Causal Inference with Multistate Models—Estimands and Estimators of the Population Attributable Fraction. (1st July 2019)

• Record Type: Journal Article
• Authors: von Cube, Maja; Schumacher, Martin; Wolkewitz, Martin
• Is Part Of: Journal of the Royal Statistical Society. Volume 183:Number 4(2020)
• Page Start: 1479
• Page End: 1500
• Publication Date: 2019-07-01
• Subjects: Causal effect measure -- Competing risks -- Excess fraction -- Measure of association -- Nosocomial infection -- Time-dependent exposure; Social sciences -- Statistical methods -- Periodicals; Statistics -- Periodicals; 300.15195
• DOI: 10.1111/rssa.12486
• Languages: English
• ISSNs: 0964-1998

Abstract

Summary: The population attributable fraction (PAF) is a popular epidemiological measure for the burden of a harmful exposure within a population. It is often interpreted causally as the proportion of preventable cases after an elimination of exposure. Originally, the PAF was defined for cohort studies of fixed length with a baseline exposure or cross-sectional studies. An extension of the definition to complex time-to-event data is not straightforward. We revise the proposed approaches in the literature and provide a clear concept of the PAF for these data situations. The conceptualization is achieved by a proper differentiation between estimands and estimators as well as causal effect measures and measures of association.