Case-Control Studies

Control definition

A sample of the source population that produced the cases. Done so in order to estimate the exposure distribution in source population that produced the cases.

Definition of a Case-Control Study

A study in which cases of disease are identified, and then a sample of the population that produced the cases is identified (the controls). Exposures are determined and compared for individuals in each group. You compare exposures in cases and controls.

Strengths of Case-Control Studies

Fewer ethical concerns than experimental studies. More efficient than a cohort: saves money, time, allows for fewer subjects to be studied to make a measure of an association, no need to wait for a long-latency disease to develop. Easy to explore the effect of many exposures on an outcome.

The "Would Criterion"

If the controls had experienced the outcome, would they have been identified as cases in your study? They have to be independent of the exposure of the cases.

Limitations of Case-Control Studies

Usually limited to studying a single outcome. Inefficient for rare exposures. More opportunity for systematic bias (recall and selection bias). May be unsure about temporal sequence between exposure and disease. Cannot calculate absolute measure of association.

Calculation in a Case-Control Study

You can only calculate an odds ratio (which is a relative measure of association). No absolute measures of association can be calculated. Why? No longer have entire denominator of population at risk (because controls represent a sample of the total population).

Three Ways to Sample Controls in a Nested-Case-Control Study

1) Case Cohort Sampling 2) Risk-Set Sampling 3) Survivor Sampling

Requirements for picking a Control group

1) Controls must come from the same source population as the cases. Random selection. 2) Controls must be selected independently from exposure

The Process of Conducting a Case-Control Study

1) Define the study hypothesis 2) Define the source population 3) Identify valid controls from the same source population 4) Ascertain exposure for cases and controls 5) Calculate a measure of association between exposure and outcome

When is it desirable to conduct a case-control study?

1) When exposure data are expensive or difficult to obtain. 2) When disease has long induction and or latent period; results may take decades to emerge 3) When the disease is rare; a cohort study would require too large a sample size 4) When underlying population is dynamic; high loss to follow up may bias results 5) When little is known about the disease

Risk-Set Sampling

Choose a control each time a case occurs from among those without a disease at case diagnosis times.

Survivor Sampling

Choose a control from the members of source population who are non-diseased at end of follow-up. Controls cannot go on to become cases.

Case Cohort Sampling

Choose controls from starting cohort at beginning of the follow up. Controls may go on to become cases

Hospital or Clinic-Based Controls

Controls selected among patients at a hospital or clinic. Requirements: Same source population as cases: "Would criterion" and illness should be unrelated to exposure under study. Illnesses chosen have to have no relation to the risk factors under study.

Nested Control

Controls selected from an existing cohort population. The controls represent a sub-set of the full source population.

Population-Based Controls

Controls selected from the general population, and are most suitable when cases are from well-defined geographic areas.