Signal Detection in Pharmacovigilance

Addendum to GVP Module IX on Signal Management

• The aim is to provide guidance on the detection of potential signals of adverse drug reactions.

Routine Periodic Summaries

• Statistical summaries of information received for each 'drug-event' combination should be produced in a routine periodic fashion to enable identification of potential signals.

Limitations of Addendum

• The Addendum does not provide details of specific implementations developed for use in EudraVigilance.

Criteria for Selecting 'Drug-Event' Combinations

• The criteria for selecting 'drug-event' combinations (DECs) for further investigation should be objectively defined to ensure transparency and consistency.

Purpose of Routine Scanning

• The purpose of routine scanning of accumulating data is to detect potential signals of adverse drug reactions.

Considerations in Detecting Potential Signals

• A key consideration in detecting potential signals is the integration of multiple methods for efficient routine signal detection.
• Integration of multiple methods is important to increase the sensitivity and specificity of signal detection.

Decision Process for Signal Detection

• The decision process for signal detection should rely on data summaries and other relevant information such as pharmacological and toxicological data.
• The decision process aims to prompt further investigation and assessment of potential signals.

Indicators of Adverse Consequences

• A potential indicator of a change in quality or use of a medicinal product with adverse consequences is a change in the frequency or pattern of adverse event reporting.

Purpose of Statistical Methods

• The purpose of statistical methods is to facilitate the detection of potential signals when the database accrual is too large for individual scrutiny of all incoming Individual Case Safety Reports (ICSRs).

Monitoring of Signal Detection

• Any system of signal detection should be monitored for its performance and validity.

Influence of Medicinal Products on Implementation

• The nature of the medicinal products in the dataset may influence the specific details of implementing proposed methods.

Disproportionate Reporting Statistics

• Disproportionate reporting statistics take the form of measures of the strength of association between a drug and an event, such as the proportional reporting ratio (PRR) or the reporting odds ratio (ROR).