Case-Control Studies Flashcards
case-control studies
observational studies allowing researcher to be a passive observer of natural events occurring in individuals with the disease of interest compared to people who do not have the disease
group assignments of case-control studies
group assignments are based upon your disease status
diseased = group of interest
no disease = control
case-control studies are useful when …..
studying a rare disease or investigating an outbreak
case-control: randomization
do not randomize people into groups
take the population and just randomly select people to include but their group allocation is decided by their disease state
in case-control we are looking at…….. (data)
odds of exposure for each group and an odds ratio
commonly use a 2x2 table to compare the data
reasons to select a case-control design
- unable to force group allocation
- limited resources
- disease of interest is rare
- prospective exposure data is hard to obtain
unable to force group allocation
unethical
not feasible
limited resources
time
money
subjects
disease of interest is rare
in occurrence
little is known about it’s associations or causes
prospective exposure data is difficult to obtain
expensive
time inappropriate
case-control studies are classically _______ studies
retrospective studies
data is already collected, all has happened in the past
ex. any long term cancer
typically know the outcome of interest
strengths of case-control studies
- good for assessing multiple exposures of one outcome
- good for rare diseases
- good for determining associations (not causes)
- less expensive than interventional/prospective studies
- less ethical issues
- good for diseases w/ long induction/latent periods
weaknesses of case-control studies
- can’t determine causes
- impacted by confounders
- retrospective-cannot control for variables or contamination
- impacted by biases - selection and recall bias
- may only have limited data available due to being retrospective
selection of cases is defined by _______
the investigator using accurate, medically reliable data sources
classifying patients correctly is ideal but misclassification is present
new term for case definition
diagnostic criteria
all else being equal, the outcome if something didn’t occur
recall counterfactual theory
requires assumption of exchangeability to allow comparability
most difficult part of case-control studies is
picking the controls
goals of control selection
to assess for the presence of an association between exposure and known condition of interest by selecting non-diseased individuals from the same population which produces the cases
expectation of the controls
that they represent the baseline risk of exposure in the population
control subjects cannot be selected upon
because of their exposure
researcher cannot look at a subjects exposure before allocating them as a control
the way controls are selected is a major determinant of …….
- validity of conclusions
- internal validity
- selection bias
differences between case and control groups
want the groups to be as equal as possible except that the control subjects do not have the disease
if the exposure truly has no effect, then ……
the odds will be exactly the same for both groups
OR = 1.0
controls must be selected ……
irrespective of exposure status
selection of controls can come from several sources:
- randomly obtained from the population
- institutional/ organization/ provider of care
- spouse/ relatives/ friends
how to select the control group for studies on an outbreak
people who participated in the same event that caused the outbreak but did not get the actual illness
an individual can function as both an exposed individual and an unexposed individual in the same study
the case-crossover design
because they are looking at different exposures
the only case-control design able to adequately attempt to address the issue of ______ .
case-crossover design
issue of temporality
case-crossover study
a case-control study (observational)
finds people who are cases and finds a temporally close earlier time period in their life when they weren’t doing the exposure so didn’t have the event
nested case-control study
a case-control study nested/comes from a study that is already done. from a cohort or interventional study.
comes from another study but set up as a normal case-control study
can be used to evaluate other exposures
selection of controls used for nested case-control studies
- survivor sampling
- base sampling
- risk-set sampling
survivor sampling
sample of non-diseased individuals at end of the study period
for control subjects in nested case-control studies
base sampling
sample of non-diseased individuals at start of study period
for control subjects in nested case-control studies
risk-set sampling
sample of non-diseased individuals during study period at same time when case subjects were diagnosed
for control subjects of nested case-control studies
another term for risk-set sampling
risk-analysis sampling
common biases in case-control studies
- selection bias — how subjects were chosen
2. recall bias — is subjects are required to recall information that will effect the study
matching
1:1 typically in case:control but can be higher
can be individual or group matching
individual matching
matching individuals based on specific patient-based characteristics
what is individual matching useful for?
controlling confounding characteristics
group matching
proportion of cases and proportion of controls with identical characteristics are matched
this requires cases to be selected first
what does group matching require
case subjects to be selected first
when matching, never match anything that ____ .
could be a risk factor
