2020/05/14 by Michail Katsoulis, Christina Bamia, Katsoulis, Michail +1
Medicine · #Applications (stat.AP) #FOS: Computer and information sciences #Methodology (stat.ME) #Nutritional Studies and Diet
paper · pdf · doi:10.48550/arxiv.2005.07142
openalex publication_date 2020/05/14 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Many studies have focused on investigating deviations from additive\ninteraction of two dichotomous risk factors on a binary outcome. There is,\nhowever, a gap in the literature with respect to interactions on the additive\nscale of >2 risk factors. In this paper, we present an approach for examining\ndeviations from additive interaction among three on more binary exposures. The\nrelative excess risk due to interaction (RERI) is used as measure of additive\ninteraction. First, we concentrate on three risk factors - we propose to\ndecompose the total RERI to: the RERI owned to the joint presence of all 3 risk\nfactors and the RERI of any two risk factors, given that the third is absent.\nWe then extend this approach, to >3 binary risk factors. For illustration, we\nuse a sample from data from the Greek EPIC cohort and we investigate the\nassociation with overall mortality of Mediterranean diet, body mass index\n(BMI), and, smoking. Our formulae enable better interpretability of any\nevidence for deviations from additivity owned to more than two risk factors and\nprovide simple ways of communicating such results from a public health\nperspective by attributing any excess relative risk to specific combinations of\nthese factors.\n