Study Design

Question 1

What is selection bias?

Answer 1

Error in the sampling of individuals for a clinical study. Ex: Subjects chosen for the case and control group differ

Question 2

What is observational bias?

Answer 2

Error in the recording of the factors of the study. Ex: inaccurate recording of a patient’s risk factor

Question 3

What is recall bias?

Answer 3

Systematic error caused by differences in the accuracy or completeness of recollections of study participants regarding past events or experiences

Question 4

What is interviewer bias?

Answer 4

Ex: interviews are not conducted in a uniform manner for all participants

Question 5

What are the disadvantages of case reports/series?

Answer 5

Does not provide explanation other than conjecture and does not establish causality of association

Question 6

What are the advantages of case reports/series?

Answer 6

Hypothesis can be formed. Easy to perform and are inexpensive

Question 7

Case-Control study

Answer 7

Study exposure in those with and without outcomes of interest

Question 8

What kind of studies are retrospective studies?

Answer 8

Observational

Question 9

What category of study should be chosen to study exposures in rare diseases that take long periods to develop?

Answer 9

Observational

Question 10

What assumptions must be made to minimize bias in observational studies?

Answer 10

-Cases are selected to be representative of those who have the disease
-Controls are representative of general population
-Information is collected from case and controls in the same way

Question 11

What kind of bias are case-control studies most vulnerable to?

Answer 11

-Observational and recall bias
-Selection bias

Question 12

What is the most common measure of association used for case-control studies?

Answer 12

OR or RR

Question 13

Can RR be calculated from prospective or case-control studies?

Answer 13

Prospective

Question 14

If RR <1 what can be the interpretation?

Answer 14

Risk of disease is lower in the exposed group

Question 15

If RR =1 what can be the interpretation?

Answer 15

No association

Question 16

If RR/OR >1 what is the association?

Answer 16

Positive association

Question 17

Calculate the OR. A total of 7 women are exposed to PPA. 1126 of women included in the study were not exposed to PPA. Out of the women who were not exposed to PPA, 377 experienced hemorrhagic stroke. Out of the women who were exposed to PPA women experienced hemorrhagic stroke.

Answer 17

OR = (6/377)/(1/749)=12

Question 18

What is the most common situation in which non inferiority trials are used?

Answer 18

When placebo administration is not possible for ethical reasons

Question 19

Do intention-to-treat vs per-protocol analysis have more information about treatment efficacy?

Answer 19

Per-protocol

Question 20

Do intention-to-treat or per-protocol analysis have greater external validity?

Answer 20

Intention-to treat

Question 21

What is the difference between systemic reviews and meta-analysis?

Answer 21

Systemic reviews perform comprehensive lit search and synthesize literature on a topic. Meta-analysis use mathematical/statistical techniques to summarize the results of evaluated studies

Question 22

What is the calculation for NNT?

Answer 22

NNT=1/ARR *generally round to the next highest whole number
ARR= (experimental event rate)-(control event rate)

Question 23

What must apply to a study in order to properly calculate a NNT?

Answer 23

-Outcomes must be dichotomous (yes/no, alive/dead)
-Outcomes must be statistically significant

Question 24

What is the difference between a cost-effectiveness analysis and cost-utility analysis?

Answer 24

Cost-utility analysis assigned utility weights to outcomes (Ex: quality-adjusted life-years) and cost-effectiveness analysis assigned cost per unit health outcome(ex:number of symptom free days)