Diagnostic Tests 1

Question 1

Characteristics of reference test

Answer 1

Best available test, as close to gold standard as possible; highest sensitivity, specificity, etc.; usually most expensive, slow, or technically challenging

Question 2

3 real world alternatives to reference tests?

Answer 2

surrogate tests, imperfect tests, or multiple tests with various characteristics

Question 3

Define sensitivity

Answer 3

proportion of disease positive animals that have a positive test result

Question 4

define specificity

Answer 4

proportion of disease negative animals that have a negative test result

Question 5

will a high sensitivity test will have: a low risk of false negatives or a high risk of false negatives

Answer 5

a high sensitivity test will have a low risk of false negatives

Question 6

will a high specificity test have a low or high risk of false positives

Answer 6

a high specificity test will have a low risk of false positives

Question 7

define accuracy

Answer 7

ability of test to give true measure of what it is testing; average of repeat testing is close to truth

Question 8

define precision

Answer 8

how consistent the results of a test are when repeated (not necessarily close to the true value)

Question 9

define agreement

Answer 9

how well two tests agree, ie. give same value or outcome (doesn’t mean that either test is accurate, and can refer to two people assessing the same parameter)

Question 10

what is Kappa used for

Answer 10

used to test agreement (beyond just chance) between two or more tests when the tests are categorical

Question 11

define analytic sensitivity

Answer 11

lowest concentration an assay can detect

Question 12

define analytic specificity

Answer 12

capacity of an assay to react to only one compound

Question 13

define positive predictive value and give 2 factors that influence it

Answer 13

proportion of animals that test positive which truly have the disease. influenced by pre-test probability for patient (or prevalence in population) and specificity of test or cut point being used

Question 14

define negative predictive value and give 2 factors that influence it

Answer 14

proportion of animals that test negative which truly do not have the disease. influenced by pre-test probability for patient (or prevalence in population) and specificity of test or cut point being used

Question 15

if prevalence decreases while Se and Sp remain the same, what is the effect on PPV? on NPV?

Answer 15

if prevalence decreases, the positive predictive value decreases and the negative predictive value increases