Bias, Chance + Confounding Flashcards
what is selection bias?
systematic differences in characteristics between those who take part in a study and those who do not.
such as the persons who are selected to take part in the study are not representative of the reference population.
what is information bias?
any error in the measurement of exposure or outcome that results in systematic differences in the accuracy of information collected between comparison groups.
within this:
- reporting bias
- observer bias
- misclassification bias
what is reporting bias?
when subjects with a specific health outcome report previous exposures with a different degree of accuracy to those without the outcome.
give the answer that they think will please the investigator, or if they tend to conceal potentially embarrassing info.
how can reporting bias be minimised/avoided?
- use exposure data before outcome - objective data sources where possible
- keep subjects unaware of association under study
what is observer bias in case-control or cross-sectional studies?
when the accuracy of exposure data recorded by the investigator differs systematically between subjects in different exposure groups.
what is observer bias in cohort/intervention studies?
when the accuracy of the outcome data recorded by the investigator differs systematically between subjects in different exposure groups (interviewer bias)
how can observer bias be avoided in cohort studies?
the people who are responsible for classifying the outcome do not know the subject’s exposure category.
how can observer bias be avoided in case-control studies?
the people responsible for classifying the exposure do not know the subjects disease status (blinding).
what is misclassification bias?
either the exposure status is incorrect or the disease status is incorrect.
how is the confounding variable dealt with in the analysis?
- stratified analysis
- direct and indirect standardisation
- statistical modelling (multiple regression)
