Drug Safety : Causality Assessment scales

Causality Assessment is Complex

Imputation/Causality is a difficult process due to the complex nature of adverse events, multiple treatments and individual clinical variability. Causality inference obtained by the clinical judgement of an expert's panel or "global introspection" is the most common approach for individual causality assessment of adverse drug reports. However the process of global introspection is know to be subjective. To overcome this there has been several (nearly 30) decision support algorithms which are available to make an explicit and reproducible assessment of causality.

Different scales used for causality assessment are as follows:

• WHO assessment scale
  
• Naranjo’s scale
  
• European ABO system
  
• Yale logarithm
  
• Spanish imputation system
  
• Kramer scale
• Karch and Lasagna’s scale
  

Naranjo’s scale

Assessment schema

• > 9 = definite
  
• 5-8 = probable
  
• 1-4 = possible
  
• 0 = unlikely
  

Is there a gold standard ?

Unfortunately no. Global introspection and causality algorithms compare poorly against each other in terms of the decisions reached. However decision support algorithms such as Bayesian networks can be used to generate data on signal detection, indicating a potential safety issue in future there by supporting robust assessment of causality.