Related, Not Related, and the Space Between

Part of Pharmacovigilance and Drug Safety (PV) Training: Causality.

About 1 minutes. Free, in your language, no account needed.

Causality is an assessment, not a fact

Nobody can observe that a drug caused an event in an individual patient. What exists is a set of considerations that make a causal relationship more or less plausible, and a judgment made on them.

This is why causality assessments differ between reasonable people, and why both the reporter's and the company's are kept.

What goes into the judgment

Temporal relationship. Did the event occur in a plausible window after exposure? An event beginning four hours after the first dose of an immediate hypersensitivity reaction is plausible. A cancer diagnosed a week after starting a drug is not caused by that week.

Biological plausibility. Is there a known mechanism, a class effect, or prior evidence?

Dechallenge. Did the event resolve when the drug was stopped?

Rechallenge. Did it return when the drug was restarted?

Alternative explanations. Underlying disease, concomitant medication, infection, procedure.

Prior experience. Has this been seen with this product before?

The categories

Terminology varies between organizations and jurisdictions. Common schemes use certain, probable, possible, unlikely, unrelated and unassessable. Some use only related and not related.

What matters is not the number of categories but that they are applied consistently and that the scheme in use is the one the receiving authority expects.

The default in trials

For the purposes of expedited reporting from a clinical trial, a reasonable possibility of a causal relationship is the threshold. "Possible" is enough. Ambiguity resolves toward reporting, not away from it.