Case Control Study Design

strengths of case control studies

-good for assessing multiple exposures of one outcome -useful when disease are rare -useful in determining associations (NOT CAUSATION) -less expensive thant international trials and prospective cohort studies -useful for ethical limitiations -useful when disease has a long indcution/latent period

commonly use

2x2 tables in case control study designs.

selection criteria is applied to

all study participants

recall bias is related to the

amount or specificity that cases or controls recall past events differently.

if the same person is exposed and unexposed in the same study, the person can be

associated with an outbreak or with multiple exposures OR in a situation with a breif, acute change in risk of the outcome in interest. this is called a case crossover.

in case control studies, the counterfactual theory must

be present. counterfactual outcome for smokers estimated by nonsmokers requires exhangeability

prospective exposure data is difficult/expensive to obtain or very time inappropriate--> use

case control study design which is usually retrospective

subjects in cohort studies ultimately developing the disease or outcome are defined as

cases for the subsequent case control study.

the most dicfficult part of the selection process is the

control selection

select cases in case control study by using

defined criteria (by investigator) using accurate, medically reliable, efficient data

in case control study design, the group assignments are based on

disease status

the control group supplies information about he

expected baseline risk factor profile in the population which the cases are drawn

an individual can actually function as both an

exposed individual and an unexposed individual in the same study

controls must be selected irrespective of

exposure status

case control studies are looking for the

exposure, because the disease is known. The goal is to see if there is any difference in exposure status to form RR, OR, HR.

weakness of case control studies may be the opposite of the

general strengths listed above.

matching

in some studies, cases are matched to controls in 1:1 or higher ratio; can be 3:1, etc.

possible to introduce bias in which ways

interviewer, recall, misclassification--> avoid by using top notch assessment/diagnostic studies.

the way the controls are selected is a

major determinant in whether any conclusion is valid internal validity and selection bias come in to play here.

indivudual matching

matches individuals based on specific patient based characteristics, useful for controlling confounding characteristics-difficult to do with high specificity.

dont match on anything that

might be a risk factor.

case control studies are useful when

studying a rare disease or investigating an outbreak.

selection bias is related to the way

subjects are chosen for study

when selecting controls used for nested case control studies

survivor sampling base sampling risk set sampling

in an outbreak, select controls form the outbreak; example

take diseased from potlock, take non diseased from potlock as well and use them as control.

clinical supportable and definable criteria are

the best

the goal of selection is to assess for

the presence of an association between exposure and known condition of interest by selecting non disease individuals from the sample population which produced the cases --> expectation is the controls represent the baseline risk of exposure in the general population

nested case control studies

these are the case control studies conducted after or from a prospective previous study type (cohort or interventional)

why do a case controls study design

when you are unable to force group allocation due to ethics/feasabiliy, limited resources, or rare disease.

how to select the control study population

you want to groups to be as equal as possible except the presence of the disease

survivor sampling

sample of non diseased individuals at the end of study period

risk set sampling

sample of non diseased individuals during study period at same time when case was diagnosed.

base sampling

sample of non disesaed individuals at start of study period

classifying pt; s correctly is ideal but there is a risk for

musclassifying subjects

selection criteria is

objective, consistent, accurately and validity

case control studies are

observational

case control studies commonly generate an

odds of exposure for each of them and then they use the odds ratio as the measure of association.

when determining controls, you base the selection on the

outcome, not on the exposure which is hypothetically unknown and trying to be associated with the study. Example: outcome head injuries exposure is helmet or no helmet case: cyclist with head injuries control: cyclists without head injuries

case controls studies allow researcher to be

passive observe of natural evens occurring in individuals with disease of interest who are compared to people who do not have the disease.

control groups can come from several sources

population --> state/neighborhood; obtained via numerous avenues including randomly instituional--> illness of controls should be unrelated to exposures being studied spouse/friends--> genetic, environmental, SES, etc.

group matching

proportion of cases and proportion of controls with identical characteristics are matched-requiem cases be selected first

case control design is useful when studying a

rare disease