Partitioning the Contribution to Death Attributable from Multiple Time-Varying Coexisting Diseases

• Activity Number: 638
• Type: Contributed
• Date/Time: Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Haiqun Lin*+
• Companies: Yale University
• Keywords: aaf ; sef ; le-aaf

Abstract

The use of the average attributable fraction (AAF) based on a regression model for a dichotomous outcome has been developed in the last decade. This approach takes into account the interrelationship between coexisting risk factors when measuring how much disease burden (or death) could be reduced if some risk factors were eliminated. The AAF method partitions the total attributable fractions due to a set of diseases into the sum of attributable fractions for the individual diseases. We extend the AAF method, which was developed for situations in which disease occurrence and exposures are fixed in time, to the longitudinal situation in which the diseases and the outcome may be time-varying. The proposed longitudinal extension of AAF (LE-AAF) preserves the attractive properties of additivity and symmetry of the AAF. In this manuscript, we describe our method and apply it to the case of de