Week 5 Flashcards

1
Q

What is probability

A

the probability of the occurrence of an event is the
proportion of times that event would (or does) occur in a
large number of repeated trials.

Values ranges from 0(never happen) to 1 (always happen)

probability = proportion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How to calculate risk?

A

D/N

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How to calculate risk difference?

A

risk difference = risk in exposed - risk in non exposed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Risk ratio interpretation

A

RR = 1 occurs when there is no association between exposure
and disorder.

RR > 1 occurs when the risk is higher among the exposed

RR < 1 occurs when the risk is lower among the expose

The further the RR is from 1, the stronger the association
between exposure and disorder

nb a RR is always a positive number

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Can you estimate risk in case-control studies?

A

No, because, in case-control studies, the cases are an unknown
fraction of the population, it is not possible to estimate risk.

Instead, you can estimate odds of exposure by disorder.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Interpretation of Odds Ratio

A

OR = 1 occurs when there is no association between exposure
and disorder.

OR > 1 occurs when the risk is higher among the exposed.

OR < 1 occurs when the risk is lower among the exposed.

The further the OR is from 1, the stronger the association
between exposure and disorder.

nb an OR is always a positive number

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Why use odds ratios?

A

common mistake is to interpret the odds ratio as if it were a
risk ratio

• this is ok for a rare outcome (i.e., probability of event not
occurring is close to 1), when the odds ratio approximates the
risk ratio (as the odds approximate the risk)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly