Diagnostic Tests

Question 1

What is sensitivity?

Answer 1

Proportion of affected individuals correctly identified.

True positives/ True positive + false negative

Question 2

What is specificity?

Answer 2

Proportion of unaffected individuals correctly identified.

True negatives/ true negative + false positive

Question 3

How to increase sensitivity?

Answer 3

Decrease cutoff score to identify more people.

Question 4

What are the benefits of high sensitivity tests?

Answer 4

Rule out diseases

Question 5

What are the benefits of high specificity tests?

Answer 5

Rule in diseases

Question 6

How to increase specificity?

Answer 6

Increase cutoff score.

Question 7

What is apparent prevalence?

Answer 7

Proportion of all tested who are tested positive.

True positive + false positive/ n

Question 8

What is true prevalence?

Answer 8

Proportion of all tested who are true positives.

True positive + false negative/ n

Question 9

What is positive predictive value?

Answer 9

Proportion of positive tests

True positives/true positives + false positives

Question 10

What is negative predictive value?

Answer 10

Proportion of negative test that are correct

True negatives/true negative + false negatives

Question 11

What is a likelihood ratio?

Answer 11

Summarises the performance/value of a test, using sensitivity and specificity

This compares the likelihood of a specific test result in people with a disease compared to people without the disease.

Question 12

What is positive likelihood ratio?

Answer 12

Sensitivity/ (1-specificty)

Probability that a person with the disease tested positive compared to person without disease testing positive.

Question 13

How can the value of the LR+ be used to determine presence of disease?

Answer 13

High positive likelihood ratio means there is a greater chance you have disease because a positive test is associated with people who have the disease. This should be a LR+ that is greater than 10 to reliably determine disease.

Question 14

How can the value of the LR- be used to determine absence of disease?

Answer 14

Low LR- means there is a greater chance you DON’T have disease with a negative test, because a negative test is associated with with people who don’t have the disease. The LR- should be less than 0.1 to reliably determine absence of disease.

Question 15

What is negative likelihood ratio?

Answer 15

1-sensitivity/specificity

Question 16

How is a disease case defined?

Answer 16

Based on:
Biological results e.g lab reports
Clinical signs or symptoms
Mental health survey response

Question 17

What is a significant likelihood ratio?

Answer 17

A likelihood ratio greater than 1 indicates that the test result is more likely to occur in those with the disease.

Tests with a likelihood ratio further from 1 have a stronger association for discriminating and providing a diagnosis.

Question 18

What does a likelihood ratio of 1 mean?

Answer 18

There is no difference in the probability of the test result between people who have the disease and those unaffected, they are equally likely to test positive or negative therefore the test is not worth performing.

Question 19

What is pre-test probability?

Answer 19

Likelihood that a patient has the disease before the results of a diagnostic test, based on clinical history and background and prevalence in the population.

Question 20

What does a low LR+ indicate?

Answer 20

A low positive likelihood ratio means that LOW amounts of people WITH the disease tested positive, therefore the test is not very accurate for determining disease.

Sensitivity/ 1-specificity.

Question 21

What does a high LR- indicate?

Answer 21

A high negative likelihood ratio indicates that many people WITH the disease tested negative, therefore the test is not useful for ruling out disease.

1-sensitivity/specificity