Question

2. Which of the following statements is TRUE (1 mark)?
a) Selection bias can be reduced by increasing sample size.
b) Loss to follow-up in a cohort study can bias findings either towards the null (reducing the magnitude of the true association) or the opposite (over-estimating the magnitude of the true association).
c) Systematic error can be reduced by taking repeated measurements.
d) Non-differential misclassification of exposure or outcome usually biases study findings away from the null (towards finding an association between exposure and outcomes when there is a smaller or no true association).
e) In a case-control study, epidemiologists are usually most concerned that errors in outcome measurement may differ between cases and controls.

3. Which of the following methods can be used to control for confounding in the analysis stage of an epidemiologic study (1 mark):
a) Multivariable modelling
b) Matching
c) Stratification
d) Restriction
e) All of the above, a through d
f) Choices a, c and d

4. Which of the following is NOT a characteristic of an effect modifier (1 mark):
a) Different effects in different strata
b) Part of nature/reality
c) Increases our knowledge of mechanism of disease
d) Needs to be prevented in study design
e) Informs targeted public health action

Answer 1

(2)

Correct option:

(b) Loss to follow-up in a cohort study can bias findings either towards the null (reducing the magnitude of the true association) or the opposite (over-estimating the magnitude of the true association).

Explanation:

Loss to follow up leads to bias and so it is necessary to mainintain high follow up rates by using techniques such as enrolling motivated subjects, using subjects who are easy to track etc.

(3)

Correct option:

(e) All of the above, a through d

Explanation:

Methods that can be used to control for confounding in the analysis stage of an epidemiologic study are: randomization, restriction, matching, stratification, adjustment and multi-variate analysis.

(4)

Correct option:

(e) Informs targeted public health action

Explanation:

Effect modifier modifiesthe observed effect of a risk factor on disease status. So,if an effect isreal but the magnitude of the effect is different for different groups of individuals, we can conclude the presence of effect modifier.