Chapter 3

The Methodology of Causality Assessment

WHO-UMC Causality Categories

Certain
  • Plausible time relationship
  • Cannot be explained by disease or other drugs
  • Positive dechallenge
  • Positive rechallenge (mandatory)
Probable / Likely
  • Reasonable time relationship
  • Unlikely due to disease or other drugs
  • Positive dechallenge
  • Rechallenge not required
Possible
  • Reasonable time relationship
  • Could also be explained by disease or other drugs
  • Dechallenge information may be lacking or unclear
Unlikely
  • Improbable (but not impossible) time relationship
  • Disease or other drugs provide plausible explanation
Conditional
  • Event reported but more data needed for assessment
  • Additional investigation ongoing
Unassessable
  • Report suggests an ADR
  • Cannot be judged — insufficient or contradictory information
  • Data cannot be supplemented or verified

Two Philosophies: Global Introspection vs. Algorithmic

🧠 Global Introspection

Relies on the holistic judgment of experienced clinicians who evaluate the case based on their medical knowledge, clinical experience, and the totality of the evidence.

Strength: Captures unique clinical nuances an algorithm might miss
Weakness: Highly subjective; poor reproducibility between reviewers

📊 Algorithmic (Naranjo)

A structured questionnaire-based method that assigns a numerical score to the likelihood of an adverse drug reaction, producing a standardized output.

Strength: Reproducible, transparent, easy to audit
Weakness: May oversimplify complex clinical scenarios

The Naranjo Algorithm — Scoring Questions

#QuestionYesNoDon't Know
1Are there previous conclusive reports on this reaction?+100
2Did the adverse event appear after the suspected drug was given?+2-10
3Did the adverse reaction improve when the drug was discontinued or a specific antagonist given?+100
4Did the adverse reaction reappear when the drug was readministered?+2-10
5Are there alternative causes that could have caused the reaction?-1+20
6Did the reaction reappear when a placebo was given?-1+10
7Was the drug detected in blood (or other fluids) in toxic concentration?+100
8Was the reaction more severe when the dose was increased, or less severe when decreased?+100
9Did the patient have a similar reaction to the same or similar drug in any previous exposure?+100
10Was the adverse event confirmed by any objective evidence?+100

Score Interpretation

≥ 9: Definite
5–8: Probable
1–4: Possible
≤ 0: Doubtful

The French Imputability Method (Dangoumau/Bégaud)

Intrinsic Accountability (I)

Determined by combining two sub-scores from the specific case:

Chronological Score (C)

Time to onset (C1 to C3), evolution after dechallenge, and response to rechallenge. Evaluates whether the temporal pattern is consistent with causation.

Semiological Score (S)

Clinical and extra-clinical signs (S1 to S3), and the exclusion of alternative causes. Evaluates the clinical plausibility of the drug-event relationship.

I = f(C, S)
Combined into a final intrinsic score
+

Extrinsic Accountability (B)

Benchmarks the case against the state of published medical literature:

B4 Notorious

Reaction listed in standard reference books (e.g., VIDAL, Martindale)

B3 Published

Reported in the literature but not yet "notorious"

B2 Sparse

Reported only once or twice in the literature

B1 Novel

Completely new reaction — no prior reports found

French Method — Key Differentiator

By separating intrinsic (case-specific) from extrinsic (literature-based) accountability, the French method prevents well-known reactions from receiving automatic high scores without proper case-level evidence. It is mandatory for spontaneous reporting in France.