Testing for Sufficient-Cause Gene-Environment Interactions Under the Assumptions of Independence and Hardy-Weinberg Equilibrium

Comparative analysis of drivers of energy consumption in China, the USA and India - A perspective from stratified heterogeneity

With the limited amount of resources, developing effective strategies to make full use of them and decrease the energy consumption without too much sacrifice of economic output requires identifying key drivers of energy consumption growth rate as a prerequisite. Meanwhile, as top three consumers of primary energy of the world, China, the United States of America, and India burn over 45% of global fuels in 2016. Conducting an empirically comparative analysis of them can also set up pilot scheme for other economies to develop more efficient strategies for energy consumption. The paper modified the original Geographical Detector model with a different sampling method to detect the key driver of energy consumption growth rate, which filling the gap that there are possible interactions of potential factors. The results show that coal intensity is the biggest driver to change overall energy consumption growth rate in China and India. In comparison, for the United States, the leading drivers of energy use are the factors of individual incomes and oil intensity. In addition, all factors have interactions and enhance each other when influencing total energy consumption growth rate. India has the strongest factor interactions when influencing the energy consumption growth rate among the three economies, all interactions between factors in US is not significant as those in China and India. Besides providing outcomes that can contribute towards developing new strategies to use energy more efficiently, this research offers a pilot example of analyzing energy issues from the perspective of stratified heterogeneity in consideration the characteristic differences of each factor.

Testing for Sufficient-Cause Interactions in Case-Control Studies of Non-Rare Diseases

Sufficient-cause interaction (also called mechanistic interaction or causal co-action) has received considerable attention recently. Two statistical tests, the ‘relative excess risk due to interaction’ (RERI) test and the ‘peril ratio index of synergy based on multiplicativity’ (PRISM) test, were developed specifically to test such an interaction in cohort studies. In addition, these two tests can be applied in case–control studies for rare diseases but are not valid for non-rare diseases. In this study, we proposed a method to incorporate the information of disease prevalence to estimate the perils of particular diseases. Moreover, we adopted the PRISM test to assess the sufficient-cause interaction in case–control studies for non-rare diseases. The Monte Carlo simulation showed that our proposed method can maintain reasonably accurate type I error rates in all situations. Its powers are comparable to the odds-scale PRISM test and far greater than the risk-scale RERI test and the odds-scale RERI test. In light of its desirable statistical properties, we recommend using the proposed method to test for sufficient-cause interactions between two binary exposures in case–control studies.

Sharp bounds on sufficient-cause interactions under the assumption of no redundancy

BACKGROUND

Sufficient-cause interaction is a type of interaction that has received much attention recently. The sufficient component cause model on which the sufficient-cause interaction is based is however a non-identifiable model. Estimating the interaction parameters from the model is mathematically impossible.

METHODS

In this paper, I derive bounding formulae for sufficient-cause interactions under the assumption of no redundancy.

RESULTS

Two real data sets are used to demonstrate the method (R codes provided). The proposed bounds are sharp and sharper than previous bounds.

CONCLUSIONS

Sufficient-cause interactions can be quantified by setting bounds on them.

Complementary Log Regression for Sufficient-Cause Modeling of Epidemiologic Data

The logistic regression model is the workhorse of epidemiological data analysis. The model helps to clarify the relationship between multiple exposures and a binary outcome. Logistic regression analysis is readily implemented using existing statistical software, and this has contributed to it becoming a routine procedure for epidemiologists. In this paper, the authors focus on a causal model which has recently received much attention from the epidemiologic community, namely, the sufficient-component cause model (causal-pie model). The authors show that the sufficient-component cause model is associated with a particular ‘link’ function: the complementary log link. In a complementary log regression, the exponentiated coefficient of a main-effect term corresponds to an adjusted ‘peril ratio’, and the coefficient of a cross-product term can be used directly to test for causal mechanistic interaction (sufficient-cause interaction). The authors provide detailed instructions on how to perform a complementary log regression using existing statistical software and use three datasets to illustrate the methodology. Complementary log regression is the model of choice for sufficient-cause analysis of binary outcomes. Its implementation is as easy as conventional logistic regression.

False Appearance of Gene-Environment Interactions in Genetic Association Studies

Under the assumption of gene-environment independence, unknown/unmeasured environmental factors, irrespective of what they may be, cannot confound the genetic effects. This may lead many people to believe that genetic heterogeneity across different levels of the studied environmental exposure should only mean gene-environment interaction--even though other environmental factors are not adjusted for. However, this is not true if the odds ratio is the effect measure used for quantifying genetic effects. This is because the odds ratio is a "noncollapsible" measure--a marginal odds ratio is not a weighted average of the conditional odds ratios, but instead has a tendency toward the null. In this study, the authors derive formulae for gene-environment interaction bias due to noncollapsibility. They use computer simulation and real data example to show that the bias can be substantial for common diseases. For genetic association study of nonrare diseases, researchers are advised to use collapsible measures, such as risk ratio or peril ratio.