Observational studies are a fundamental part of epidemiological research. They are called observational studies because the investigator observes individuals without manipulation or intervention. This is in contrast to randomised controlled trials where investigators do intervene and look at the effects of the intervention on an outcome. Although randomised controlled trials are useful in determining causal relationships between treatment and outcome, there are often instances where randomised controlled trials are not appropriate, so observational studies are needed.
Randomised controlled trials have to be ethical, but often there are questions that it would be unethical to do a trial on. For example, if one wanted to conduct a hypothetical trial to determine if smoking caused lung cancer, individuals would be randomized to smoke or not smoke, and then followed up until death to subsequently determine the effect of smoking on lung cancer. Observational studies can determine if there are associations between smoking and lung cancer, but instead of the investigator controlling who smokes or not, smoking status is observed. With appropriate control of other factors related to lung cancer and smoking status, such as gender and age, observational studies have shown a strong association between smoking and lung cancer events.
Often observational studies lead the way in research as associations from these studies help to formulate hypotheses to be tested in subsequent controlled experiments. Observational studies can also help inform probable cause and effect associations before randomised data are available, as randomised controlled trials often take years to complete.
Observational studies are also useful in studying rare events where we can retrospectively collect data to determine probable causes.
Additionally, observational studies help researchers know what happens in real life situations. These studies serve as a collection of data from standard practice.
In conclusion, observational studies are a key part of research. They help to investigate relationships that are unable to be tested under randomised controlled experiments, can help provide insight and develop hypotheses on what subsequent randomised evidence is needed for future research, and can provide an understanding on how things work in clinical practice.