Introduction
Epidemiology is the medical science that asks what causes a disease and how it spreads among people; it studies the distribution and determinants of disease in human populations. Its main aim is to find and establish a causal relationship between an exposure to a risk factor and an outcome, and the main instruments for this are epidemiological studies. This topic covers the experimental studies: intervention trials, types of control groups, and the randomised clinical trial (RCT).
Experimental studies
- Always have an experimental group and a control group.
- The investigator controls the exposure, often in a randomised way.
- Kinds: randomised control trials, field trials, community trials.
Intervention trials
An intervention trial is a follow-up experimental study in which the exposure under investigation is applied by the investigator. The aim is to investigate a possible causal relationship by exposing one or more groups of patients to a factor (a risk factor or treatment) and comparing the results with one or more control groups that are unexposed to this factor. The investigator decides which subjects are "exposed" and which are not, so in epidemiology intervention trials are the closest analogue of a laboratory experiment.
Intervention trials can be
- Community trials: a research study that involves an individual-level intervention, a community-level intervention, or both.
- Clinical trials: a research study to test new methods of screening, prevention, diagnosis or treatment of a disease, carried out in hospital settings; randomised or non-randomised.
- Laboratory experimental trials: a research study carried out on animals or tissues in laboratories.
Types of interventions
- Prophylactic: targets prevention (e.g. vaccines).
- Diagnostic: targets evaluation of a new diagnostic procedure (e.g. a new lab test compared with the gold standard).
- Therapeutic: targets treatment (e.g. a new drug).
Types of control groups
- Placebo concurrent control group
- Dose-response concurrent control group
- Active concurrent control group
- No-treatment concurrent control group
- Historical control