Following on from the previous chapter, we discuss cohort studies in this chapter and compare them with case–control studies. The difficulties in assessing aetiology are discussed in Chapter 5.
Cohort studies are probably the most intuitive study designs to assess aetiology. In cohort studies, the outcome of interest has not occurred at the start of the investigation. People with and without the exposure of interest are followed up over time to examine the proportion in each group who go on to develop the outcome of interest. Occasionally, in cohort studies, data on exposure in the past are collected – usually from documented records or by measuring biomarkers from previously collected samples. The outcome may or may not have occurred at the time of the study in such instances; it is of course essential that the outcome has not occurred at the time of exposure. This type of cohort study is sometimes called a ‘retrospective’ cohort study although some epidemiologists dislike the term.
Figure 10a illustrates the basic design of a cohort study.
In cohort studies, we can compare the risk of disease in the exposed and unexposed groups and calculate risks and risk ratios, and rate and rate ratios. Direct measurement of incidence allows us to compare absolute differences too. We can also calculate odds ratios if necessary but presenting results as risks or risk ratios is preferred.