Pharmacovigilance methods overview 

A key component of pharmacovigilance is signal detection, which involves searching for indications that a drug may be associated with a particular adverse side effect. One common approach involves comparing how often a specific side effect is reported for one drug versus others. If it appears more frequently than expected, it may suggest a potential link. This type of analysis is known as disproportionality analysis, and it helps identify early safety concerns that may require further investigation. 

In short, disproportionality analysis aims to identify drugs that have a higher number of adverse drug reaction (ADR) reports compared to others. This is illustrated in the chart, where each dot represents a drug and the number of reports for a specific adverse event. Most drugs cluster around the expected level of reporting (marked by the red dashed line), showing no unusual pattern. However, one drug—Drug E—stands out with a significantly higher number of reports. This suggests a disproportionate signal that may warrant further attention.

It's important to note that this signal does not prove the drug causes the side effect. Rather, it acts as an early warning that helps prioritize drugs for more detailed investigation.

To quantify disproportionality, several statistical measures can be used, such as the Reporting Odds Ratio (ROR) and the Proportional Reporting Ratio (PRR). Generally, values greater than 1 suggest a potential signal, but these methods must be interpreted with caution, especially to avoid false positives. More advanced techniques have been developed to improve reliability—especially when dealing with low case numbers—and to better control for background noise in large datasets. Further detail is available in the following paper: Evans et al, 2006: Statistical Methods of Signal Detection.