Confounding and effect modification: possible explanation for variation in the results on the association between oral and systemic diseases

Abstract

Objectives: There is large variation in the results of studies on the association between periodontitis and systemic diseases. The variation might be explained by the fact that the association between periodontitis and systemic diseases is confounded, or the association might be modified by extraneous factors. In this article, we show, using simple examples, how confounding and effect modification may cause variation in results. In addition, these examples show that uncontrolled or partially controlled confounders can induce spurious associations.

Conclusion: Confounding and effect modification may explain the variation in the results of studies on the association between periodontitis and systemic diseases.