Retrospective Case-Control Studies in Epidemiology

Retrospective Epidemiological Studies

Case-Control Studies

Retrospective = effect → cause. Case-control studies identify two existing groups that differ in outcome and compare them based on a supposed causal attribute.

• They work by comparing subjects who have the condition/disease (the cases) with patients who do not have the condition/disease but are otherwise similar (the controls).
• Advantages: relatively inexpensive; can be carried out by small teams or individual researchers; shorter in duration.
• Disadvantage: difficulty obtaining reliable information about an individual’s exposure status over time.
• Example of a successful study: demonstration of the link between tobacco smoking and lung cancer.

Characteristics

• Outcome has already occurred; exposure is measured retrospectively.
• Controls are selected on the basis of not having the outcome.
• Good for rare outcomes.
• Relatively inexpensive.
• Smaller numbers required.
• Quicker to complete.
• Prone to selection bias.
• Prone to recall (retrospective) bias.

The Groups

Two groups are selected based upon their history:

• Those who were exposed to the studied risk factor.
• Those who were not exposed.

At present, the current state of the patient is divided into:

• Those who have the disease (a + b).
• Those who do not have the disease (c + d).

Thus, four groups will ultimately be formulated:

• Those who were exposed and develop the disease (a).
• Those who were exposed and do not develop the disease (b).
• Those who were not exposed but develop the disease (c).
• Those who were not exposed and do not develop the disease (d).

Calculations and Trials

Calculations: as for prospective studies.

Randomized Clinical Trial: a model of the experimental method (used before treatment with new drugs, new methods, etc.).

Trial Types and Controls

• Controlled: observational study (two groups).
• Uncontrolled (least certain).
• Blind Experiment (BE): accounts for the placebo effect.
• Double Blind Experiment (DBE): neither participants nor investigators know treatment allocation.