Increased frequency algorithm

For interactive signal configurations, Oracle Empirica Signal includes an algorithm to monitor events for increased frequency.

Reporting proportion (RP) is used as a proxy for exposure adjusted reporting rates and is defined as the number of cases for a product-event combination divided by the total number of cases for the product within the time period.

Increased frequency counts are based on the Data configuration for 2D runs specified in the signal configuration. Increased frequency is computed for all product-event combinations, and Oracle Empirica Signal sets a YES flag for the product-event combinations that have at least three new reports in the most recent 12 months and meet at least one of the following conditions:

• At least a five-fold increase in the RP for the most recent 12 months, compared to the cumulative RP excluding the most recent 12 months.
• At least a two-fold increase in the RP for the most recent 12 months and for the time period of 13 to 24 months ago, both compared to the cumulative RP excluding the most recent 24 months.

The Product-Event Combinations and the Products tables maintain information for increased frequency for three time periods, based on the initial reporting or received date of each case, and follow-up reports are not considered. For example, Argus Signal Management uses the following time periods:

• 1-12 months
• 13-24 months
• 25 or more months