A Structured Methodology to Assess Safety Signal Strength and Inform Causality Assessment

Abstract

Causality assessment of safety signals observed with medicinal products is a foundational element of pharmacovigilance and regulatory practice, typically performed by a global introspection process. We have developed a novel, structured methodological framework to support the global introspection process for safety signal causality assessment. This Signal Assessment Guide (SAGe) tool was developed by AstraZeneca and is used internally, both to assess safety signal strength and to inform causality decisions related to safety signals. The term ‘safety signal’ refers to information arising from one or multiple sources, which suggests a new potentially causal association, or a new aspect of a known association, between an intervention and an adverse event. The key concept underlying the SAGe tool is that safety signal data can be reliably sorted into one of three categories: aggregate safety data, plausibility data, and case-level data. When applying the tool, an evidence grade score (Levels A, B, C, and D) is transparently assigned to the available data in each category. This information can then be summarised and presented for formal decision making regarding causality for safety signals. By using a transparent method to categorise the grade of evidence for causal association, with an option to additionally derive a quantitative strength of safety signal score, the SAGe tool can support the global introspection process for causality decisions, contributing to the quality of safety information for medicinal products provided to healthcare professionals and patients. Our anecdotal experience of using the SAGe tool at AstraZeneca is that it has resulted in more efficient and robust conversations regarding the strength of safety signals and the causality question. Wider use of the SAGe tool may bring increased levels of transparency and consistency to the evaluation of safety signals.