Managing risk in statistics - "Attributable Risk"

Abstract

In a previous article we discussed the value of determining the relative risk of developing a disease in relation to a specific exposure and thus its utility for assessing the exposure's etiological role in disease causation. This article focuses on attributable risk, another important measure, which assists us in determining what the impact would be of effectively intervening against a specific causative factor. It is determined by subtracting the risk in the unexposed group from the risk in the exposed group, that is, Risk (exposed) - Risk (unexposed). The underlying or background risk without the exposure is assumed to be the same in both groups. When the level of risk is the same in both groups, then the risk difference is 0 and one can conclude that the exposure makes no difference to the disease risk. However, if the risk difference is greater than 0, there is increased risk of the outcome in relation to the specific exposure under investigation. If it is less than 0, that is, a negative number, then the exposure under investigation would appear beneficial with a lower disease risk in the exposed group.