Diagnostic Tests

Question 1

What is the difference between dichotomous tests vs. continuous tests?

Answer 1

Dichotomous only gives you 2 poss answers - normal or not, present or absent, etc. A continuous test has a continuum of possible answers, but still needs to be classified as pos or neg (bloodwork for ex.)

Question 2

Immunological diagnostic tests use what to measure what? What are some examples? Is it dichotomous or continuous?

Answer 2

use antigen to measure antibodies, or antibodies to measure antigen. SNAP FIV/FeLV combo test. Dichotomous (only pos or neg)

Question 3

What are some examples of pathogen detection tests?

Answer 3

Virus isolation/bacterial culture, direct visualization, PCR based tests to detect nucleic acid.

Question 4

Do diagnostic tests measure the disease itself?

Answer 4

Not usually. They measure something that is needed in certain quantity to confirm dz (this something is called a test value)

Question 5

What is a cut off value?

Answer 5

A value where values that fall on one side are considered positive, and values on another side are considered negative.

Question 6

How are cut off values determined?

Answer 6

experimentally. The value gone with is one that minimizes false positives/false negatives.

Question 7

What are the steps in determining cut off values?

Answer 7

test animals proven (w/ gold standard test) to have and not have the dz. Then, pick a value that best separates the 2 groups.

Question 8

What are some issues with gold standard tests?

Answer 8

they are often very labor intensive, impractical, highly invasive, slow, and/or expensive.

Question 9

What is sensitivity?

Answer 9

how many sick animals a test can ID as sick (be sensitive to animals with dz)

Question 10

What is specificity?

Answer 10

how many healthy animals a test can ID as healthy.

Question 11

T/F: Sensitivity and specificity are proportions.

Answer 11

True.

Question 12

Will sensitivity give you false negatives or false positives?

Answer 12

Sensitivity will give you false NEGATIVES. If low sensitivity, then sick animals were not identified as sick.

Question 13

When can sensitivity be low?

Answer 13

When there are very few/small amounts of whatever the test measures, or when samples are degraded before testing.

Question 14

Will specificity give you false negatives or false positives?

Answer 14

False POSITIVES. If specificity is low, it can incorrectly ID healthy animals as sick.

Question 15

When can specificity be low?

Answer 15

when there is cross reactivity or when samples are contaminated.

Question 16

What does specificity tell you about sick animals?

Answer 16

Nothing. It only tells you how a test performed on healthy animals.

Question 17

Why does a very sensitive test rule dz out?

Answer 17

Because you are confident that it correctly identifies sick animals as such. So, if a highly sensitive test comes back NEGATIVE, you can be confident that animal does not have that dz.

Question 18

Why does a very specific test rule dz in?

Answer 18

because you are confident that it identifies healthy animals as such. So, if a highly specific test comes back as positive, then you know the animal is probs sick.

Question 19

When would you care more about sensitivity than specificity?

Answer 19

When you don’t want any false negatives, and false positives don’t end up in the animal dying or something (so like importing animals into a country).

Question 20

What is PPV?

Answer 20

Positive predictive value. The probability that test positive animal is truly diseased

Question 21

What is NPV?

Answer 21

negative predictive value. The probability that a test negative animal is truly disease free.

Question 22

With PPV, are you dealing with diseased or healthy animals?

Answer 22

BOTH. You are dealing with all animals that tested positive. So, true and false positives.

Question 23

With NPV, are you dealing with diseased or healthy animals?

Answer 23

BOTH. You are dealing with all animals that tested negative. So, true and false negatives.

Question 24

PPV and NPV are determined by what?

Answer 24

the Se and Sp of the test, as well as the prevalence of disease in population. PPV increases as prevalence increases, and NPV decreases as prevalence increases.