Invited Commentary: How Big Is That Interaction (in My Community)—and in Which Direction?

Abstract

In an accompanying article, Turner et al. (Am J Epidemiol. 2014;180(12):1145-1149) compare the joint effects of smoking and air pollution to make inferences about the reduction in lung cancer mortality achieved when reducing each exposure separately and when reducing both together. In this commentary, we use first principles to quantify the difference between the risk or mortality reduction obtained from reducing each of 2 exposures together and the sum of the risk differences obtained from reducing the 2 exposures separately. Metrics of the impact of joint effects or comparisons of joint effects presented in units of absolute risk, such as Rothman's I, can provide more meaningful quantitative measures of public health impact than unitless metrics (e.g., ratios) and standardized metrics (e.g., the population attributable fraction) of potential interventions for reducing smoking and air pollution exposure. In particular, the venerable attributable community risk metric can provide an estimate of the community impact of such interventions in units of absolute risk. A spreadsheet we provide demonstrates the calculation of the various metrics for hypothetical data similar to those reported by Turner et al. Using algebra, graphics, and examples, we show that positive interaction, or synergy, on the additive scale implies that the impact on risk reduction from a program that applies both interventions will be lesser than the sum of the impacts of the separate interventions.

Inferential challenges when assessing racial/ethnic health disparities in environmental research

“…logistic, Poisson) and by default represent deviations from multiplicative joint effects and not deviations from additivity. Many papers discuss why the absolute scale is a more appropriate scale for inferring public health policy implications [ 71 ]. For authors interested in quantifying EMM on the additive scale from multiplicative models, some tools have been proposed to facilitate this calculation.…”

How do we assess a racial disparity in health? Distribution, interaction, and interpretation in epidemiological studies

“…For all examples, we assume no unmeasured confounding and sufficient power to detect group-specific prevalence differences and modification of the exposure-outcome association. While we believe this framework applies on all scales, examples refer to modification on the additive scale as this scale is supported in interaction literature due to its collapsibility and interpretability [12,14].…”

“…Instead, we present a more comprehensive framework for assessing health disparities, focusing on the distribution of the outcome and exposure across racial groups as a critical companion to assessment of interaction terms and stratum-specific effects. This paper expands upon a long tradition in epidemiology of interaction assessment by adding a specific focus on racial health disparities, which is a central component for the investigation of health disparities [11] and an area of research in which there may remain an over-reliance upon statistical significance testing [12–14]. We promote a framework of three guiding questions for disparity investigation by exploring the relationship of exposure, outcome, and interactions to assess meaningful differences in population health and the resulting implications for interventions.…”

Association between ambient particulate matter concentration and fetal growth restriction stratified by maternal employment

“…Non-employed mothers were defined as the retired, housewives, students, or those without work-related activity. Based on the notion of superiority of additive scale to multiplicative scale in terms of estimating the impact of intervention [39], we evaluated additive interaction between maternal employment and PM exposure in the odds of fetal growth restriction. If the strata-specific effect estimates are not homogenous across strata, the effect modification is considered to be present.…”