what is the purpose of a case-control study?
To measure disease incidence in the exposed vs. the unexposed.
what can the odds ratio be interpreted as?
- the risk of disease given exposure in most circumstances
when would you carry out a case-control study?
- to study rare outcomes
- to study the effects of multiple exposures on outcomes
- to study time-varying exposures
why is sample size important in case-control studies?
- it determines the effect size that can be detected as statistically significant
- it determines how many confounding variables can be considered at the analysis stage
what is selection bias?
when the relation between exposure and disease is different for those who participate and those who are theoretically eligible but do not.
how can selection bias be reduced/prevented?
- using population based sampling
which type of sampling is vulnerable to selection bias?
clinic-based sampling
what is information bias?
occurs when there are errors in the measurement of characteristics and the consequences of the errors are different for exposed vs unexposed/cases vs controls.
how is exposure data collected in case-control studies?
example: smoking status
- GP records
- self report
- saliva cotinine
methods of data collection in case-control studies
- case-note review
- postal questionnaire
- interview
- physical measurement
problems with data collection in case-control studies?
- observer bias
- subject bias (e.g. recall bias of mothers with cerebral palsy babies)
what is misclassification bias?
- when either there is a true case but a wrongly labelled control
OR
- there is a true control and wrongly labelled case
what is observer bias?
when there is knowledge of case/control status and it influences data collection
what is recall bias?
when the case and controls recall prior exposure differently
how can the c
