1

 Critically Appraising
the Evidence:
 Statistics for Diagnosis

2

 Clinical Statistics Calculator (Excel)
 Statistics for:
 Diagnosis
 Sensitivity (Sn)
 Specificity (Sp)
 Likelihood Ratios (LR)
 Pretest Probabilities/Prevalence
 PostTest Probabilities/Predictive Values

3

 If available, find the best evidence in secondary sources where analysis
has already occurred.
 If not preassessed, use critical appraisal worksheets to help you
through the process.

4

 Understanding the Limitations of the Author’s Analyses and
Interpretations of the Data
 Assessing Internal Validity
 Assessing External Validity
 Identifying Potential Confounding Variables

5

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

6


7

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

8


9

 Answers the question: When
the target disorder was present, what fraction of the time did the
diagnostic test produce positive results?
 Excludes all trials where the diagnostic test produced negative results

10

 Step 1:
 How many times was the target disorder actually present?

11

 How many of these times did the diagnostic test produce positive
results?
 Note: We do not count the
number of times the diagnostic test produced positive results when the
target disorder was absent (b). We are only interested in the cases
where target disorder was present .

12

 Step 2:
 How many of these times did the diagnostic test produce positive
results?:

13

 Step 3:
 So what fraction of the total number of times the target disorder was
present did the diagnostic test produced positive results?:

14

 Answers the question: When
the target disorder was not present, what fraction of the time did the
diagnostic test produce negative results?
 Excludes all trials where the diagnostic test produced positive results

15

 How many times was the target disorder absent?: b+d
 How many of these times did the diagnostic test produce negative
results?: d
 Note: We do not count the
number of times the diagnostic test produced negative results when the
target disorder was present (c). We are only interested in the cases
where the target disorder was absent.
 So what fraction of the total number of times the target disorder was
absent did the diagnostic test produced negative results?: b/(b+d)

16

 Specificity – SpPin (spin)
 When a test has a high Specificity, a Positive test rules in the
diagnosis.
 Sensitivity – SnNout (snout)
 When a test has a high Sensitivity, a Negative test rules out the
diagnosis.

17

 Specificity
 1  α
 Probability of not getting
a Type I Error/α Error/False Positive
 Sensitivity
 1  Β
 Probability of not getting
a Type II Error/Β Error/False Negative
 Also know as the statistical power of the test

18

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

19

 Likelihood Ratio for a Positive Result (LR+)
 The number of patients testing positive for the target disorder who
have the disease for every patient testing positive that does not have
it
 Likelihood Ratio for a Negative Result (LR)
 The number of patients testing negative for the target disorder who
have the disease for every patient testing negative that does not have
it

20

 Likelihood Ratio for a Positive Result (LR+)
 sensitivity/(1specificity)
 Likelihood Ratio for a Negative Result (LR)
 (1sensitivity)/specificity

21

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

22

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

23

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

24

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

25

 Fraction of the population at risk at a given time who have the target
disease

26

 Probability that a patient has the target disease without knowing the
results of a diagnostic test

27

 (a+c)/(a+b+c+d)
 The number of those with the disease divided by the total number of
people

28

 Positive Predictive Value
 Probability that a patient with a positive diagnostic test actually has
the disease
 Negative Predictive Value
 Probability that a patient with a negative diagnostic test actually
does not have the disease

29

 Positive Predictive Values (+PV)
 Negative Predictive Values (PV)

30


31

 Critical Appraisal Practice Exercises

32

 EBM Glossary
 Critical Appraisal Practice Exercises

33

