Automating Signal Detection to Efficiently Manage Safety Data

In order to ensure patient safety and comply with pharmacovigilance regulations, massive amounts of safety data must be analyzed. Arithmos explains how automating pharmacovigilance systems can improve the identification of adverse events through real time data results.

Pharmacovigilance Regulations: Signal Management

Regulatory authorities and governing bodies have taken steps to improve patient safety. The European Union took action in July 2012 by implementing the first seven modules of new pharmacovigilance legislation. The changes are designed to tackle the startling statistic that 200,000 patients in the EU die each year from adverse drug reactions (ADRs).

Part of the implementation plan is to better analyze and understand data from clinical studies – especially post-market studies – to identify risks to patients. One of the more significant modules to pass in July 2012 was Module IX: Signal Management which includes requirements on the statistical analysis and systems used to detect signals. It also outlines how data should be assessed by a Qualified Person in a “timely” manner, how the process should be documented and how urgent action should be taken whenever a safety issue arises.

The Industry Need

Due to the increased regulation on safety data and the need to improve data quality, signal detection becomes very data intensive and difficult to manage. Pharmaceutical companies are searching for a signal detection solution that can produce real time results with accurate signal identification at an affordable operational cost.

Technology can facilitate the automation of signal detection to help reduce workload.

Combining Statistical Programming & Data Integration Technology

Combining statistical analysis, statistical programming and IT support under a commonly used platform like SAS is an ideal solution for automating data collection and analysis from multiple sources to implement an efficient signal detection process. In order to automate the process, a system needs to be in place to pull and analyze data from the safety database.

The identification of signal criteria and the implementation of standardized programs automates the signal detection process. It also produces structured data which speeds up the task of finalizing Eudravigilance submissions.

This approach is appropriate even for small companies, or for products with  small amounts of safety data, because automation can be done without a complex or expensive Business Intelligence platform.

Arithmos Case Study: Producing Pharmacovigilance Reports Using Data Integration Technology

A large pharmaceutical company contacted a CRO to do statistical analysis and reporting on adverse events for a suspect drug reported in individual case safety reports (ICSR). The Sponsor was using a third party adverse reporting system along with SAS to collect post-marketing adverse reactions data.

Signal Detection 2

The methodological statistician defined a Proportional Reporting Ratio for the detection of “Signals of Disproportionate Reporting (SDRs)”. To produce line listings and summary tabulation for signal detection and reporting, the Sponsor needed a set of SAS programs which allowed data to be retrieved from ARES database instantly and in a structured manner. To use these programs, SAS programming knowledge is not required because the programs are automatically set up for all the variables required. Therefore, pharmacovigilance officers do not need to have a knowledge of SAS.

The end result was an automated signal detection process which leads to better analysis, real time results, and structured and formatted data for efficient preparation of regulatory submissions.

Related Links

Arithmos Pharmacovigilance

Arithmos Pharmacovigilance Services – AERS vs Argus



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s