ABSTRACT
In this chapter we consider when it is possible to draw causal conclusions from a case-control study. This requires special attention because causal conclusions make stronger claims than ordinary predictions. For instance, it may not be sufficient to note that people with a higher 88alcohol intake are at a higher risk of stroke; rather, we may wish to establish that if we are able to reduce people’s alcohol intake, this will have the effect of reducing the number of instances of stroke. In other words, we wish to establish that certain interventions in the level of an exposure or in a lifestyle choice will result in improved health, and to quantify the strength of such effects.
