Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 289 - Contributed Poster Presentations: Section on Statistics in Epidemiology
Type: Contributed
Date/Time: Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #320872
Title: Subclass-Wise Measures and Tests for Interaction Detection in Epidemiological Studies
Author(s): Eiji Nakatani*
Companies: Japan
Keywords: interaction test; risk ratio; risk difference
Abstract:

Scales such as the RRR (Ratio of Risk Ratio) and RERI (Relative Excess Risk due to Interaction) have been used in epidemiology to measure the interaction between two exposure variables, but these scales are unable to assess the magnitude of the effect in some subpopulations, and it is difficult to compare between subpopulatios. In this study, in response to these problems, we will create new scales, RRRp (Ratio of Risk Ratios [RRs] with RR in the population as a denominator) and DRDp (Difference of Risk Differences [RDs] with RD in the population as a comparator). Tests and confidence intervals can be constructed for each subpopulation, and heterogeneity can be determined for all subpopulations. In addition, the RRRp and DRDp for each subpopulation can be interpreted. With respect to adjustment for confounding, we define RRRp and DRDp using the latent outcome variables and estimate the average causal effect using regression standardization under the assumption of consistency and conditional independence of the two exposure variables and outcomes under conditioning on confounders.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program