Why Causality Assessment Is the Hinge of Pharmacovigilance

Every activity covered so far in this course quietly depends on causality assessment having been done consistently somewhere upstream. Signal detection in Module 3 assumes cases coded as "related" to a drug were actually assessed as related using a defensible method, not a coder’s gut feeling. Aggregate reports summarise causality patterns across hundreds of cases. Labelling decisions rest on a body of individually assessed cases adding up to a pattern regulators trust. None of that works if causality assessment itself is inconsistent from one assessor to the next.

The core difficulty is that causality is a probability judgment, not an observable fact. Unlike a lab value or a MedDRA code, "did this drug cause this event" usually can’t be directly verified — a patient developed a rash three days after starting a new drug, but people develop rashes for other reasons too, and there’s rarely a controlled way to prove the drug did it in this specific patient. That’s exactly the gap structured causality scales exist to close: not by making the judgment purely objective, which isn’t possible, but by making it consistent — two trained assessors working from the same criteria on the same case should land on the same, or a very similar, category.

This module works through that structure in the order it’s actually applied in practice: the WHO-UMC scale and its key concepts (challenge, dechallenge, rechallenge) first, then the Naranjo algorithm as a second, complementary method, then how company and regulatory causality assessments relate to each other. From there it moves to seriousness — a related but genuinely separate judgment — and expectedness, before showing how all three combine into the SUSAR decision Module 1 introduced the vocabulary for. Every lesson works from the actual assessment criteria rather than a simplified summary, because real fluency here means being able to apply the criteria to a new case, not just recite the category names.