Gordis Chapter 15: Bias, Confounding,
Interaction Test Questions Fully
Solved.
What is bias? - Answer any SYSTEMATIC error in the design, conduct or analysis of a study that
results in a mistaken estimate of an exposure's effect of the risk of disease
It can occur at every stage of investigation
-before the study (lit review, study design)
-during the study (sample selection)
-after the study (inferences made, subgroup analyses, publication bias)
Random error - Answer -NOT bias
-differences in measurement around the true value from sampling error or variability in
measurement (unreliable)
-can minimize by: larger sample, more reliable measures, improved data collection
Bias effects the ___________ validity of epidemiologic studies - Answer internal validity
Two major types of bias - Answer selection bias
information bias
What is selection bias? - Answer if the way in which cases and controls were selected distorts
the relationship between the exposure and the disease
In cohort studies, selection bias occurs when...... - Answer loss to followup is associated with
both the exposure and risk of the outcome
Give some examples of selection bias (BERNS) - Answer -non-Response bias (non-responders
are inherently different from responders)
-Exclusion bias (systematic difference in eligibility criteria to the cases and the controls)
, -Confouding by indication
Neyman's bias aka incidence prevalence bias - Answer This type of bias often happens when a
significant amount of time has passed between exposure and investigation; patients who have
died or recovered will be erroneously excluded from any analysis, skewing the results towards
individuals who are more "average." For example, a study of patients hospitalized for the flu will
miss those patients who have died, and those who have been discharged after recovery.
Neyman bias is less of a problem with acute, short-lived cases than with long-term diseases like
HIV or tuberculosis.
This type of bias is also called prevalence-incidence bias from the fact that it's preferable to use
incident cases instead of prevalent cases. Incident cases are newer cases — like first time
admissions. Prevalent cases are pre-existing cases, which are usually sicker with more
progressed disease than incident cases. Combining prevalent and incident cases can actually
make prevalent-incidence bias worse, obscuring the true relationship between your study
variables (i.e. the variables in your experiment or study)
Response bias: When cases are more likely to participate than controls, what happens to OR of
exposed cases? - Answer Overestimation of OR
Response bias: When controls are more likely to participate than cases, what happens to OR of
exposed controls? - Answer Underestimation of OR
Effectiveness - Answer A measure of the extent to which a specific intervention, procedure,
etc. when deployed, will do what is it intended to do for a specific population
Confounding by indication (type of selection bias) - Answer individuals receiving treatment for
a disease are receiving treatment because they are indicated for treatment (sicker individuals
are more likely to be treated)
Survival bias (type of selection bias) - Answer individuals whose prognosis allows them to
survive longer will be more likely to receive newer treatments that become available (less sick
more likely to receive new treatments)
Confounding by indication and survival bias complicate _____________ studies that aim to
characterize treatment outcomes among those receiving new treatments - Answer cohort
studies
Interaction Test Questions Fully
Solved.
What is bias? - Answer any SYSTEMATIC error in the design, conduct or analysis of a study that
results in a mistaken estimate of an exposure's effect of the risk of disease
It can occur at every stage of investigation
-before the study (lit review, study design)
-during the study (sample selection)
-after the study (inferences made, subgroup analyses, publication bias)
Random error - Answer -NOT bias
-differences in measurement around the true value from sampling error or variability in
measurement (unreliable)
-can minimize by: larger sample, more reliable measures, improved data collection
Bias effects the ___________ validity of epidemiologic studies - Answer internal validity
Two major types of bias - Answer selection bias
information bias
What is selection bias? - Answer if the way in which cases and controls were selected distorts
the relationship between the exposure and the disease
In cohort studies, selection bias occurs when...... - Answer loss to followup is associated with
both the exposure and risk of the outcome
Give some examples of selection bias (BERNS) - Answer -non-Response bias (non-responders
are inherently different from responders)
-Exclusion bias (systematic difference in eligibility criteria to the cases and the controls)
, -Confouding by indication
Neyman's bias aka incidence prevalence bias - Answer This type of bias often happens when a
significant amount of time has passed between exposure and investigation; patients who have
died or recovered will be erroneously excluded from any analysis, skewing the results towards
individuals who are more "average." For example, a study of patients hospitalized for the flu will
miss those patients who have died, and those who have been discharged after recovery.
Neyman bias is less of a problem with acute, short-lived cases than with long-term diseases like
HIV or tuberculosis.
This type of bias is also called prevalence-incidence bias from the fact that it's preferable to use
incident cases instead of prevalent cases. Incident cases are newer cases — like first time
admissions. Prevalent cases are pre-existing cases, which are usually sicker with more
progressed disease than incident cases. Combining prevalent and incident cases can actually
make prevalent-incidence bias worse, obscuring the true relationship between your study
variables (i.e. the variables in your experiment or study)
Response bias: When cases are more likely to participate than controls, what happens to OR of
exposed cases? - Answer Overestimation of OR
Response bias: When controls are more likely to participate than cases, what happens to OR of
exposed controls? - Answer Underestimation of OR
Effectiveness - Answer A measure of the extent to which a specific intervention, procedure,
etc. when deployed, will do what is it intended to do for a specific population
Confounding by indication (type of selection bias) - Answer individuals receiving treatment for
a disease are receiving treatment because they are indicated for treatment (sicker individuals
are more likely to be treated)
Survival bias (type of selection bias) - Answer individuals whose prognosis allows them to
survive longer will be more likely to receive newer treatments that become available (less sick
more likely to receive new treatments)
Confounding by indication and survival bias complicate _____________ studies that aim to
characterize treatment outcomes among those receiving new treatments - Answer cohort
studies
