Interaction between two different factors can occur which influence the relationship with another factor. For example, there could be twice the risk of developing a disease for a smoker compared to a non-smoker, and twice the risk of developing the same disease if the person is overweight compared to someone who is within the ‘desirable’ weight category, but for an overweight smoker the risk of developing the disease may be ten times greater than a person who is a non-smoker and not overweight. This type of effect occurs for oral cancers, where the risk among smokers who drink alcohol is much higher than either one alone. Therefore, examining the relationship between two factors is not straightforward, and can be further complicated by confounding and effect modification.
Also see: Bias, Causality, Confounding, Confidence Intervals, Effect Modification, and Small Numbers.