Quiz 5 - Bias, Confounding

Case-Control study designs are ideal for rare diseases.

True

Suppose that you interview a sample of older individuals living in a senior citizen center and ask if they currently have hypertension and if they currently smoke. What type of study design is this?

Cross-sectional

Suppose that you interview a sample of older individuals living in a senior citizen center and ask if they currently have hypertension and if they currently smoke. What type of study design is this?

Cross-Sectional

True or False? In a case-control study, if there is an association between exposure and disease, you would expect a higher proportion of exposure among cases than the proportion exposed in the control group.

True

If the true risk ratio is 1.0 and the confounded risk ratio is 2.0, this is an example of which of the following?

Positive confounding

True or False? A confounding variable can be an intermediate step in the causal pathway between the exposure and the disease.

False

True or False? When the direction of a potential bias is towards the null, it means that the error causes the true measure of association to be underestimated.

True

True or False? Loss to follow-up in a cohort or RCT study design can introduce important selection bias if the loss of people over time is related to their exposure and disease status (i.e. differential loss and not random loss).

True

True or False? Epidemiological studies of the association between an exposure and a disease are susceptible to the disturbing influences of extraneous factors called confounders.

True

True or False? In an RCT study design, randomization does nothing to balance known and unknown confounders across the intervention and control group.

False

What criteria should a confounder meet?

1. Must be associated with exposure

2. Must be associated with outcome

3. Must not be a mediator between exposure and disease