Lecture 8

Question 1

Sensitivity

Answer 1

The ability of a test to correctly detect a condition.
True positive.

True positive / True positive + False negative

Question 2

Specificity

Answer 2

The ability of a test to exclude subjects without the condition. True negatives.

True negative / True negative + False positive

Question 3

2 tests that make statement about the test in general

Answer 3

Sensitivity and specificity

Question 4

2 tests that makes a statement about what a test means to the patient

Answer 4

Positive and negative predictive value.

Question 5

Positive predictive value

Answer 5

Probability of having the condition when testing positive.

True positive / True positive + False positive

Question 6

Negative predictive value

Answer 6

Probability of not having the condition when testing negative

True negative / True negative + False negative

Question 7

Accuracy

Answer 7

The proportion of all tests providing the correct result

True positive + true negative / All

Question 8

Likelihood ratio of a positive test

Answer 8

Odds of a disease if the test result is positive

Sensitivity / 1 - specificity

Question 9

Likelihood ratio of a negative test

Answer 9

Odds of a disease if the test is negative

1- sensitivity / Specificity

Question 10

10 questions to evaluate a paper claiming to validate a diagnostic or screening test

Answer 10

1. Is the test relevant to clinical practice?
2. Has it been compared with the gold standard?
3. Was an appropriate spectrum of participants included?
   - PPV/NPV are dependent on prevalence.
4. Verification bias avoided?
   - Confirm all subjects were tested with both the gold standard and the new test.
5. Was expectation bias avoided?
   - Do this by blinding the interpreter of the tests.
6. Test reproducible both within and between observers ?
   - Intra observations: same observer, same result.
   - Inter observations: different observers, same results
7. What are the test features?
   - Sensitivity, specificity, PPV, NPV
8. Were confidence intervals provided?
9. Normal range derived from results?
10. Has the test been placed in the context of other similar tests?

Question 11

Intra observations vs inter observations

Answer 11

Intra: Same observer, same result.

inter: Different observer, same result.

Question 12

Confidence interval

Answer 12

Possible range of results within which the true value will lie.

Ideal: narrow and based on large sample size.

Question 13

What happens to the CI with a small sample size?

Answer 13

CI will be sensitive to small changes.

Question 14

Variables (parameters)

Answer 14

What we are measuring or manipulating in a study

Question 15

Two types of variables

Answer 15

Numeric - continuous and discrete

Categorical (non numerical)- ordinal and nominal

Question 16

Continuous variable (branch off numeric)

Answer 16

One that takes on an infinite number of values in a certain range.
Ex: distance, temp, time, some biomarkers.

Question 17

Discrete variable (branch off numeric)

Answer 17

Countable whole number. Ex: 10 people in the room. Not 10.5

Question 18

Dichotomous variable (apart of categorical nominal)

Answer 18

only have 2 values.
Male-female
Employed- non employed

Question 19

Ordinal (apart of categorical)

Answer 19

Distance between categories is unknown.
Usually questionnaires use this.
Ex: Leikert scale.

Question 20

Likelihood ratio

Answer 20

Ratio of the likelihood of a test result in one with disease to likelihood of the same test result in one without the disease.

Preferred way of expressing and comparing the usefulness of different tests.

Defines risk of actually having the condition.
Useful for continuous variables.
Calculated using sensitivity and specificity.

Question 21

Likelihood ratio is calculated using

Answer 21

Sensitivity and specificity

Question 22

Likelyhood ratio defines ____ of having the condition

Answer 22

Risk

Question 23

Clinical prediction

Answer 23

Numerical estimate of chance of person having or developing disease. Can use likelihood ratio to compare pre-test probability to post-test probability.

Question 24

Three stages of developing a clinical prediction rule (standard of care)

Answer 24

1. Establish indecent and combined effects of symptoms, signs, and diagnostic tests.
2. Assess explanatory variables above in different populations to be more generalizable.
3. RCT to assess the impact on patient outcomes applying the rule. Most critical part.

Question 25

ROC graphically evaluates what?

Answer 25

The trade offs between sensitivity (y is sensitivity) and specificity (x is 1-specificity)

Question 26

When looking at ROC graph, how do you know what the overall accuracy of ability of a test to distinguish between two groups?

Answer 26

Look at AUC.

Question 27

Ideal ROC graph

Answer 27

in upper left corner. AUC= 1.00

Question 28

Line of no difference in ROC graph

Answer 28

AUC guessing = 50%

Question 29

Grading ROC AUC

Answer 29

90-100: excellent A
80-90: Good B
70-80: Fair C
60-70: Poor D
50-60: Fail F

Question 30

Bland-Altman PLots

Answer 30

Graphical analysis process used to assess agreement between two methods of measurement.

Requires 2 methods measuring the same characteristic with the same scale.

When plotted, the points should line up along the line of equality (y=x)

Question 31

Bland Altman compared to correlation plots

Answer 31

Not the same. Bland Altman compares the same characteristics with same scale.

Correlation compares two different characteristics on different scales.

Question 32

Name of difference vs mean plot

Answer 32

Bland Altman