CK040 - Diagnosis Flashcards

1
Q

What is the odds at the ‘no treat-treat threshold’ ?

A

Harm / Benefit

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

What is the probability at the ‘no treat-treat threshold’ ?

A

Harm / (Harm + Benefit)

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

What is the ‘net harm’ ?

A

Treatment if no disease (so ‘difference in outcome with/without treatment if disease is NOT present’)

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

What is the ‘net benefit’ ?

A

Treatment if disease (so ‘difference in outcome with/without treatment if disease is present’)

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

What is the ‘sensitivity’ ?

A

True Positive Rate
(so the probability of getting a positive test result, GIVEN disease is present)

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

What is the ‘specificity’ ?

A

True Negative Rate
(so the probability of getting a negative test result, GIVEN disease is NOT present)

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

What is the ‘PPV’ ?

A

Probability of having the disease, GIVEN a positive test result

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

What is the ‘NPV’ ?

A

Probability of having NO disease, GIVEN a negative test result

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

If you want to make sure you do not miss any diagnosis, what should be high?

A

Sensitivity

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

What is the Bayes’s formula for tests?

A

Posterior odds = Prior odds x Likelihood Ratio

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

What are the formulas for the ‘Likelihood Ratio’ ?

A
  • p(R | D+) / p(R | D-), where R is a certain test result
  • TPR / FPR for a positive test result
  • FNR / TNR for a negative test result
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What does a ‘LR > 1’ mean?

A

That you are more likely to find this test result in diseased

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

What does a ‘LR < 1’ mean?

A

That you are more likely to find this test result in non-diseased

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What happens to the thresholds when the specificity gets bigger?

A

‘No treat-test threshold’ gets smaller

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What happens to the thresholds when the sensitivity gets bigger?

A

‘Test-treat threshold’ gets smaller

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

What is on the axis of a ‘ROC curve’ ?

A
  • x-axis –> FPR (1 - specificity)
  • y-axis –> TPR (sensitivity)
17
Q

What is the KEY step in constructing a ‘ROC curve’ ?

A

Ordering the test results according to increasing Likelihood Ratio

18
Q

What is the assumption of ‘conditional independence’ ?

A

Both for presence & absence of disease, the probability of combination of results is the same as the product of the results independenctly

19
Q

When combining tests, what happens to the ‘sensitivity’ and ‘specificity’ when we us ‘OR’ ?

A
  • Sensitivity goes up
  • Specificity goes down
20
Q

When combining tests, what happens to the ‘sensitivity’ and ‘specificity’ when we us ‘AND’ ?

A
  • Sensitivity goes down
  • Specificity goes up
21
Q

What is ‘calibration’ ?

A

Is our model predicting what we think it should?

22
Q

How to check for ‘calibration’ ?

A

Plot the predicted probabilities (x-axis) against the observed probabilities (y-axis). This line should be linear, and go through the origin.

23
Q

What is ‘discrimination’ ?

A

How good does the prediction model discriminate diseased from non-diseased?

24
Q

How to check for ‘discrimination’ ?

A

Use the AUC of the ROC-curve

25
Q

What happens on the ROC curve when there is a high ‘prior odds’?

A

You move to the top right (because the LR decreases)

26
Q

What happens on the ROC curve when there is a ‘large loss due to FP’?

A

You move to the bottom left (because LR increases)

27
Q

What happens on the ROC curve when there is a ‘large loss due to missed diagnosis’?

A

You move to the top right (because LR decreases)

28
Q

What is ‘verification bias’ ?

A

Selective use of the reference test conditional (based) on the index test result

29
Q

What is ‘test-review bias’ ?

A

When the interpretation of the results of a test is influenced by knowledge of prior tests

30
Q

What is ‘uninterpretability bias’ ?

A

When tests that yield uninterpretable or inconclusive results are excluded from the analysis