain, the patient has several competing risk factors and the event is not particularly characteristic of the proposed mechanism.

The correct response is not to force the case into a confirmed interaction category. The interaction hypothesis can be documented and may justify follow-up, but the uncertainty should remain visible. A weak individual case may still become informative if similar cases subsequently establish a coherent pattern.

Common Analytical Errors

Several errors repeatedly weaken interaction assessment.

Treating concomitant use as proof of interaction. Two medicines taken together establish exposure overlap, not necessarily a pharmacological interaction.

Using a theoretical mechanism without clinical context. A CYP, transporter or receptor mechanism may establish plausibility but does not demonstrate that a clinically meaningful interaction occurred in the patient.

Ignoring dose and exposure. Interaction magnitude can depend strongly on dose, duration, adherence, organ function and other exposure conditions.

Overinterpreting dechallenge. Improvement after withdrawal is not specific for an interaction when several treatments or other factors changed at the same time.

Treating disproportionality as incidence or causality. A disproportional reporting pattern can identify a hypothesis but does not establish the frequency or cause of an interaction.

Ignoring prescribing context. Commonly co-prescribed medicines create substantial opportunities for coincidental associations and confounding by indication.

Failing to distinguish evidence levels. A recognised interaction in a reference source, a plausible interaction in an individual case, and a validated pharmacovigilance signal are different propositions and should not be presented as equivalent.

Governance and Quality-System Considerations

An organisation should be able to demonstrate that interaction information is handled consistently across case processing, medical review, signal management and aggregate safety assessment. The exact operational controls are organisation-specific, but the underlying objective is traceability: an interaction hypothesis identified during case processing should not disappear when the case enters later stages of pharmacovigilance evaluation.

Useful controls can include appropriate coding conventions, targeted medical review, quality-control checks, structured search strategies, reconciliation between relevant safety datasets and documented escalation criteria. These are examples of operational practice rather than universal legal requirements. Their design should be proportionate to the organisation's products, data sources and risk profile.

The quality system should also make clear who is responsible for evaluating emerging interaction information and how conclusions are communicated to relevant safety governance bodies. Where an interaction has implications for product information or risk management, the decision and its supporting evidence should be traceable through the applicable governance process.

Inspection Considerations

An inspector evaluating interaction management could reasonably examine whether the organisation can demonstrate how suspected interactions are recognised, assessed, coded, followed up and incorporated into aggregate safety evaluation. The inspection focus would be on evidence and control rather than on whether every possible interaction has been predicted in advance.

Illustrative inspection questions include:

• Can the organisation identify how suspected drug interactions are recognised during case processing?
• Are interacting medicinal products represented consistently in the safety database and applicable ICSR fields?
• Can the organisation explain how interaction hypotheses are medically assessed?
• Are important interaction patterns considered during signal detection and aggregate safety review?
• Can decisions to escalate, monitor or close an interaction concern be reconstructed from contemporaneous evidence?
• Where interaction information changed product information or risk management, is the evidence and decision-making traceable?

These are illustrative inspection questions, not a list of prescribed inspection findings. An organisation may use different controls while still meeting applicable regulatory requirements.

Key Takeaways

Drug–drug interaction assessment in pharmacovigilance is a layered evidence problem. The starting point is concomitant exposure, but the assessment must determine whether a plausible mechanism could have changed exposure or pharmacological effect, whether the clinical event is compatible with that mechanism, and whether competing explanations have been adequately considered.

At the individual-case level, interaction evidence should be documented with an appropriate degree of uncertainty. At the aggregate level, recurring cases should be integrated with mechanistic, clinical, epidemiological and exposure evidence rather than evaluated by case counts or disproportionality alone.

The regulatory significance of an interaction is a further judgement. Evidence that supports mechanistic plausibility is not necessarily sufficient to justify regulatory action, while a well-characterised clinically important interaction may require changes to product information, monitoring or risk minimisation. Maintaining these distinctions allows pharmacovigilance professionals to recognise meaningful interaction risks without overstating uncertain evidence.

References

1. European Medicines Agency. Guideline on good pharmacovigilance practices (GVP), Module VI: Collection, management and submission of reports of suspected adverse reactions to medicinal products (Rev. 2). EMA/873138/2011 Rev. 2. EMA.
2. European Medicines Agency. GVP Module VI Addendum I: Duplicate management of suspected adverse reaction reports. EMA/405655/2016. EMA.
3. European Medicines Agency. GVP Module IX: Signal management (Rev. 1). EMA/827661/2011 Rev. 1. EMA.
4. European Medicines Agency. Guideline on good pharmacovigilance practices (GVP). Current GVP framework and modules. EMA.
5. International Council for Harmonisation. E2B(R3): Electronic Transmission of Individual Case Safety Reports (ICSRs), Implementation Guide. ICH.
6. International Council for Harmonisation. E2D(R1): Post-Approval Safety Data: Definitions and Standards for Management and Reporting. ICH.
7. Uppsala Monitoring Centre. WHO Drug Dictionary and pharmacovigilance resources. UMC.

Regulatory Note

This article explains the EU pharmacovigilance framework and associated scientific principles. References to GVP describe the applicable regulatory guidance and, where the underlying text uses legally binding language, should be distinguished from recommendations for implementation. Operational examples, inspection questions and analytical approaches in this article are illustrative unless explicitly identified as regulatory requirements.

The EU pharmacovigilance framework continues to evolve. EMA states that amendments introduced by Commission Implementing Regulation (EU) 2025/1466 affect the pharmacovigilance framework and that relevant GVP modules will be updated. In particular, the updated signal-detection obligations for marketing authorisation holders apply following the entry into force of that Implementing Regulation, while EMA has indicated that corresponding GVP Module IX updates will follow. Users should therefore verify the current legislation, GVP text and EMA procedural guidance when applying this article to a live regulatory decision. citeturn0search0turn0search1