18.iii. LISTEDNESS/ EXPECTEDNESS ASSESSMENT – Guidelines and Examples

The determination of whether an ADR is or is not expected is not an exact science; there are many grey areas. A decision in many instances will have to be based on clinical evaluation of inadequate case information. Evaluation of expectedness will probably remain subject to high variability between assessors. 

The Council for International Organizations of Medical Sciences (CIOMS) provided guidelines to improve consistency and reproducibility in the process. 

Determining whether a reported reaction is expected or not involves two levels of inquiry: 

1st Question

• If the actual term is not listed in the RSI – is a synonym listed?

• Is the AE/ADR mentioned in the appropriate section of the reference safety information (RSI)? Any reaction not mentioned is supposedly new and therefore unexpected.

2nd Question

• Based on the medical information contained in the case report, is the AE different regarding its nature, severity, specificity or usual outcome relative to the term or description used in the RSI?

Guidelines and Examples:

1. AEs are expected if their nature, severity, specificity and frequency are consistent with the information in the relevant Reference Safety Information (RSI).

• If reported event specificity is clinically important than expected term then medical judgement should be used.

– Example: cerebral thromboembolism and cerebral vasculitis would both be unexpected (by virtue of greater specificity) if the RSI document only listed cerebral vascular accidents 

• When a term mentioned in the reference information encompasses situations with distinguishable and recognised levels of severity, a significantly more serious case should be considered unexpected 

– Example: fulminant hepatitis should not be considered expected if ‘‘liver injury’’ is mentioned in the reference information. 

• If reported event frequency/duration vary from expected RSI – A case will usually be considered ‘‘unexpected’’ if the RSI lists an ADR which is specified as transient or acute, but it persists in the new case. 

– Example: if the RSI document refers to acute elevated liver function tests, a raised level lasting three months would be unexpected. Thus, prolonged cholestatic liver injury should not be considered expected when acute cholestatic liver injury is mentioned in the RSI, since prolonged forms may not be reversible. 

2. AEs associated with specific indications, dose specifications, dose forms or circumstances are only expected for those specific conditions (i.e. “situational” expectedness)

3. Generally, further anatomical specification or synonym of a expected AE should be considered expected.

Example: Further anatomical specification:

• left-sided chest pain is equivalent to chest pain; it should not be assessed as unexpected if chest pain is expected;
• fibrosis of upper left lobe is equivalent to lung fibrosis
• if arteritis is expected, temporal arteritis should be considered unexpected due to the associated additional risks and poorer prognosis. 

4. Further histological or diagnostic specification does not make an expected ADR unexpected 

– Example: A liver biopsy shows hepatic necrosis (expected) with the presence of eosinophils (not mentioned in RSI document). Though presence of eosinophils is not present in RSI documents hepatic necrosis need to be considered as expected as specificity is not clinically important. 

5. If a diagnosis is an expected ADR, then the signs and symptoms which comprise the diagnosis are also considered to be expected, when they are reported as associated. For example, if anaphylactic reaction is labeled, then a report of a patient who experienced hypotension, wheezing, and urticaria together would be considered an expected event. 

6. The reverse is not true however; a diagnosis relating to a group of symptoms or signs which are each individually labeled would not usually be considered expected. A reported anaphylactic reaction is unexpected if only isolated hypotension, or wheezing, or urticaria are labeled. 

7. If a diagnosis or syndrome is expected, if the usually accompanying signs and symptoms are reported in the absence of a clear diagnosis (i.e., as one or more isolated signs and symptoms), those terms should not be considered as expected unless already in the RSI. It is impossible to ascertain that their appearance alone or together necessarily reflects a mechanism similar to that of a labelled diagnosis (e.g., isolated nausea, or asthenia, or gastralgia, when liver injury is labelled; isolated pallor, or hypotension or pruritus when anaphylactic reaction is labelled).

8. If the label lists an AE which is specified to be transient, but it persists in the new case, the case is unlabeled and should be reported: E.g., prolonged elevated liver function tests, when labelling states transient elevated liver function tests

9. Unless the RSI specifies a fatal outcome, then the case should be considered as unexpected as long as there was an association between the adverse reaction and the fatality. 

10. A fatal outcome to a suspected ADR should not be mentioned in the RSI unless it has been reported to occur and is thought to be causally related to the ADR. 

11. In the absence of special circumstances, once the fatal outcome is itself expected (labeled/listed), reports involving fatal outcomes should be handled as for any other serious suspected ADR in accord with appropriate regulatory requirements. 

12. AE with causality disclaimer – are considered expected (including those listed in a frequency table). E.g., “The following adverse events have been reported in association with the drug, but a causal relationship has not been established” However: if the AE disclaimer states that the events have been reported but not been considered drug-related (negative causality statement), then the AE should be assessed as unexpected

13. If an AE/ADR has been reported only in association with an overdose, then that same AE/ADR at usual doses should be considered unexpected.

14. If an AE/ADR occurs in a different indication, it is assessed unlabelled unless it is described in the adverse events section applicable for a specific indication or patient population.

15. If an AE/ADR follows a different route/formulation of drug administration, the event labelled for one presentation cannot be considered labelled for another presentation.

16. Drug exposure during pregnancy: Abortion, stillbirth, congenital abnormalities and maternal/new-born hazards are considered unlabelled unless explicitly specified in the RSI.

17. Pregnancy/drug exposure in utero or normal babies are considered labelled.

18. Drug abuse, drug dependence, maladministration: unlabelled unless explicitly specified in the relevant sections of the RSI.

19. Lack of efficacy and resulting signs or symptoms are considered labelled events. However, if treatment directly exacerbates the treated condition then exacerbation is considered unlabelled unless specifically mentioned in the RSI.

20. If an AE/ADR is due to a specific drug-drug interaction, it is considered labelled but only in the context of the drug-drug interaction.

21. A laboratory value that is representative of the definition of the diagnosis (labelled in the RSI) is expected, for example, if the expected term is thrombocytopenia then decreased platelet count is expected. While PI is reporting decrease platelet count but RSI says thrombocytopenia, it is expected even though the terms are different but the meaning is the same.

22. Symptoms/signs listed in the RSI are expected but diagnoses not listed in the RSI equals unexpected. Diagnoses are not equivalent to lab values for example,: if the expected term is raised liver enzymes (meaning if the RSI states exactly this), then hepatitis (diagnosis) is not expected because it is not listed in the RSI.

• Only a medic can provide a diagnosis.

Sources:

CIOMS V – Current Challenges in Pharmacovigilance:Pragmatic Approaches.

CIOMS VI – Management of Safety Information from Clinical Trials.