SCREENING

Dr.Syed Aftab Rahim

[email protected]
 "iceberg
phenomenon"

<1% Classical Clinical

Disease
 "iceberg
phenomenon"

10 -30% Less Severe

Disease
 "iceberg
phenomenon"

70-90% Sub Clinical Cases

Or Asymptomatic Infection
Carriers
Natural History of Disease
Different types of Medical Tests
 Diagnostic
 specifically looking for a suspected
condition which is tested for and confirmed
or excluded
 Screening
 usually no specific exposure or indication
that the individual has disease.
Screening
 “...the identification of unrecognized
disease or defect by the application of
tests, examinations or other procedures...”
 “...sort out apparently well persons who
probably have disease from those who
probably do not.”
 “...not intended to be diagnostic...”
Screening
Screening is the process
to detect among
healthy people
disorders or risk factors
of which they are
 Wilson’s criteria for screening

• The condition should be an important health problem

• The natural history of the condition should be understood
• There should be a recognisable latent or early symptomatic stage
• There should be a test that is easy to perform and interpret, acceptable
accurate, reliable, sensitive and specific
• There should be an accepted treatment recognised for the disease
• Treatment should be more effective if started early
• There should be a policy on who should be treated
• Diagnosis and treatment should be cost-effective
• Case-finding should be a continuous process
Screening, examples
 Physicians assess blood pressure and
cholesterol as screening tools for the
development of cardiovascular disease
 Women use home pregnancy tests to screen
for presence of an embryo or fetus
 Some Common Screening
Screening,
Tests examples

 Pap smear for cervical dysplasia or cervical cancer

 Fasting blood cholesterol for heart disease
 Fasting blood sugar for diabetes
 Blood pressure for hypertension
 Mammography for breast cancer
 PSA test for prostate cancer
 Fecal occult blood for colon cancer
 Ocular pressure for glaucoma
 PKU test for phenolketonuria in newborns
 TSH for hypothyroid and hyperthyroid

1
Types of Screening
1. Mass
2. Targeted
3. Multiphasic
Types of screening
1. Mass screening, no selection of
population (e.g., checking all infants for
hearing problems)
2. Targeted / Selective screening (e.g., by
age and sex: mammograms for women
aged over 40)
3. Multiphasic screening (a series of tests,
as family doctors do at annual health
Multiple-stage screening

Two-stage screening test

1. Screen individuals using test with high
sensitivity
2. Follow-up with test with high specificity
Screening program Screening test
 Survey Questionnaire
comprehensive disease  Physical observation
control activity based  Measurement
on the identification and  laboratory test
treatment of persons  radiological procedure
with either used to help identify persons
unrecognized disease or with unrecognized disease or
unrecognized risk unrecognized risk factors for
disease.
factors for disease.
Characteristics of a good
screening test
 Valid (e.g., sensitive and specific)
 Reliable (gives consistent results; no random errors)
 Yield (number of cases identified per thousand screened)
 Cost – benefit (compare costs avoided due to early
detection of the disease against cost of the screening.
Does the test merely uncover more disease that is
expensive to treat without appreciable advantage?)
 Acceptable (discomfort, hassle, cost of obtaining test)
 Follow-up services (plan needed to deal with positive
results)
 Validity
 Accuracy of a test
 How good is the test at discriminating between who has disease and

who doesn’t have disease

 Two components:
 Sensitivity: ability of a test to identify correctly those who have the
disease
 Specificity: ability of a test to identify correctly those who do not have
the disease
Sensitivity and Specificity

Population
Disease No Disease
Test + True Positive False Positive
(TP) (FP)
Test - False Negative True Negative
(FN) (TN)
Sensitivity = TP__ Specificity = TN__
TP + FN TN +
FPcalculate Sensitivity and Specificity we need to know
To
the true state of disease
 Calculating Sensitivity and
Specificity
 Comparing your test or measure to a “gold standard”

 Gold Standard is usually another test

 More definitive
 More invasive
 More expensive
 Takes longer

 To limit the number of people exposed to these drawbacks, we develop other “tests”
Calculating Sensitivity
Disease No Disease
Test + True Positive (TP) False Positive (FP)

Test - False Negative (FN) True Negative (TN)

Test Results Disease No Total

Disease
Positive 80 (TP) 100 (FP) 180

Negative 20 (FN) 800 (TN) 820

Total 100 900 1,000

Sensitivity = TP/Have disease = %
Sensitivity = 80 (TP)/100 (have disease) = 80%
 Calculating Specificity
Disease No Disease
Test + True Positive (TP) False Positive (FP)

Test - False Negative (FN) True Negative (TN)

Test Results Disease No Total

Disease
Positive 80 (TP) 100 (FP) 180

Negative 20 (FN) 800 (TN) 820

Total 100 900 1,000

Specificity = TN/No disease = %
Specificity = 800 (TN)/900 (no disease) = 89%
 False Positives vs. False Negatives

 False Positives (FP): Burden on patient and health care system

 Expensive, invasive tests and procedures
 Anxiety and worry
 False Negatives (FN): Depends on the nature and severity of the disease

 Patient access to care

 Likelihood of another test
 Availability of effective interventions/therapies
 Timeliness of intervention
 Cancer

 In general: high sensitivity for a severe or frequent disease, high specificity for a rare disease (limits
FPs)
 Screening Test
POSITIVE PREDICTIVE VALUE
S
Actually Have Condition? S
E
N P
SENSITIVITY E
S Totals
I Screening Yes No C
T Test Results: I
I F
a b I
V Positive TP (True FP (False a+b
I C
Positives) Positives)
T c d I
Y Negative FN (False TN (True c+d T
Negatives) Negatives) Y
Totals a+c b+d a+b+c+d
NEGATIVE PREDICTIVE VALUE
Errors
HYPOTHESIS
TESTING
DEFINITION OF
HYPOTHESIS
Is a statement which shows a probable relation
between the variables.
It is of two types:
 Alternate

 Null
 Suppose a study is being conducted to
answer questions about differences
between two regimens for the management
of diarrhea in children:

- the sugar based modern ORS and the time-

tested indigenous herbal solution made
from locally available herbs.
 One question that could be asked is:

 "In the population is there a

difference in overall improvement
(frequency and consistency of the
stools) after three days of treatment
between the ORS and the herbal
 There could be only two answers to this
question:

 No or Yes
1. "There is no difference between such
improvement” (null hypothesis).

A null hypothesis usually states that there is no

difference between groups or that one factor is not
dependent on another and corresponds to the No
answer.
Null Says:
 In populations there is no association (or there is no
difference) between A and B; The observed results from
the investigator's sample are therefore due to

sampling variation (we got the sample was just by

chance alone).
 Associated with the null hypothesis there is always
another hypothesis or implied statement
concerning the true relationship among the
variables or conditions under study if ‘no’ is not a
possible answer.
 This statement is called the
alternate hypothesis and
corresponds to the Yes answer.
 2. "There is a difference
between the improvement
achieved by a three days
treatment with the ORS than
that of the herbal solution"
(alternative hypothesis).
 ALTERNATE HYPOTHESIS

 It does not say anything with reference to

sampling variation.

 It just says what the investigator wishes it to

say i.e. A<B, A>B, or A not related to B.
 The issue now is that if you
could prove beyond doubt that
null is false, then the alternate
must be true (because there are
only two answers to the
question).
ALPHA AND BETA ERRORS

True Disease Status

Clinical Yes No
Diagnosis

Yes Correct Decision Wrong Decision

 (α) Alpha Error

No Wrong Decision Correct Decision

(β) Beta Error 
Type I Error (False Positive Error)

 A type I error occurs when the

null hypothesis is true, but is rejected.
 Shepherd and wolf example. Let’s say
that our null hypothesis is that there is “no
wolf present.” A type I error (or false
positive) would be “crying wolf” when
there is no wolf present.
Type II Error (False Negative)

 A type II error occurs when the null

hypothesis is false, but erroneously fails to
be rejected. Let me say this again, a type
II error occurs when the null hypothesis is
actually false, but was accepted
Example of Type II Error
 null hypothesis is that there is “no wolf
present.” A type II error (or false
negative) would be doing nothing (not
“crying wolf”) when there is actually a
wolf present.
PPV
 Positive predictive value tells you the
odds of you having a disease if you have
a positive result.
NPV
 The negative predictive value is the
probability that people who get a
negative test result truly do not have the
disease. In other words, it’s the
probability that a negative test result is
accurate.
 Predictive Value of a Test
 Predictive value changes with prevalence of
disease
 Predictive Value and Prevalence
 As prevalence increases, the Positive Predictive Value increases and the Negative
Predictive Value decreases
Predictive Value and Specificity

 Another value that affects the predictive value is the

specificity

 If specificity increases then predictive value will also

increase.
Reliability
 Reliability tells you how consistently a
method measures something. When you
apply the same method to the same
sample under the same conditions, you
should get the same results. If not, the
method of measurement may be
unreliable.