Data Analysis for Adverse Drug Events Reported to CVM
When an ADE report arrives at FDA’s Center for Veterinary Medicine (CVM), it is logged into the Center’s Document Control Unit, assigned a submission number, and forwarded to CVM’s Division of Surveillance for processing. CVM safety reviewers, who are experienced clinical veterinarians, then review the individual ADE reports. The safety reviewers enter all relevant information such as age, breed, history, pre-existing problems, and concomitant drugs into an Oracle database called the Submission Tracking and Reporting System (STARS).
The reviewers evaluate each clinical manifestation reported using a scoring system that is a modified version of the Kramer algorithm. This scoring system has the following components:
Under the scoring system, information about age, breed, gender, pre-existing conditions, and concomitant drugs are considered. A summary causality assessment score for each clinical manifestation is determined and entered into the STARS database. The summary score corresponds to the strength of the association between the drug and the clinical manifestation and ranges between -9 and +7. Clinical manifestations with summary scores of > or = 0 are considered possibly, probably, or definitely drug-related.
Every clinical manifestation (CM) is reviewed and evaluated for each of the following criteria in the Kramer algorithm.
1. Previous Experience with the Drug:
|
+1 |
CM generally recognized to occur in this species at the dosage received. | |
|
0 |
CM is not generally recognized to occur in this species at the dosage received, but has been previously reported in veterinary and/or human medicine. | |
|
Or, drug has limited accumulated clinical experience (time and/or quantity marketed). | ||
|
-1 |
CM previously unreported and drug has substantial accumulated clinical experience (time and/or quantity marketed). |
|
+2 |
There is no good candidate or no change in a candidate which can explain the CM, exclusive of drug administration. | |
|
0 |
An alternative candidate(s) exists, but not a good one(s) which can well explain the CM. | |
|
CM commonly occurs spontaneously in this type of patient and situation, usually in the absence of any recognizable alternative candidate(s). | ||
|
-1 |
There is a good candidate or a change in a candidate which can well explain the CM, exclusive of drug administration. |
|
+1 |
Timing was consistent and as expected for this type of CM to this drug. | |
|
0 |
Do not know what timing to expect. | |
|
-2 |
|
Timing was inconsistent for this type of CM to this drug. |
|
+1 |
CM is clearly a dose-related type of manifestation, and there is unequivocal evidence that the amount of drug received was an overdose for this animal. | |
|
0 |
CM is not a dose related type of manifestation or there in no evidence of an overdose. |
5. Dechallenge (whether the problem disappears after withdrawal of the drug):
|
+1 |
CM diminished or disappeared after discontinuation of suspect drug or administration of a specific antidote. CM is known to be dose-related, and CM diminished or disappeared after dosage reduction. | |
|
0 |
Dechallenge difficult, impossible, or inappropriate to assess. A non-specific agent or maneuver (non-antidotal) was administered that was directed against the CM and that usually produces the degree and rate of improvement observed in the case. A "non-antidotal” is a remedy that is not specifically used for treating the animal exhibiting toxic signs. For example, giving intravenous fluids might help reduce an illness because of the general effect of cooling, re hydration, and supplying needed electrolytes but this is not considered an "antidote". By contrast, an antidote is a drug that is specifically used for treating the animal with these toxic signs. CM characteristically transient and episodic, and there is no established pattern episodes (regardless of what occurs after discontinuing the drug). CM known to be dose-related, and CM did not diminish or disappear after dosage was reduced. | |
|
-1 |
|
CM did not diminish or disappear after discontinuation of suspect drug. CM improved without dechallenge and the improvement cannot be attributed to the development of tolerance. |
6. Rechallenge (whether the problem reappears when the drug is reintroduced):
|
+1 |
CM unequivocally recurred or exacerbated after rechallenge. | |
|
0 |
There was no rechallenge attempted. A non-specific agent or maneuver (non-antidotal) was administered that obscured the response of the CM. CM failed to recur or exacerbate on rechallenge, but the dosage or duration of drug administration on rechallenge was substantially less than that suspected of causing the original CM. Recurrence or exacerbation of CM was impossible to assess because it was progressing or was at a level of severity that any further increase would be difficult to appreciate. | |
|
-1 |
|
CM failed to recur or exacerbate on rechallenge. |
The component scores above are then summed to reach a final causal score:
|
Scores |
Category of Association |
|
-9 |
Not applicable |
|
-8 |
Information lacking |
|
-7 |
No conclusion |
|
-1 to -6 |
Remotely drug-related |
|
0 to 2 |
Possibly drug-related |
|
3 to 5 |
Probably drug-related |
|
6 to 7 |
Definitely drug-related |
|
-1 |
Ineffectiveness - Remotely drug-related |
|
0 to 1 |
Ineffectiveness - Possibly drug-related |
|
3 |
Ineffectiveness - Probably drug-related |
|
6 |
Ineffectiveness - Definitely drug-related |
|
||||||
 |
||||||
|
||||||