 Confounding factor is a third factor which is independently associated with both the outcome of interest and the exposure. When this third variable is distributed unequally among the study and control groups, it results in confounding. Examples of confounders include observed associations between coffee drinking and heart attack, where smoking can be a confounding factor. Demographics, such as age or gender, and risk factors, such as body mass index or smoking, can usually result in confounding. However, the effect of confounders can be controlled by restriction, matching, and proper randomization in a study with a larger sample size. Stratified analyses and adjustments in multivariable regression analysis can identify confounders during data analysis. The role of confounding or effect modification by an independent factor can be identified by the stratified analysis. The association will disappear in case of the confounding variable in the stratified analysis, while in case of the effect modification, the association will get stronger.