Lecture 8-10: Measures Of Association Flashcards

1
Q

What are the main measures of association?

A

Exposed/Treated vs. Non-Exposed/Non-Treated

Disease/Outcome vs. Non-diseased/No-Outcome

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

What is Counterfactual Theory?

A
  • The Outcome of Something “Not” occurring.
  • For example the counterfactual for smoker’s as estimated by using Non-Smokers.
  • Requires assumption of EXCHANGEABILITY. And EXCHANGEABILITY = COMPARABILITY with respect to the exposed/treated group.
  • Can also be stated as the Outcome of something Not occurring.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How would one find the Absolute Difference in a descriptive measure of association?

A

Absolute Difference is done by adding or subtracting.
So if Males had 45 surgeries and Females had 17 surgeries,
You could say Males had 28 more surgeries or Females had 28 fewer

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

What are two ways to find the Relative Difference amont measures of association?

A

Ratio of Frequencies (Counts)
Or Ratio of Proportions
So if males had 45 surgeries and Females had 17 surgeries,
Ratio of Frequencies: Males had 2.6x number of surgeries Females Did
Ratio of Proportions: (additional totals would be needed: See slide 8)
Males: 45/50 total surgeries in males, (90%)
Females: 17/39 total surgeries in females (44%)
Note: with Ratio of Proportions, we would say males have a (90/44) 2.04x greater Proportion of having surgery.

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

How does absolute difference compare with a relative difference?

A

Relative Difference is what most companies use…because Absolute Difference looks smaller.

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

Define Risk as a probability of Outcome

A

Risk is a type of Proportion
Easiest description of this is (Treated/Exposed divided by (Treated/Exposed Plus Non-Treated/Non-Exposed))

Or (A/(A+B))
This can also be described as Incidental Risk

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

Define Absolute Risk Reduction

A
  • The Absolute Risk Reduction, or ARR, or just AR, defines the risk difference of the outcome attributable to exposure difference between groups. So basically this takes the Risk Probability of Exposed minus Probability of Non-Exposed (Absolute value always positive)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the Relative Risk Reduction?

A

Relative Risk Reduction, or RRR

is the ARR / (Probability Risk of Non-exposed)

So if the result was 23%, we would say relative to the placebo, this drug drops the risk by 23%.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Give the definition of
Number Needed to Harm
Number Needed to Treat

A

NNT or NNH is the number of patients needed to treat before target outcome hypothetically occurs.

To do this, we take (1/ARR).
With a 3.8% ARR, the NNT would be 27

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What’s Risk Ratio Again?

A

RR:

Risk of Outcome In Exposed / Risk of Outcome in Nonexposed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What does the value of the ratio say about the outcome of the exposed group?

A
  • All Ratios have the property that if ratio is 1.0, then the event/outcome is equally likely for both groups
  • If Ratio is >1.0 then the event/outcome is more likely to occur in the comparison group (numerator group)
  • If Ratio is
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are synonyms for the risk ratio?

A

Hazard Ratio, or Odds Ratio

HR, or OR

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What type of numeral thing do we give a ratio based on it’s value

A
  • If the Odds are between 99% less and 99% more, number should be presented as a fraction.
  • If the Odds are 1.0, we stick with the phrase no difference or change in outcome
  • If the Odds are 2.0 or higher, we present it as a times greater (2x greater)
  • Remember, evaluate the outcome carefully. If exposed outcome is 80% compared to non-exposed, then the risk reduction is 20% Reduced Risk
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

When preparing your “statement/report” of the findings, what 3 things should you be sure to provide?

A
  1. Group Comparison: This Drug was being compared to This Placebo OR (careful) This Placebo is being compared to This Drug
  2. Direction of Words: Increased, or Decreased
  3. Magnitude: 80% greater, 30x greater, etc.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Give 2 ways to find the RRR

A

Way #1:
ARR / RRunexposed

Way #2:
Possibly by taking (1 - RR). Not positive about this one.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are Odds generally defined as?

A

The frequency of an outcome Occurring vs. Not Occurring.

So within the exposed group, the Odds would be:
Number of Cases in those Exposed) / (Number of Cases in Unexposed

17
Q

What is the Odds Ratio (OR)

A

Now we’re taking the:
(Odds of Exposure In Diseased) / Odds of Exposure in Non-Diseased

Alternatively, cross multiplying:
(A * D) / (B * C) supposedly works too.

Remember, there IS a difference between odds, and odds ratio.

18
Q

Wouldn’t be a bad idea to practice drawing out that standard 2 x 2 box. It might help visualize some problems.

A

Do It!