Exam #1: Appraising Diagnostic Studies

Question 1

Review the steps in EBM.

Answer 1

1) Ask answerable questions
2) Search for evidence
3) Critical appraisal of evidence
4) Integration of evidence with expertise and patient values + application
5) Evaluate results

Question 2

What are diagnostic tests used for?

Answer 2

• Screening a population
• Diagnosis in an individual
• Patient follow-up and management

Question 3

What are the disadvantages of diagnostic testing?

Answer 3

• May be expensive
• Can risk morbidity or mortality
• May cause discomfort

*****False positives can cause psychological distress and lead to risk from unnecessary therapy

Question 4

Are studies of diagnosis random?

Answer 4

NO

Question 5

How are diagnostic studies designed?

Answer 5

1) Subject population to test
2) See what happens
3) Compare to reference standard

Question 6

What are the questions you need to ask when appraising a diagnostic study?

Answer 6

1) What are the results?
2) Is the study valid?
3) Are the results in the study important?
4) Will the results help my patients?

Question 7

What are the questions you need to ask to determine if the study is valid?

Answer 7

1) Is the research question clearly defined?
2) Presence or absence of disorder confirmed by “gold standard?”
3) Was the test evaluated on an appropriate spectrum of patients?
4) Reference standard applied to all patients?

*****Also need to know how the investigators knew the disorder was present AND that the comparison was blinded.

Question 8

How should you determine the clinical importance of a diagnostic test?

Answer 8

Determine the sensitivity and specificity of the test

*****An ideal test will produce only true positives and true negatives

Question 9

What is sensitivity?

Answer 9

How well a test finds true positives

Question 10

What is specificity?

Answer 10

How well a test finds true negatives

Question 11

What is a likelihood ratio?

Answer 11

Likelihood of a result in a patient with a disease vs. a patient without

Question 12

What is the equation for sensitivity?

Answer 12

TP/ TP + FN

Question 13

What is the equation for specificity?

Answer 13

TN/ TN + FP

Question 14

Match the terms: sensitivity/ specificity and rule-in vs. rule out.

Answer 14

Sensitivity= Rule OUT
- SnNout

Specificity= Rule IN
- SpPin

Question 15

What is pre-test probability?

Answer 15

An estimate of prevalence

Question 16

What is post-test probability?

Answer 16

Probability of disease after diagnostic test

*****Note that a diagnostic test should increase this number

Question 17

What does a likelihood ratio of 1 mean?

Answer 17

Pre and post-test probability are the same

*****This test is USELESS–it has not told you anything

Question 18

What is a likelihood ratio greater than one mean?

Answer 18

Probability of finding the disease by test is increased

Question 19

What is a likelihood ratio less than one mean?

Answer 19

Probability of finding the disease by test is decreased i.e. the test does NOT help you

Question 20

Generally, what does Bayes Theorem state?

Answer 20

Likelihood of a patient having a disease is equal to the prevalence of the disease in the population

Question 21

How can you determine if you should implement a test into practice?

Answer 21

Ask:

• Will the test be satisfactory in my setting?
• Are the results applicable?
• Will the results change my management?*
• Will patients be better off as a result of the test?