EPI 201 Midterm

Question 1

Relationship between CI and IR

Answer 1

CI = 1 - e^(-IR*t)

Question 2

Which interval does a censored person go in?

Answer 2

The one that ends soonest AFTER they are censored

Question 3

What do you need to include with a CI?

Answer 3

Time period

Question 4

What do you need to include with an IR?

Answer 4

Person-time

Question 5

What do you need to include with a prevalence?

Answer 5

Point in time

Question 6

Measures of association

Answer 6

Cumulative incidence ratios/differences, prevalence ratios/differences, incidence rate ratios/differences

Question 7

Traditional definition of a confounder

Answer 7

• Must be associated with exposure in the study base
• Must be a cause or correlate of a cause of the outcome among the unexposed
• Cannot be a downstream consequence of exposure or outcome

Question 8

Definition of a prospective cohort study

Answer 8

Exposures measured for research purposes before the OUTCOME occcurs

Question 9

Study base types

Answer 9

Primary, secondary

Question 10

Types of case-control sampling

Answer 10

Density - IRR, risk-set (subset of density) - IRR, case-crossover - IRR, cumulative incidence - CI (under rare disease assumption), case-cohort - IRR

Question 11

Measures of association for a closed cohort study

Answer 11

IRs, CIs, ORs…

Question 12

Hill Criteria

Answer 12

Strength, consistency, specificity, temporality, biologic gradient, plausibility, coherence, experimental evidence, analogy

Question 13

Popper postulates

Answer 13

Knowledge accumulates by falsification
Scientific theories can only be falsified, never proven
Scientific hypotheses must have empirical content and be falsifiable

Question 14

Problems with random digit dialing

Answer 14

Probability of having a phone may relate to exposure
Probability of answering may relate to exposure
Different ratios of people to phone
Area codes no longer relate to geography

Question 15

Key point with hospital matched controls

Answer 15

The control condition must be UNRELATED to the exposure of interest

Question 16

Attributable risk

Answer 16

CID = CI_E+ - CI_E-

Question 17

Population attributable risk percent

Answer 17

Pr[A=1|Y=1] = (CI_tot = CI_E-)/CI_tot

Question 18

Number needed to treat

Answer 18

1/CID

Question 19

Population attributable risk

Answer 19

CI_tot - CI_E-

Question 20

Attributable rate

Answer 20

IRD

Question 21

Attributable risk percent

Answer 21

(CIR-1)/CIR * 100

Question 22

Measures of association for an open cohort study

Answer 22

Odds, IR, only approx CIs using exponential formula

Question 23

Definition of D-separation

Answer 23

Two variables in a DAG are D-separated if all paths between them are blocked

Question 24

What happens if you condition on a descendant of a collider?

Answer 24

This is just like conditioning on the collider itself – opens a path

Question 25

Problems with ecological studies

Answer 25

Often have poor measure of exposure
Often have no info on who is exposed
No info on if exposed are the ones getting the outcome
No data on individual-level confounding

Question 26

In what kind of study can you directly measure CI?

Answer 26

Closed cohort study (however, you would often use IR in a closed cohort situation due to LTFU, changes in exposure over time, etc). In an open cohort, you cannot measure CI directly, but you can estimate it using the exponential formula.

Question 27

Problems with CI

Answer 27

Tends towards 1
No info on when event occurred
Does not allow for changing population size
Does not allow for time-varying exposures
Events may not be observable due to censoring

Question 28

Probability tree notes

Answer 28

L=0 (0.33) 20

Non-treatment variables in circles

Question 29

Relationship between prevalence and IR

Answer 29

prevalence odds = IR * duration (under steady state)

Question 30

Difference between confounding and selection bias

Answer 30

Confounding occurs when there is a lack of exchangeability between exposed and unexposed groups. Selection bias occurs when we condition on a collider.

Question 31

When is it appropriate to use the NON-exponential formula?

Answer 31

To approximate CI using the IR, you can use the NON-exponential formula C=I*t when the product of I and t is small.