The role of gene x lifestyle interactions in the development of cardiometabolic diseases is often highlighted, but very few robustly replicated examples of interactions exist in the literature. The slow pace of discoveries may largely be due to interaction effects being generally small in magnitude and/or more complex than initially thought. However, the progress may also be hindered by the poor accuracy in large-scale epidemiological studies to estimate the true interaction effect sizes. Often, this bias tends to underestimate the interaction effect, leading to inadequate statistical power to detect the interaction. In this review, I will discuss the most common sources of bias in the estimation of gene x lifestyle interactions, and will discuss how such factors could be addressed in the future to enhance our potential to identify and replicate interactions for cardiometabolic diseases.