Application of flexible matching strategy to detect gene-environment interactions for increasing the study power / 中华流行病学杂志

ABSTRACT

Flexible matching has recently been proposed as a method of improving interactions efficiency. In this study, the concept of flexible matching has been introduced, and the applicability of this strategy has also been described based on the power calculation of interaction between HER-2 polymorphism and smoking with breast cancer. A large-sample approximation method is used to estimate the power and efficiency of gene-environment interactions. In the basic scenario, power of interaction between HER-2 polymorphism and smoking of unmatched case-control study appears to be 30% while in the frequency matching case-control study it is 56%. However, when increasing the smoking prevalence in controls, greater power can be obtained (power=74%). Conclusions: Flexible matching strategies can increase the power and efficiency of case-control studies to detect and estimate the gene-environment interactions when compared with traditional frequency matching and it is especially useful under those scenarios when low environmental exposure of population, adverse gene-environment interactions or less paired controls are seen. Optimal matching design should be made available by weighing the benefits and loss due to flexible matching.