Deck 5: Study Design

Question 1

In both cohort studies and case-control studies, the crude) measure of excess risk can be calculated by setting up a 2x2 table.

Answer 1

True

Question 2

The following data are based on a study of 375 people sampled in the following way: 125 people who have been diagnosed with throat cancer and 250 people who have not been diagnosed with throat cancer were selected from hospital records. Among other exposures, these individuals with and without throat cancer were asked about their smoking history. What is this study design?

A) Cross sectional study
B) Retrospective cohort study
C) Case-control study
D) Randomized trial

Answer 2

Case-control study

Question 3

The Framingham Heart Study, which has followed 3 generations of Framingham, Massachusetts since 1948 to assess the risk factors for heart disease and other health outcomes, is an example of what type of epidemiological study?

A) Ecological study
B) Cross-sectional study
C) Cohort study
D) Randomized trial

Answer 3

Cohort study

Question 4

A cross-sectional study is one in which both exposure and outcome are assessed at the same point in time, providing a snapshot of experience rather than information over time.

Question 5

If the results of your study are the opposite of most previous studies published, which aspect of causal inference should you be concerned about?

A) Temporal sequence
B) Strength of association
C) Dose-response
D) Consistency

Question 6

What is the relative risk for throat cancer among smokers compared to non-smokers based on the numbers below?

A) 0.75
B) 1.5
C) 2.0
D) 3.5

Question 7

The correct measure of excess risk to be used in case-control studies is the relative risk.

Question 8

We are using cross-sectional data to investigate whether heavy drinking increases the risk of unemployment. Which of the following aspects of causal inference concerns you based on this information?

A) Temporal sequence
B) Strength of association
C) Dose-response
D) Consistency

Question 9

If increasing levels of exposure cause increasing levels of the outcome, we would say that there is evidence of biological plausibility when assessing causal inference.

Question 10

A prospective cohort study is one that uses historical exposure information - for example, employment or medical records - to construct the exposed and unexposed groups.

Question 11

In order to investigate whether nutritional information combined with telephone counseling is better at increasing weight loss among women who are obese than telephone counseling alone, researchers randomly assign one group of obese women to the counseling only group and assign another group of obsess women to the counseling plus nutritional information group. What type of epidemiological study is this?

A) Cross sectional study
B) Retrospective cohort study
C) Case-control study
D) Randomized trial

Question 12

Which of the following epidemiological study designs makes use of group-level rather than individual level) data?

A) Ecological study
B) Cross-sectional study
C) Cohort study
D) Case-control study

Question 13

One reason that observational studies are common in public health research is that many exposures in which we are interested are not ethical to assign people to experience like radiation, smoking during pregnancy, or being in a car crash).

Question 14

Making individual-level inferences about the relationship between exposure and outcome based on group-level data is called confounding.

Question 15

Bias is the ability to correctly identify cases people with the outcome) as cases in a case- control study.

Question 16

A research team is interested in whether warfarin a medication to prevent clotting) reduces the risk of stroke. They ask warfarin's manufacturer for the number of annual warfarin prescriptions in the US from 1990-2000 and use data from death certificates to assess the numbers of deaths due to stroke in the US during the same time period. The resulting graph showed that the number of annual warfarin prescriptions increased in the US from 1990-2000 while stroke mortality in the US declined from 1990-2000. What type of study is this?

A) Ecological study
B) Cross-sectional study
C) Case-control study
D) Randomized trial

Question 17

Data from the Behavioral Risk Factor Surveillance System - a telephone survey that asks adults about a variety of health behavior and health conditions at a single point in time - could be used for what type of epidemiological study?

A) Ecological study
B) Cross-sectional study
C) Cohort study
D) Randomized trial

Question 18

If you calculate a relative risk of 1.0 RR=1.0) in a cohort study, what is the correct interpretation?

A) Exposed individuals and unexposed individuals are equally likely to experience the outcome.
B) Exposed individuals are twice as likely 100% more likely) to experience the outcome compared to unexposed individuals.
C) Exposed individuals are half as likely 50% less likely) to experience the outcome compared to unexposed individuals.
D) It is not possible to assess the excess risk among the exposed because there are no unexposed people with the outcome.

Question 19

Observational studies are considered to be better than experimental studies because they are not plagued by confounding.

Question 20

An observational epidemiological study is one in which the researcher randomly assigns participants to an exposure group.