Screening

Under supervision of Dr. Hebaallah Abd el maasoud

Prepared by dr. Remonda Nabil Haries
Community department
KFS University
 Topics
• Objectives
• Definition
• Examples
• Screening Vs diagnostic tests
• Criteria
• Validity assessment
• Cut off value
• How to choose screening test
 Definition of Screening

oTests done among apparently well people to identify

those at increased risk of a disease or disorder .
oPeople with positive screening test (+Ve) are subjected
for further diagnostic test , procedure , treatment or
preventive medication .
Iceberg phenomenon
of Disease
 is .ory o D11sease
Detec :nbl ubc inic:nl 1sease

I
Ii.a IDO
p -of re 111ired
Exposure

c ee •
 Objectives
• Early diagnosis
• Case detection (prescriptive screening ) ex: neonatal screening for
hypothyroidism
• Control of the diseases ( prospective screening ) ex: screening blood donor
for HBV,HCV and HIV .
• Research purposes
• Educational opportunities
 Examples
• In newborn : T3 measurement for iodine deficiency
• In women : - breast self examination for cancer breast
- cytology screening for cervical cancer
- VDRL in pregnant woman for venereal diseases
• In adults : blood pressure for hypertension
 Screening Vs. diagnostic tests
Screening tests Diagnostic tests
Done to those who are apparently Done to those with suggestive
healthy or asymptomatic signs or symptoms

Applied to a group of individuals Applied to a single person

Results are based on one Results are based on the evaluation

criterion of a number of symptoms, signs
and investigations
Results are not conclusive Results are conclusive and final
(positive – negative ) (diseased –non diseased )

Less accurate More accurate

Less expensive More expensive
Not a basis for treatment Basis for treatment 9
 Flow diagram for
a screening program
Population

Screening test

Test -ve Test +ve

Diagnostic events

Unaffected Affected

Re-screen Re-screen Intervention

10 Prof. El-Gilany
 Criteria for screening test
• There are 3 requirements for the screening test to do well more than harm

Disease Policy Test

 DISEASE
• 1-The disease should be :
- Medically important
- Clearly defined
- Prevalence reasonably well known
- Natural history known
- Effective intervention for prevention and control
 POLICY
2- The policy :
- Cost effective
- Facilities for diagnosis and treatment must be available
- The course after a positive result must be generally agreed upon and
acceptable to those screened
 Test
• The test should be :
- Simple and safe
- Acceptable
- Reasonable cut-off level defined
-Valid ( The ability of the test to differentiate between those with and those without
the disease )
- Reliable ( The ability of the test to give the same result with repetition )
- Good performance
 Validity
• Definition : The ability of the test to differentiate between those with and
those without the disease
• Assessment of screening test effectiveness ( validity ) : 4 indices

Sensitivity Negative
Positive predictive
Specificity predictive value
value
 Disease
Screening test Present Absent
Positive a b
TP FP
Disease present +test
• •
Non diseased+ positive
positive test
Negative C d
FN TN
Diseased with Non diseased with
negative test negative test
 Test cut-off
Moving cut-off to left Moving cut-off to right
redue es false negatives reduces false positives
(higher sensitivity) (higher spe cifi.c ity)
at the cost of reduced at the cost of reduced
specifi.city sensitivity

TN TP

Healthy Sick
 Validity assessment
1-Sensitivity it is ability of test to detect disease in those who are actually
diseased .
TP
Sn = ---------------- X 100
TP + FN
2-Specificity is the ability of test to exclude disease in those who actually
do not have it.
TN
Sp =--------------------------X100
TN + FP
 Validity assessment
• 3-- Positive predictive value : it is the proportion of correctly defined cases
among test positive cases .
• TP
PPV=------------------------X100
TP +FP
• 4- Negative predictive value : It the proportion of correctly excluded of
non diseased cases among test negative cases .
TN
NPV =------------------------X100
TN +FN
 D1ia�;nos:tIC test Total
vie -ve
,....J•

(]J, A B A. + B
.J; ' .

Of)
.....
C vie Tl'ue +ve FaJsie +ve
(]J,
(]J,
�
u,
- C D C +D
-ve False· --ve True-ve

:\·+ C
.J; ' B ·o A B +C+ D1
'

Total Total +ve Total-ve· Grand total

 Surgical biopsy
Mammography Total
Cancer No cancer
Positive 14 8 22
True False-
positive positive

Negative 1 91 92
False- True
negative negative

Total 15 99 114

Sensitivity = Specificity =
14/14+1 X100 91/91+8 X100
= 93 % = 92 %
 Surgical biopsy
Mammography Total
Cancer No cancer
Positive 14 8 22
True positive False-positive

PPV=14/22
=63.6%
Negative 1 91 92
False-negative True negative

NPV=91/92=
98.9%

Total 15 99 114
 Cut off level
So if ↓↓ cut off level → ↓↓ FN and ↑↑ FP → ↑↑ Sensitivity
And if ↑↑ cut off level → ↑↑FN and ↓↓ FP → ↑↑ specificity
‫العالقة بين‬sensitivity & cut off level ‫عكسية‬
‫العالقة بين‬specificity & cut off level ‫طردية‬
 Selecting suitable screening test :
• If the purpose of the screening test is to find diseases ( HIV –HCV ) cases
for treatment << select the test with …… sensitivity and ……. predictability
• If the purposes of the screening test is to employ healthy people ,select the
test with …… specificity and …... predictability
 Incidence and prevalence

Incidence Prevalence

Number of new cases of a particular disease Number of all cases (new +old) diagnosed in
over certain time and locality per at risk certain population in certain year and locality
population of certain time and locality per the population in the same year and
locality
Measure acute cases Measure chronic cases
 Rec rre ce

P evale ce

ortal-ty
 Predictive Value and Disease Prevalence
Mammo Surgical biopsy Mammo Surgical biopsy
graphy graphy
Cancer No cancer
Total Total
Cancer No cancer

Positive 113 15 128

Positive 14 8 22

Negative 8 181 189

Negative 1 91 92

Total 15 99 114 Total 121 196 317

Prevalence = 13 % Prevalence = 38 %
Sensitivity = 93% Sensitivity = 93%
Specificity = 92 % Specificity = 92 %
PV+ = 64% PV+ = 88% 12/27/2022 29
PV- = 99 % PV- = 96 %
 NB
• Sn test gives few false Negatives.
Sp test gives few false Positives.
Sn and Sp are the true indicators of validity because they are
independent of disease prevalence
• NB: In perfectly valid test FN and FP will be zero i.e. 100% agrees
with the reference test (gold standard). In fact no screening test is
perfect i.e. 100% Sn and 100% Sp .