case control studies Flashcards
what is a case control study
select diseased cases and undiseased controls
measure exposure in cases vs controls
what does an odds ratio of one mean
theres no difference
odds ratio more than 1
exposure if harmful
what is the calculation of odds ratio
A+C/ B+D
how to interpret odds ratio
the odds of disease given the exposure
when to do case control vs cohort
case control useful for rare outcomes
- dont require large number of cases (unlike cohort)
multiple exposures
time varying exposures
when to do cohort instead of case control
rare exposures
recall bias
bias definition
deviation of results from the truth or processes leading to such deviation
selection bias
the relation between exposure and disease is different for those who participate and those who are theoretically eligible but do not (because of procedures used to select subjects or factors that influence study participation)
how to avoid selection bias
selected cases should represent all the cases in the source population
- population based source
- clinic based source (more difficult)
information bias
occurs when there are errors in the measurement of characteristics and the consequences of the errors are different for exposed vs unexposed
examples of information bias
the way the disease has been measured in the past?
prevalent cases
existing cases of disease
incident cases
newly diagnosed
advantages and disadvantages of prevalent cases
advantage: increases sample size
disadvantage: reflects determinants of disease onset or duration
