2.4.5. Basic Biostats I

Question 1

Experimental vs. Observational Study

Answer 1

Experimental means that the investigators assigned exposure, while observational means that they did not

Question 2

Types of experimental studies

Answer 2

Randomized clinical trial (preferred) and non-randomized clinical trial

Question 3

Types of Analytical Studies

Answer 3

Cohort (exposure → outcome)

Case-Control (outcome → exposure)

Cross-Sectional (exposure & outcome at the same time)

Question 4

Types of Descriptive Studies

Answer 4

NO Comparison Group

Ecologic Study
Case Reports
Case Series

Question 5

Selection of Control Group

Answer 5

Controls represent population at risk from which the case arose

Restrictions:

1. Free of disease of interest at time
2. Must have had opportunity for both exposure & disease

Question 6

Odds Ratio

Answer 6

Odds a case was exposed = A/C

Odds a control was exposed = B/D

Odds ratio = A/C / B/D = AD/BC

Question 7

Interpreting Odds Ratios

Answer 7

OR ranges from 0 to infinity

Question 8

if OR is < 1

Answer 8

decreased risk

Question 9

if OR = 1

Answer 9

equal risk

Question 10

if OR > 1

Answer 10

increased risk

Question 11

If outcome is rare

Answer 11

OR ~ RR

Question 12

Confounding Requirements

Answer 12

1. A risk factor for the outcome
2. Associated with the exposure
3. Not caused by the exposure

Confounding factors distort or mask the true effect of exposure in an epidemiologic study

Question 13

How to address confounding?

Answer 13

1. Matching
2. Stratification
3. Adjustment

Question 14

Features of Cohort Study

Answer 14

1. Start with exposure
2. Look forward in time from the point of exposure to determine outcomes of interest
3. Compare incidence of outcome between exposed and unexposed groups

Question 15

Bias in cohort study

Answer 15

Selection: exposed and unexposed subjects are not equally susceptible; subjects may move from one exposure group to another during follow up; subjects in one group may be more likely to drop out

Measurement: subjects in one group are more likely to have outcome detected

Question 16

What is cohort useful for?

Answer 16

Rare exposures

Question 17

What does cohort measure?

Answer 17

Incidence and relative risk

Question 18

What is case-control good for?

Answer 18

Rare diseases

Question 19

What does case-control measure?

Answer 19

Odds ratio

Question 20

Advantages of Cohort

Answer 20

1. Approximates RCT design
2. Can evaluate multiple outcomes
3. Provides an actual measure of risk of the outcome of interest
4. Can extract incidence and relative risk

Question 21

Disadvantages of Cohort

Answer 21

1. Potential for loss to follow-up
2. Requires large number of subjects
3. Not intended for the study of rare outcomes
4. Takes a long time
5. Expensive

Question 22

Advantages of CC

Answer 22

1. Study of rare diseases
2. Study of disease with long latency
3. Requires relatively few subjects
4. Requires less time than Cohort
5. Allows for evaluation of multiple exposures
6. Well-suited for outbreak investigation

Question 23

Disadvantages of CC

Answer 23

1. Relies on recall of subjects
2. Problems of recall bias
3. Selection of appropriate control group may be difficult
4. Yields the odds ratio which is only an estimate of relative risk
5. Cannot calculate incidence rates