2

1. A _____________ is a type of study in which the investigator takes advantage of existing circumstances that serve effectively as an experiment.

   Case-control study

   Randomized clinical trial

   Natural experiment

   Field trial

1 points

QUESTION 3

1. Match each of the following study types to the appropriate characteristics

   - A. B. C. D.	Compares exposed with nonexposed persons
   - A. B. C. D.	Compares people who have a disease with people who do not have a disease
   - A. B. C. D.	Subjects are randomized to exposure or non-exposure
   - A. B. C. D.	Odds ratios can be calculated

   A.	Both cohort and case-control studies
   B.	Case-control study
   C.	Neither cohort nor case-control studies
   D.	Cohort studies

4 points

QUESTION 4

1. In a case-control study of stroke, where cases are patients treated for stroke at a particular hospital, the best group of controls would come from:

   Anybody from the population who would be treated at that hospital if they had a stroke

   Stroke patients at another hospital

   Diabetes patients treated at that hospital

   People who had died from other causes

1 points

QUESTION 5

1. If you are conducting a case control study and you select your cases based on existing (prevalent) cases of disease instead of incident (newly diagnosed) cases, how is this likely to influence your results?

   Cases are more likely to have shorter survival times

   Cases are more likely to have longer survival times

   Controls will have longer survival times

   The results will be biased away from the null hypothesis

1 points

QUESTION 6

1. Nondifferential misclassification (where cases and controls are equally likely to forget details about their exposure histories) tends to produce estimates of effect that are:

   Underestimated (closer to the null value than they should be)

   Exaggerated (further from the null value than they should be)

   Sometimes exaggerated, sometimes underestimated

   Highly variable

1 points

QUESTION 7

1. When investigating an outbreak of a disease, where patients do not come from a well-defined group of people, which of the following study designs makes the most sense?

   Prospective cohort

   Retrospective cohort

   Case-control

   Case-crossover

1 points

QUESTION 8

1. In a nested case-control study,

   Cases and controls come from a previously defined cohort study

   Cases and controls come from an existing case-control study

   Cases act as their own controls

   None of the above

1 points

QUESTION 9

1. The incidence of disease in exposed divided by the incidence of disease in the unexposed group is known as the:

   Absolute risk

   Attributable risk

   Odds Ratio

   Relative Risk

1 points

QUESTION 10

1. Relative risk can be calculated:

   In cohort studies

   In case-control studies

   In cross-sectional studies

   All of the above

1 points

QUESTION 11

1. All of the following need to be satisfied for the odds ratio to represent a good estimate of the relative risk, EXCEPT:

   Cases are representative of all people with the disease

   Controls are representative of all people without the disease

   The disease does not occur frequently

   The exposure does not occur frequently

1 points

QUESTION 12

1. In aMatched Pairscase control study represented by the following 2x2 table, what is the formula for the odds ratio?

   Controls
   Exposed	Not Exposed
   Cases	Exposed	A	B
   Not Exposed	C	D

   A/D

   AD/BC

   (A/B)/(C/D)

   B/C