Description of performance of

diagnostic test and calculation

1
Description of performance of diagnostic and
calculation
• Gold standard
• Graphic and tabular representation of diagnostic test result
• Definition, notation and calculation of test performance characteristics
• Sensitivity
• True positive
• False positive

• Specificity
• True negative
• False negative

• False negative rate

• False positive rate
• Positive predicative value
• Negative predicative value
2
Content contd……
• Relation between test performance
characteristic
− Sensitivity and FNR
− Calculating sensitivity in terms of FNR
− Calculating of FNR in terms of sensitivity
− Specificity and FPR
− Calculating specificity in terms of FPR
− Calculating of FPR in terms of specificity
• Likelihood ratio and serial likelihood ratio
• Uses of sensitivity and specificity test
• Deference between sensitivity and specificity
Gold standard (test)
• In medicine and statistics, gold standard
test usually refers to a diagnostic test or benchmark
that is the best available under reasonable
conditions.

• Other times, gold standard is used to refer to the

most accurate test possible.

• In clinical research "Gold standard" can refer to the

criteria by which scientific evidence is evaluated.

4
• A gold standard medical test is not
necessarily perfect, but one that is recognized as
good enough so that all other tests can be
compared to it. 
• This type of evidence is used to evaluate new
methods of diagnosing disease

• as often the "gold standard" test may have the

potential for complications or is expensive or time
consuming.
5
• Other times, "gold standard" does not refer to the
best performing test available, but the best
available under reasonable conditions.
• For example, in this sense, a MRI is the gold
standard for brain tumor diagnosis, though it is not
as good as a biopsy.
• In this case the sensitivity and specificity of the
gold standard are not 100% and it is said to be an
"imperfect gold standard" or "alloyed gold
standard"
6
Examples
• the gold standard for diagnosing pulmonary
embolism (PE) is the in which dye is injected into the
bloodstream and X-rays are taken of the arteries in
the lungs.
•  Angiography is presumed to be nearly 100%
sensitive and specific, and then CT can be compared
to it, and its accuracy calculated.
• this is an invasive test, other methods of diagnosing
the disease are desirable.
• Computed tomography (CT) scanning is an
increasingly common method used to diagnose PE.
To know how accurate it is, it must be compared to a
test that is known to be good. 7
• Pulmonary embolism is
a blockage in one of the
pulmonary arteries in
your lungs. In most
cases, pulmonary
embolism is caused by
blood clots that travel
to the lungs from the
legs or, rarely, other
parts of the body (deep
vein thrombosis).
CT image of PE Angiography images
Graphical and tabular presentation
of diagnostic test result
• Graphical displays of results allow researchers to
summaries and communicate the key findings of their
study.

• Diagnostic information should be presented in an

easily interpretable way, which conveys both test
characteristics (diagnostic accuracy) and the potential
for use in clinical practice (predictive value).

• It can be presented in graph (line, Pie bar and

histogram), diagram and table
 Tabular presentation
Bed Time BP Pulse Tempr Vitals Be 6:00 10:0 2:00 6
No. d AM 0 PM PM
1 6:00 AM 140/10 no AM
0
2 10:00 AM 130/90 BP 1 160/ 130/ 140/ 150/
90 80 85 90
3 2:00 PM 130/90
4 6:00 PM 120/80 Pulse
5 10:00 PM 130/90 Tempr
6 2:00 AM 130/90 SpO2
7 6:00 AM 130/90
Objective of graphical and
tabular presentation
• To summarize the date
• To interpreted the finding
• To make easily understandable and
Criteria for Evaluating a Test

1. Acceptability
2. Repeatability
3. Validity
4. others- yields, simplicity, safety,
rapidity ease of administration and
cost.

14
Validity of a Diagnostic Test
• Test validity is the ability of a screening test
to accurately identify diseased and non-
disease individuals.

• An ideal screening test is

exquisitely sensitive  (high probability of
detecting disease) and
extremely specific(high probability that
those without the disease will screen
negative). 15
Validity of a Diagnostic Test

• Sensitivity and specificity are statistical measures

of the performance of a test also known in statistics
as classification or function

Test gives two answers

1. Correct answers- true positive and true negative,
and
2. wrong answers- false positive and false negative.

16
 Sensitivity
• The term sensitivity was introduced by
Yerushalmy in 1940s as a statistical index of
diagnostic accuracy.
• It has been defined as the ability of a test to
identify correctly all those who have the disease,
that is "true positive".
• A 90 per cent sensitivity means that 90 per cent
of the diseased people screened by the test will
give a "true positive" result and the remaining 10
per cent a "false negative" result.

17
 Sensitivity Contd..
• Sensitivity is the
ability of the test
to identify
correctly those
who have the
disease (a) from all
individuals with
the disease (a+c)
18
• Sensitivity is the test detecting true
cases of disease? or
• sensitivity is the proportion of truly ill
people in the screened population who
are identified as ill by the screening test.
• Ideal is 100%: 100% of cases are
detected.

19
Sensitivity = Probability of a positive test in a
people with the disease
= a / (a + c ) all persons with
the disease
= True Positive/ disease+

Sensitivity is a fixed characteristic of the test

(not fluctuated in incidence and prevalence of the
disease)
20
Specificity
• It is defined as the ability of a test to
identify correctly those who do not have
the disease, that is, "true negatives".
• A 90 per cent specificity means that 90
per cent of the non-diseased persons will
give "true negative" result,
• 10 per cent of non-diseased people
screened by the test will be wrongly
classified as "diseased" when they are
not. 21
 Specificity
• Specificity is the ability
of the test to identify
correctly those who do
not have the disease (d)
from all individuals free
from the disease (b+d).
• Specificity is the
proportion of truly
healthy people who are
so proved by the
diagnostic test
22
Specificity: Is the test excluding those without
disease? or specificity is the proportion of truly
healthy people who are so identifies by the
screening test.

Ideal is 100%: 100% of non-cases are negative

23
• Specificity= probability
of a negative test in a
people without the
disease
= d / (b + d) all persons
without disease
= True Negetive/ Disease -
= (T-/D-)
• Specificity is also a fixed
characteristic of the test
24
True positive and false positive
• True positive (a): the patient has the
disease and the test is positive. Sick
people correctly identified as sick

• False positive (b): the patient does not

have the disease but the test is positive.
Healthy people incorrectly identified as
sick.
25
True negative and false negative
• True negative (d): People does not have
the disease and the test is negative.
Healthy people correctly identified as
healthy.
• False negative (c): the patient has the
disease but the test is negative. Sick
people incorrectly identified as healthy

26
Sensitivity or true positive and specify
and true negative
Screening Diagnosis Total
test result Diseased Not diseased

Positive a (true b(false a+b

positive) positive)
Negative c(false d(true c+d
negative) negative)
Total a+c b+d a+b+c+d
• Sensitivity (TPR) = (a/a+c)x100%
• Specificity (TNR) = (d/b+d)x100%
27
Content contd……
• Relation between test performance characteristic
• Sensitivity and FNR
• Calculating sensitivity in terms of FNR
• Calculating of FNR in terms of sensitivity
• Specificity and FPR
• Calculating specificity in terms of FPR
• Calculating of FPR in terms of specificity
• Likelihood ratio and serial likelihood ratio
• Uses of sensitivity and specificity test
• Deference between sensitivity and specificity
Validity of a Diagnostic Test

Disease status
Present Absent Total
Positive a b a+b
Negative c d c+d
Total a+c b+d a+b+c+d
a = no. of true positives, b=
no. of false positives, c=no. of
false negative, d=no. of true
negatives

29
Sensitivity or true positive rate

Screening Diagnosis Total

test result Diseased Not diseased

Positive a (true b(false a+b- 60

positive)- 40 positive)-20
Negative c(false d(true c+d
negative) negative)- 9940
100 9840

30
Sensitivity or true positive rate

Screening Diagnosis Total

test result Diseased Not diseased

Positive a (true b(false a+b- 60

positive)- 40 positive)-20
Negative c(false d(true c+d
negative) negative)- 9940
100 9840
Total a+c- 140 b+d- 9860 a+b+c+d
(1000)

31
Sensitivity or true positive rate
Screening Diagnosis Total
test result Diseased Not diseased

Positive a (true b(false a+b- 60

positive)- 40 positive)-20
Negative c(false d(true c+d
negative) negative)-
100 9840
Total a+c- 140 b+d- 9860 a+b+c+d
• Sensitivity or True Positive Rate = a/a+c x 100=
(40/140)x100= 28.6%
• Speficity or TNR= d/b+d = (9840/9860)x100 =
99.97 % 32
FN or FNR

• False negative: the term false negative

means that patient who actually have
the disease are told that they do not
have the disease (test inaccurately gives
negative result).
• It means to giving them a false
reassurance.

33
FNR Contd…

How to find FNR? FNR

belongs to which
population ?
• FNR belong to diseased
population
• False negative rate =
c/a+c,
• FNR = c/a+c,
100/140x100= 71.4%
• The patients with a false negative test result
might ignore the development of signs and
symptoms and may postpone the treatment.
• This could be detrimental if the disease in
question is a serious one and the screening
test is unlikely to be repeated within a short
period of time.
• A screening test which is very sensitive has
few false negatives.
35
Relationship of
sensitivity and FNR?

36
Relationship of FNR and Sensitive

37
To calculate FNR in terms of sensitivity

FNR
• False negative rate = c/a+c, 100/140x100=
71.4%
• FNR in terms of sensitivity
• TPR+FNR= 100% (28.6+71.4=100)
• Sn+FNR= 100 (Sn=TPR)
• FNR=100-Sn
• FNR= 100-28.6=71.4%
Specificity or True Negative Rate
• Specificity (TNR) Screening Diagnosis Total
test result Diseased Not
= d/b+d x100 diseased
Positive a (true b(false a+b-
positive)- positive)- 60
• TNR= d/b+d x 40 20
Negative c(false d(true c+d
100 negative) negative)-
• =9840/9860x100 100 9840
Total a+c- 140 b+d- 9860 a+b+c
=99.9 +d

39
Relationship of FP or FPR and
Specificity
• False positive: the term false positive means
that people who do not have the disease are
told that they have, or test gives positive
result falsely.
• In this case, normal healthy people may be
subjected to further diagnostic test, at some
inconvenience, discomfort, anxiety and
expense until their freedom from disease is
established.
• A screening test with a high specificity will
have few false positive.
40
• False positive not only burden the
diagnostic facilities, but also bring
discredit to screening programs.
• In fact, no screening test is perfect,
i.e., hundred percent sensitive and
hundred percent specific.

41
FPR Contd…

How to find FPR? FPR

belongs to which
population ?
• FPR belong to non
diseased population
• False positive rate =
b/b+d,
• FPR =20/9860x100=
0.2%
Specificity in terms of FPR
• 

43
To calculate Specificity in terms of false
positive rate

• FPR= b/b+d x100= 20/9860 x100= 0.20%

FPR in terms of specificity

• TNR+FPR=100%
• FPR=100-TNR (Sp)
• FPR=100-99.8= 0.20

44
 Sensitivity or True Positive Rate (TPR) = % of people with the disease who
are detected by the test
TP
TPR=
TP+FN
= a/ a + c x 100
• False positive Rate ( FPR)= % of people without the disease who were
incorrectly labeled by the test as diseased
FP
• FPR=
FP +TN
= b / b + d X100
45
• Specificity or True Negative Rate (TNR) = % of people without the disease who
were correctly labeled by the test as not diseased
TN
TNR=
TN + FP
= d/ b + d x 100
• False negatives rate (FNR) = % of people with the disease who are not detected
by the test
FN
FNR=
FN +TP
= c / a + c x 100

46
• Sensitivity and FNR

• False negative rate (FNR) FN

=--------------- X 100
FN+TP
• FNR= 100 - FPR

• Specificity and FPR

FP
• False positive rate (FPR) =-------X 100
TN+FP FPR= 100 – TNR

47
DENOMINATORS OF THESE RATES

• The denominators of sensitivity and the false

negative rate is PEOPLE WITH DISEASE

• The denominators of specificity and the false

positive rate is PEOPLE WITHOUT DISEASE

48
• sensitivity and specificity
depends upon the
• Incidence of disease
• Prevalence of disease
• Both
• None

49
 Predictive value
• There are two kinds of tests used for assessing
people’s health: diagnostic tests and screening tests.
• screening tests typically have advantages over
diagnostic tests such as placing fewer demands on
the healthcare system and being more accessible as
well as less invasive, less dangerous, less expensive,
less time-consuming, and less physically and
psychologically discomforting for clients.
• Screening tests are also, however, well-known for
being imperfect and they are sometimes ambiguous.

50
 Predictive value
• It is, therefore, important to determine the extent to which
these tests are able to identify the likely presence or absence of
a condition.
• In addition to sensitivity and specificity, the performance of a
screening test is measured by its “predictive value” which
reflects the diagnostic power of the test.
• Sensitivity and specificity measure the accuracy of the test (not
any relation to the disease or population), whereas Positive
Predictive Value and Negative Predictive Value measure the
proportion of people whose test results reflect their health
status and therefore are affected by the disease prevalence.
51
 Two types
• Positive predictive value refers to the probability of the person
having the disease when the test is positive.
• Negative predictive value refers to the probability of the person
not having the disease when the test is negative.
• The positive and negative predictive
values (PPV and NPV respectively) are the proportions of
positive and negative results in statistics and diagnostic
tests that are true positive and true negative results,
respectively.
• The PPV and NPV describe the performance of a diagnostic test.
• The predictive accuracy depends upon sensitivity,
specificity and prevalence.
• The predictive value of positive test indicates the
probability that the patient with a positive test
result has, in fact, the disease is question.
• The more prevalence of disease is in a given
population, the more accurate will be the
predictive value of a positive screening test.
• The predictive value of positive result falls as
disease prevalence declines. 53
• When evaluating the feasibility
or the success of a screening
program, one should also
consider the positive and
negative predictive values.

54
 PREDICTIVE VALUE OF TEST
• The predictive value of a diagnostic test is
the most important aspects of a test.
• The ability of a test to predict the presence
or absence of a disease is a determinants
of test's value.
• Predictive values are of two types
−Positive predictive value (PPV)
−Negative predictive value (NPV)

55
Positive predictive value:
•  is the probability that subjects with a
positive screening test truly have the
disease.
• Positive predictive value ( Predictive
value of positive test ) is the proportion
of all people with positive tests who have
the disease.
56
• Negative predictive value ( Predictive
value of negative test ) is the
proportion of all people with negative
tests who do not have the disease.

57
Example

58
 Example contd….
• One way to avoid confusing this with sensitivity and
specificity
• to imagine that you are a patient and you have just
received the results of your screening test (or
imagine you are the physician telling a patient
about their screening test results).
• If the test was positive, the patient will want to
know the probability that they really have the
disease, i.e., how worried should they be?
• If a test subject has an abnormal screening test
(i.e., it's positive), what is the probability that the
subject really has the disease? 59
PREDICTIVE VALUES DEFINED
• POSITIVE PREDICTIVE VALUE =
people with disease ( true positives ) x 100
people with a positive test ( True positive + false
positive )
• Interpretation: Among those who had a positive
screening test, the probability of disease was …….%.
• If the value is 80% the interpretation is “Among those
who had a positive screening test, the probability of
disease was 80%.”

60
Here, the positive predictive value =TP/Test +ve
=(132/1,115)x100% = 0.118, or 11.8%.
Interpretation: Among those who had a positive
screening test, the probability of disease was 11.8%.
61
 Example contd….
Positive Predictive Value
• In the example we have been using there were
1,115 subjects whose screening test was positive,
but only 132 of these actually had the disease,
according to the gold standard diagnosis.
• Therefore, if a subject's screening test was
positive, the probability of disease was
132/1,115 = 11.8%.

62
Example of NPV
Negative Predictive Value
• Negative predictive value: If a test
subject has a negative screening test,
what is the probability that the
subject really does not have the
disease?

63
• NEGATIVE PREDICTIVE VALUE =
people without disease ( true negatives ) x 100
people with a negative test ( true negatives and false
negatives )
• Among those who had a negative screening test,
the probability of being disease-free was ……%.
• If the –ve predictive value=90 than interpretation is
“Among those who had a negative screening test,
the probability of being disease-free was 90 %.”

64
Here, the negative predictive values is
(63,650/63,950)*100=99.9%.  

Interpretation: Among those who had a negative screening

test, the probability of being disease-free was 99.9%.
65
• In the same example, there were 63,695
subjects whose screening test was negative,
and 63,650 of these were, in fact, free of
disease.
• Consequently, the negative predictive value
of the test was 63,650/63,695)100% = 99.9%.
• Predictive value depends upon the prevalence of
the disease.
• PPV is directly proportional to the prevalence of the
disease or condition.
• In the above example, if the group of people tested
had included a higher proportion of people with
bowel cancer, then the PPV would probably come
out higher and the NPV lower.
• If everybody in the group had bowel cancer, the
PPV would be 100% and the NPV 0%.
67
 POINTS TO NOTE
• Note that the numerators and denominators are reversed
compared to sensitivity and specificity. In predictive
values, the denominator is the test result, and the
numerator is disease or non-disease.
• In general, the positive predictive value is the one most
used.
• Positive predictive value and sensitivity are perhaps the
two most important parameters in understanding the
usefulness of a test under field conditions.

68
• Sensitivity and specificity do not vary according to
the prevalence of the disease in the population.

• Predictive value of a test, however is HIGHLY

DEPENDENT on the prevalence of the disease in the
population

69
Validity of a Diagnostic Test

Disease status
Present Absent Total
Positive a b a+b
Negative c d c+d
Total a+c b+d a+b+c+d
a = no. of true positives, b=
no. of false positives, c=no. of
false negative, d=no. of true
negatives

70
Exercise 1

A test is used in 50 people with disease and 50

people without. These are the results:
Disease
+ -
+ 48 3 51
Test 2 47 49
-
50 50 100

71
Sensitivity =
Specificity =
FNR= Disease
FPR= + -
Positive predictive value =
+ 48 3 51
Negative predictive value = Test
- 2 47 49
50 50 100

72