9. MTB Step 3 - Sensitivity & Specificity, NPV & PPV, ARR & RRR) Flashcards

Cards Complete:

1
Q

SENSITIVITY & SPECIFICITY

Does the Sensitivity & Specificity of a test change based on the Prevalence, or Rate of Disease in a community?

A

NOOOOOO!!!!!

Sensitivity & Specificity start with the Disease

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

CALCULATION

What is the equation for Sensitivity?

A

Sensitivity = TP / TP + FN

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

CALCULATION

What is the equation for Specificity?

A

Specificity = TN / TN + FP

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

DEFINITION

What is the definition of Sensitivity?

A

SENSITIVITY

The likelihood that patients WITH the disease will test POSITIVE

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

DEFINITION

What is the definition of Specificity?

A

SPECIFICITY

The likelihood that patients WITHOUT the disease will test NEGATIVE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

CALCULATION

Positive Predictive Value (PPV)

A

PPV = TP / TP + FP

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

CALCULATION

Negative Predictive Value (NPV)

A

NPV = TN / TN + FN

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

CALCULATION

What is the equation for Number Needed to Treat (NNT)?

A

NNT = 1 / ARR

  • where Attributable Rate Reduction (ARR) = (C/C+D) - (A/A+B)*
  • where Risk with Treatment = A/A+B*
  • where Risk of NO Treatment = C/C+D*
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

CALCULATION

What is the equation for Number Needed to Harm (NNH)

A

NNH = 1 / ARI

  • where Attributable Risk Increase (ARI) = (C/C+D) - (A/A+B)*
  • where Risk with Treatment = A/A+B*
  • where Risk of NO Treatment = C/C+D*
  • (a.k.a. Attributable Risk)*
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

SAID ANOTHER WAY

What are (5) different ways of explaining Sensitivity?

A
  1. All the people WITH a disease should have a POSITIVE test.
  2. A NEGATIVE result of a highly sensitive test will EXCLUDE a disease.
  3. If the test is perfectly sensitive, there will be NO False Negatives.
  4. A NEGATIVE test rules a disease OUT (SnOut)
  5. If you HAVE the disease, will you have a POSITIVE test?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

SAID ANOTHER WAY

What are (6) different ways of explaining Specificity?

A
  1. All the people WITHOUT a disease should have a NEGATIVE test.
  2. A POSITIVE result of a highly specific test will INCLUDE a disease.
  3. If the test is perfectly specific, there will be NO False Positives.
  4. A POSITIVE test rules a disease IN (SpIn)
  5. If you DON’T have the disease, will you have a NEGATIVE test?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

CALCULATION

What is the equation for Absolute Risk Reduction (ARR)?

A

ARR = (A / A + B) - (C / C + D)

  • where the Risk of exposure = A/A+B*
  • where the Risk of no exposure = C/C+D*
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

CALCULATION

What is the equation for Relative Risk Reduction (RRR)?

A

RRR = ARR / (C / C + D)

where ARR = (A/A+B) - (C/C+D)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

CALCULATION

What is the equation for Relative Risk (RR)?

A

RR = (A / A + B) / (C / C + D)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

EXAMPLE

What is the Absolute Risk Reduction (ARR) if the following are true?:

  • The Mortality Rate for an Anterior Wall MI with NO treatment is 40% (40 out of 100 will die within first year post-MI).
  • The Mortality Rate for an Anterior Wall MI with Angioplasty in first 90 mins of arrival in ED is 20% (20 out of 100 will die within first year post-MI)
A

ARR = (A / A + B) - (C / C/ + D)

  • where Risk with NO treatment = A/A+B*
  • where Risk with Angioplasty = C/C+D*

Therefore…

ARR = 40% - 20% = 20%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

NEGATIVE & POSITIVE PREDICTIVE VALUES (NPV & PPV)

Does the NPV & PPV of a test change based on the Prevalence, or Rate of Disease in a community?

A

YEEEEESSSS!!!!

NPV & PPV start with the Test

23
Q

DEFINITION

What is the definition of Negative Predictive Value (NPV)?

A

NPV

If you have a NEGATIVE test, what is the likelihood you really DON’T have the disease?

24
Q

DEFINITION

What is the definition of Positive Predictive Value (PPV)?

A

PPV

If you have a POSITIVE test, what is the likelihood you really DO have the disease?

25
Q

DEFINITION

What is the definition of Absolute Risk Reduction (ARR)?

A

ARR

The percentage decrease in the risk of death/disease from treatment compared with 100% of the people in a population.

26
Q

DEFINITION

What is the definition of Relative Risk Reduction (RRR)?

A

RRR = ARR / (C / C + D)