Incidence, Prevalence, and Duration in Alternating Events Models

• Activity Number: 642
• Type: Invited
• Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
• Sponsor: SSC
• Author(s): Mei-Cheng Wang*+ and Russell T. Shinohara
• Companies: The Johns Hopkins University and Johns Hopkins Bloomberg School of Public Health
• Keywords: Incidence ; Prevalence ; Duration ; Alternating Events ; Recurrent Events

Abstract

In studies of a non-recurrent disease, it is understood that incidence, prevalence and duration are three important components for formulating the magnitude and changes of the disease over time. In this talk we study a recurrent disease with emphasis on incidence, prevalence and duration through useful formulas in alternating events models. Statistical methods are developed for estimating the three components. In modeling of alternating events data, interests could focus on probability structure of time-to-events or time-between-events. In general, a point process model gives time-to-events interpretation and a gap time model leads to time-between-events interpretation. While the proposed work can be viewed as a compromise between time-to-events and time-between-events analyses, it offers additional and important interpretations that the other two models do not offer.