Application of Change Point Analysis in Post-Market Surveillance

• Activity Number: 581
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Zhiheng Xu*+ and Greg Campbell
• Companies: FDA and FDA
• Keywords: post market surveillance ; change point analysis ; cusum

Abstract

The Food and Drug Administration (FDA) launched the Sentinel Initiative in 2008 in order to increase FDA's post market surveillance capability in tracking the safety of medical products after licensure. A critical component in the Sentinel is to develop active surveillance methodologies for signal detection of an increased risk of adverse events or a possible safety outbreak related to medical products. A number of signal detection methods have been evaluated by FDA for the use in post market surveillance. In this talk, we first evaluate the use of change point analysis (CPA) in detecting anomaly signals and assessing the risk of adverse events in the post market surveillance time series data. Three CPA methods: cumulative sum (CUSUM), structural change model and Bayesian CPA will be discussed. Application of the methods will be illustrated using data from available national surveillance system.