Aim This paper discusses the methodological considerations inherent in the use of the cohort study as a research design.

Background Cohort studies are observational in design and are generally concerned with information regarding the prevalence distribution and inter-relationship of variables in a population. They are also used to identify risk factors and to collect information to describe the natural history or progression of disease. As with any research study, there are many sources of bias that threaten the credibility of cohort studies. Biases may arise from poor measurement (information bias), the sample being unrepresentative of the target population (selection bias) or the differential effects of other determinants on the association of interest (confounding).

Implications for practice Cohort studies provide a wealth of valuable information about population health- which informs the planning and implementation of health policy. Cohort designs are thus ideal for many of the health-related areas that interest nursing and midwifery researchers.