Epidemiology has produced some useful effect measures, the etiologic fraction and the potential impact fraction, with which the effects of changes in risk factor prevalence on disease specific incidence can be estimated. However to be able to use them for decisions in health policy, adjustments need to be made, in order to incorporate elements that can be avoided in carefully designed epidemiologic studies but cannot be ignored if the results are applied to the "real world". Such necessary adjustments include an approach in which the multi factorial aspects of chronic disease epidemiology and time dimensions can be taken into account. In this article we show the effect of including a time dimension in epidemiologic effect measures. Such a time dimension is important, e.g. to be able to incorporate the effect of a gradual reduction in excess risk following cessation of exposure to a risk factor such as cigarette smoking.