Basic Epidemiology End of term MCQ
1. A cohort study was conducted to assess the association between exposure
to a certain risk factor and the development of two diseases: A and B. The
investigators found a rate difference of 5 per 1000 person-years for disease
A and a rate difference of 30 per 1000 person-years for disease B
comparing “exposed” people to those who are “unexposed”. Which of the
following are true?
a) for both diseases, the rate in the exposed is greater than the rate in the
unexposed
b) the rate ratio for disease A must be greater than the rate ratio for disease
B
c) the rate ratio for disease B must be greater than the rate ratio for disease A
d) among exposed individuals, more cases of disease A are attributable to
the exposure than cases of disease B
Correct answer:
a) Feedback:
a) Correct: both attributable rates indicate an excess rate of disease in the
exposed group
attributable to the exposure.
b) Incorrect: we cannot deduce anything about the magnitude of the rate ratio
(RR) for either disease from the information given. The RR for disease A may be
less than, more than, or equal to the RR for disease B.
c) Incorrect (see b)
d) Incorrect: the rate difference for disease B is greater than that for disease
A. Thus the attributable risk for exposure is greater for disease B than for
disease A.
, 2. In a study of obesity as a risk factor for development of coronary heart
disease (CHD), the Population Attributable Fraction (PAF) is interpreted as
estimating:
a) the proportion of the population who develop CHD
b) the proportion of CHD among obese members of the population attributable to
obesity
c) the proportion of CHD in the population that could be avoided if obesity was
eliminated
d) the risk of CHD in the whole population minus the risk of CHD among
members of the population who are not obese
Correct answer:
c) Feedback:
a) Incorrect: this is the risk of CHD in the population
b) Incorrect: this is the attributable fraction: (risk in exposed – risk in
unexposed)/risk in exposed
c) Correct
d) Incorrect: this is the Population Attributable Risk (PAR)
3. Which of the following is true for a Risk Ratio?
a) it compares the proportion who develop disease among the
exposed with the proportion who develop disease among the
unexposed
b) it can never have a value less than 1.0
c) it takes into account that individuals in a study can be at risk of disease
for different lengths of time
d) if its 95% CI excludes 1.0 there is conclusive evidence for a causal
association
Correct answer:
a) Feedback:
a) Correct
b) Incorrect: the RR can be less than 1, for example when an exposure protects
against an outcome (e.g. a vaccine or a health education programme)
c) Incorrect: The Risk of disease (and hence the Risk Ratio) assumes that each
individual is at risk for the same amount of time. The Rate of disease (and hence
the Rate Ratio) takes into account that individuals may be at risk for different
amounts of time (due to development of disease, death, loss to follow-up, etc.)
This is because the denominator of a Rate is the total person-time at risk rather
than the total number of individuals at risk at the start of the study.
1. A cohort study was conducted to assess the association between exposure
to a certain risk factor and the development of two diseases: A and B. The
investigators found a rate difference of 5 per 1000 person-years for disease
A and a rate difference of 30 per 1000 person-years for disease B
comparing “exposed” people to those who are “unexposed”. Which of the
following are true?
a) for both diseases, the rate in the exposed is greater than the rate in the
unexposed
b) the rate ratio for disease A must be greater than the rate ratio for disease
B
c) the rate ratio for disease B must be greater than the rate ratio for disease A
d) among exposed individuals, more cases of disease A are attributable to
the exposure than cases of disease B
Correct answer:
a) Feedback:
a) Correct: both attributable rates indicate an excess rate of disease in the
exposed group
attributable to the exposure.
b) Incorrect: we cannot deduce anything about the magnitude of the rate ratio
(RR) for either disease from the information given. The RR for disease A may be
less than, more than, or equal to the RR for disease B.
c) Incorrect (see b)
d) Incorrect: the rate difference for disease B is greater than that for disease
A. Thus the attributable risk for exposure is greater for disease B than for
disease A.
, 2. In a study of obesity as a risk factor for development of coronary heart
disease (CHD), the Population Attributable Fraction (PAF) is interpreted as
estimating:
a) the proportion of the population who develop CHD
b) the proportion of CHD among obese members of the population attributable to
obesity
c) the proportion of CHD in the population that could be avoided if obesity was
eliminated
d) the risk of CHD in the whole population minus the risk of CHD among
members of the population who are not obese
Correct answer:
c) Feedback:
a) Incorrect: this is the risk of CHD in the population
b) Incorrect: this is the attributable fraction: (risk in exposed – risk in
unexposed)/risk in exposed
c) Correct
d) Incorrect: this is the Population Attributable Risk (PAR)
3. Which of the following is true for a Risk Ratio?
a) it compares the proportion who develop disease among the
exposed with the proportion who develop disease among the
unexposed
b) it can never have a value less than 1.0
c) it takes into account that individuals in a study can be at risk of disease
for different lengths of time
d) if its 95% CI excludes 1.0 there is conclusive evidence for a causal
association
Correct answer:
a) Feedback:
a) Correct
b) Incorrect: the RR can be less than 1, for example when an exposure protects
against an outcome (e.g. a vaccine or a health education programme)
c) Incorrect: The Risk of disease (and hence the Risk Ratio) assumes that each
individual is at risk for the same amount of time. The Rate of disease (and hence
the Rate Ratio) takes into account that individuals may be at risk for different
amounts of time (due to development of disease, death, loss to follow-up, etc.)
This is because the denominator of a Rate is the total person-time at risk rather
than the total number of individuals at risk at the start of the study.
