In an observational study, the researchers do not intervene in the care of the patient, but simply observe and record what happens. During investigational clinical trials, the researchers intervene in the care of the patient in a pre-planned way and record the outcome. Observational studies often use questionnaires and survey to record information about the life styles and care of the people in the studies.

Observational studies may also be called epidemiological studies. An epidemiological study looks at the causes and patterns of disease, and whether a particular factor causes cancer or not.

There are two main types of observational study; case-control studies and cohort studies.

Case control studies

These are retrospective studies in which groups of patients with cancer (cases) are compared to groups of patients without cancer (controls). The aim is to see whether exposure to any factor occurs more or less frequently in the cases compared to the control group.

An example of a case control study would be the comparison of smoking habits in people with lung cancer (cases) and people without lung cancer (controls) to determine a link between smoking and lung cancer.

Case control studies are relatively cheap, quick and easy to carry out. There is no loss-to-follow-up and they are suitable for rare events. However, the timing of events cannot be reliably established and it is difficult to assess causality. Also, case control studies rely on people remembering events, and are, therefore, subject to recall bias.

Cohort studies

Cohort studies follow a group of people over time to assess the incidence of a disease (or some other event). Cohort studies are used to describe the effect of being exposed to one or more risk factors on the incidence of the disease. Cohort studies can be prospective (planned) or retrospective (historical).

Cohort studies can be used to assess the relationship over time between exposure to certain risk factors and the development of the disease. They can also be used to assess cause and effect, although randomised clinical trials are preferable for this. However, for rare diseases, the cohort may have to be very large with a long follow-up, resulting in an expensive study and many people being lost-to-follow-up.