Pharmacovigilance & Drug Safety

Signal detection:

Signal detection involves a range of techniques. The WHO defines a safety signal as: "Reported information on a possible causal connection between an adverse event and a drug, the relationship being unknown or incompletely documented previously". Commonly more than a single report is required to generate a signal, depending upon the event and quality of the information available.

Data mining pharmacovigilance databases is one approach that has become increasingly popular with the availability of extensive data sources and low-cost computing resources. The data sources may be owned by a pharmaceutical company, a drug regulatory authority, or a large healthcare provider. Individual Case Safety Reports (ICSRs) in these databases are retrieved and converted into structured format, and statistical methods (usually a mathematical algorithm) are applied to calculate statistical measures of association. If the statistical measure crosses an arbitrarily set threshold, a signal is declared for a given drug associated with a given adverse event. All signals deemed worthy of exploration, require further analysis using all available data in an attempt to confirm or refute the signal. If the analysis is inconclusive, extra data may be needed such as a post-marketing observational trial.