The scope of epidemiology
Epidemiology has been usefully defined as the study of the distribution and determinants of disease in human populations.
Most epidemiologists would regard the randomised controlled trial (RCT) as the ‘gold standard’ of experimental design, which produces the most reliable of results in the pursuit of the aims of epidemiology. Unfortunately, there are many situations in which the RCT is not feasible: either it is too expensive to run, or it is impossible to recruit enough patients, or it is ethically unjustifiable. In such circumstances, observational research methods are needed.
Types of epidemiological study
These can be either descriptive or analytic.
‘Descriptive’ encompasses quantitative studies, e.g. censuses and surveys, and qualitative studies, e.g. focus groups, and supply essential data for many analytic studies.
Analytic studies seek to establish relationships between diseases and their causes, and deal with quantitative data.
There are four main categories of analytic studies:
(i) Cohort
(ii) Case-control
(iii) Cross-sectional
(iv) Ecological
All these types of study are known as observational: that is, they involve observing what is happening, without interfering with the ‘natural’ situation. Randomisation in such a study means the selection of a random sample of the population for observation, and has nothing to do with random allocation of treatment, which is seen in an RCT.
Cohort study
As can be seen from the diagram, the design of the study is prospective – i.e. moves forward in time from identification of the ‘exposure’ (the putative causal factor) to measurement of the occurrence of disease (‘outcome’).