The Information Component (IC) is a measure of the strength of the quantitative dependency between specific drug and specific ADR. It is a logarithmic measure of disproportionality used to evaluate strength of association between drug and ADR. It is mathematically expressed as:
Where, p (x) is probability of specific drug ‘x’ being listed in a case report; p (y) is probability of specific ADR ‘Y’ being listed in a case report and p (x, y) is probability of specific drug- ADR combination being listed in a case report.
The value of information component depends on number of case reports with drug ‘x’, on number of case reports with ADR ‘y’, on number of case reports with combination of specific drug and ADR and total number of case reports. It can also be stated that information component is the logarithm of ratio of observed rate of drug-ADR reporting to its expected rate with the assumption of null hypothesis.
- If a particular drug-ADR combination is reported more often than expected from the rest of the database then the value of IC will be positive.
- For no quantitative dependency the value of IC will be zero, while the combination is occurring less frequently than statistically expected it will be negative.
- The higher the IC value, more the combination stands out from the background.
The system is able to handle the fact that there are more suspected ADRs than cases (that is, a total of more than one reported ADR per patient); and that more than one drug may be involved in any given case. The change in value of IC may occur with the addition of new data. The value of IC is less likely to fluctuate when it is calculated large numbers. The standard deviation for each IC provides a measure of the robustness of the value
I saw some studies adds 0.5 of correction value to the probaility of target drug or ADR reported when they calculate IC. Isn’t that the correction value does impact the result of the study?