Controls must have as similar a background as possible to the cases, except that they do not have the outcome in question. They should come from the same population as the cases. Their selection should be independent of the exposures of interest. Objective measures of the presence of risk factors are best, ideally carried out in a 'blind' assessment or before the cases and controls are identified

Define a study population

Identification of cases can be made from the general population using health register and data or from a particular medical setting. The criteria for diagnosis of a case should be defined as well as the eligibility criteria used for selection. The diagostic criteria should be sensitive and specific (i.e. strict!). Information on diseases can be got from death certificates, disease registers, medical records or population survey. For rare diseases, cases may have to sought from large areas or over many years.

Define and select cases

This is a very important step. Get this wrong and you introduce bias into the study. Controls should represent the population that the cases come from (i.e. they should be at risk of becoming new cases). Ratio to cases is usually 1:1. If cases are limited, you can have up to 4 controls: 1 case. Some time will be needed in considering the way in which the cases and controls, which make up the study will be chosen. More heterogeneity in the cases, less likelihood of being able to link a specific risk factor to the disease causation. But, narrower the category of disease for inclusion as 'cases', less general applicability the findings will have.

Define and select controls

The measurement of the exposure(s) must be collected in a comparable way for cases and controls. It is worth 'blinding' the data gatherers to case or control status of participants or at least blind them to the main hypothesis of the study. This should help prevent measurement or researcher bias. Exposure information can come from records (though, obvious disadvantage is that records can be inaccurate, incomplete and were not originially collected for the study purposes) or can be via an interview or questionnaire (this can introduce recall bias, where cases have more vested interest in recalling the exposures than controls, and sometimes rely on 'proxy' respondents, e.g. carers, or parents of children).

Measure exposure

1. Define a study population (source of cases and controls) 
2. Define and select cases.
3. Define and select controls.
4. Measure exposure.

University of Michigan - Ann Arbor

Case-control studies are observational studies where subjects' previous experiences are used as evidence to assess how those who had experience with the specified condition varied from those who did not and why.

Case-control studies

James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning (Vol. 112, pp. 3-7). New York: springer.

An Introduction to Statistical Learning with Applications in R

Four steps in conducting a case-control study

1. To determine whether or not an association exists between something and a particular risk factor.
2. It may be the first step in testing a hypothesis. If positive, it can then be tested in a cohort study.
3. To start with a group of people with a certain characteristic and work back to see whether a possible risk factor may be the cause.

Purpose of case control studies

A case control study is an observational study method that selects people with a disease and those without a disease to investigate their exposure to a risk factor and its degree respectively, so as to determine whether the exposure risk factor is related to a disease and the degree of association.

Drawbacks: (Tobacco case as an example)
-  Retrospective: We know the subject has cancer, and in that context we have to go back and figure out why
-  Reversed logic of probability: What these data tell us is the probability that cancer patients are smokers, not the probability that smokers get cancer. For those who want to know if they should smoke, the probability of cancer in smokers is the probability they really care about
-  Recall bias: While the interviewers may be blinded to cancer vs. control, the patients themselves know whether or not they had cancer. Having already gone through this event can affect their recollections
-  Selection bias: Hospitalized cancer patients (aka those on record as having cancer and those specifically hospitalized for it) are not necessarily representative of the population at large or the smoking population at large.

Advantages of RCTs over Case Control Studies

Advantages of Case Control Studies:
- Quicker, cheaper and require less time and effort than cohort studies
- Case-control studies can study multiple risk factors/exposures
- They are useful for studying outcomes that take a long time to develop

Drawbacks of RCTs: 
- Ethics: RCT approaches are usually controversial and must be handled with care. The experimental group (intervened group) and control group (not- intervened group) are likely to have different outcomes and impact people who get involved in the trial. For example, you cannot assign people to smoke for decades just to figure out if they would get cancer. 
- Feasibility: Running RCTs can be very expensive. They cost time, effort and money. It can take decades to assess the long-term effect of a vaccine. Due to the high costs, most RCTS only inspect one or very few variables. Therefore they rarely reflect the full picture of a complicated issue.


Advantages of Case Control Studies over RCTs


An observational study method that selects people with a disease and those without a disease to investigate their exposure to a risk factor and its degree respectively, so as to determine whether the exposure risk factor is related to a disease and the degree of association.
Drawbacks: (Tobacco case as an example)
- Retrospective: We know the subject has cancer, and in that context we have to go back and figure out why.
- Reversed logic of probability: What these data tell us is the probability that cancer patients are smokers, not the probability that smokers get cancer. For those who want to know if they should smoke, the probability of cancer in smokers is the probability they really care about.
- Recall bias: While the interviewers may be blinded to cancer vs. control, the patients themselves know whether or not they had cancer. Having already gone through this event can affect their recollections.
- Selection bias: Hospitalized cancer patients (aka those on record as having cancer and those specifically hospitalized for it) are not necessary representative of the population at large or the smoking population at large. 

Learn Before

Related

Learn After