Screening in

Public Health Practice

 Disease detection procedures
• Screening purpose
– Applied on apparently healthy population to
identify disease before producing symptoms
• Diagnostic purpose
– Applied on a person with symptoms to confirm
the presence of disease

2
 Questions on validity of a test
• What proportion of the true cases are identified as cases by
this test?
– Sensitivity
• What proportion of the true non-cases are identified
as non-cases by this test?
– Specificity
• What proportion of the test positive cases are true
cases?
– Positive predictive value
• What proportion of the test negative cases are true
non-
cases?
– Negative predictive value 3
 Measures of validity
• Sensitivity: Probability that a test will be positive when the test
performed on the people who actually have the disease.
Proportion of individuals with the disease correctly classified by
the screening test as having the disease

= # with disease who test positive

# tested who truly have disease
• Specificity: Probability that a test will be negative when the test
performed on the people who actually do not have the disease.
Proportion of individuals without the disease correctly
classified by the screening test as being disease-free

= # without disease who test negative

# tested who do NOT have disease

4
 Indices of accuracy
• Sensitivity and Specificity describe test accuracy and
are independent of disease prevalence
• The probability that a case is positive on a test (the
detectability of the case) depends on the point in
the preclinical phase at which screening is done.
• A disease which is very early in preclinical phase is
not detectable by most tests.
– the detectability of the disease tends to increase as
the
preclinical phase progresses
– Thus, sensitivity of a test can also be considered a function
of disease progression. 5
 The 2x2 Table
Gold Standard
Disease Disease Total
Absent
Present
Test Positive a b a+b
Result ++ (TP) -+ (FP)
Negative c d c+d
+- (FN) -- (TN)
Total a+c b+d a+b+c+d
6
 Calculations

• Sensitivity
– a/(a+c)
– TP/(TP + FN)

• Specificity
– d/(b+d)
– TN/(TN + FP)

7
 Life is not always easy!
Tests of Continuous Variables
•We often test for a continuous variable, such as blood pressure for
which there is no “positive” or “negative” result
•establishing a cutoff level above which a test result is considered
positive and below which a result is considered negative

8
Tests of Continuous Variables

9
Tests of Continuous Variables

10
Tests of Continuous Variables

11
 100%
sensitivity

17%
specific
ity
 39%
sensitivity

100%
specific
ity
 Sensitivity vs. Specificity
High Sensitivity High Specificity
(Few false negatives) (Few false positives)
• Serious disease (don’t • Subsequent diagnostic
want to miss cases) test is high risk and high
• Potential for person- cost
to-
person-transmission
• Subsequent diagnostic
test is low risk and
low cost
• Psychological burden on
individual is high
It’s a trade off between sensitivity and specificity
15
Receiver Operating Characteristic (ROC) curves

16
 Receiver Operating Characteristic (ROC) curves

• Plots true positive rate (sensitivity) against false

positive rate (1-specificity) demonstrating the trade
off that exists between them.

• The area under the curve is a measure of the test

accuracy.
– Area =1 indicates perfect test

17
 Guidelines for interpreting
ROC curve
• .90-1 = excellent (A)
• .80-.90 = good (B)
• .70-.80 = fair (C)
• .60-.70 = poor (D)
• .50-.60 = fail (F)

18
 Positive predictive value
• Positive predictive value : % of individuals classified
by the screening test as having the disease who
actually have the disease
– % of people who do have the disease among those
classified as having the disease
– TP /(TP +FP)

19
 Negative predictive value
• Negative predictive value: % of individuals
classified by the screening test as disease-free who
do not have the disease

– % of people truly disease-free among those who are

classified as disease-free
– TN/(FN + TN)

20
 Predictive Value Positive and
Prevalence
Of the 1,000 subjects, 180 have a positive test result. Of these 180 subjects, 80 have the
disease. Of the people who have a negative test result, 800 do not have
the disease

21
Effects of Disease Prevalence on the
Predictive Value of a Screening Test
• NPV and PPV are determined by both the test
characteristics (i.e. sensitivity and specificity) and the
prevalence of the disease in the population that you are
testing the screening test

• When the prevalence of a disease increases, the positive

predictive value increases, and the negative predictive
value decreases.

• Predictive value helps in deciding where the test will be

applied

22
Relationship Prevalence vs. PPV

23
 When to decide what
The head of a firefighters’ union consulted a university cardiologist because the fire
department physician had read an article in a leading medical journal reporting that a
certain ectrocardiographic finding was highly predictive of serious, generally
unrecognized, coronary heart disease. On the basis of this article, the fire department
physician was disqualifying many young, able-bodied firefighters from active duty. The
cardiologist read the paper and found that the study had been carried out in
hospitalized patients. What was the problem?

• Hospitalized patients have a much higher prevalence of heart disease than does a
group of young firefighters.

• Fire department physician had erroneously taken the high predictive value
obtained in studying a high-prevalence population and inappropriately applied it to
a low-prevalence population of healthy
 Who is correct?
A physician visited his general internist for a regular annual medical examination,
which included a stool examination for occult blood. One of the three stool
specimens examined in the test was positive. The internist told his physician-patient
that the result was of no significance because he regularly encountered many false
positive test results in his busy practice. The test was repeated on three new stool
specimens, and all three of the new specimens were now negative. Nevertheless,
sensing his patient’s lingering concerns, the internist referred his physician-patient
to a gastroenterologist. The gastroenterologist said, “In my experience, the positive
stool finding is serious. Such a finding is almost always associated with pathologic
gastrointestinal disorders. The subsequent negative test results mean nothing,
because you could have a tumor that only bleeds intermittently.

•both the general internist and the gastroenterologist were correct

•internist gave his assessment of predictive value based on his experience in his general
medical practice—a population with a low prevalence of serious gastrointestinal
disease
•the gastroenterologist gave his assessment of the predictive value of the test based
on his experience in his referral practice—a practice in which most patients are referred
because of a likelihood of serious gastrointestinal illness—a high-prevalence
population
 Other Measures from the 2 x 2 Table

• Accuracy of a screening test:

(TP+TN)/(TP+FN+TN+FP)
# of correctly identified individuals (regardless of disease
status)
total # of individuals

• Prevalence of disease:
(TP+FN)/ (TP+FN+TN+FP)
# of individuals who are truly diseased
total # of individuals

26
 Screening programs

• The organized application of early diagnosis

and treatment activities in large groups is
mass screening or population screening

27
 Suitable screening program
• Acceptable to population
– Free of discomfort and risk
– Convenient and attractive to target population
• Low cost
• High PPV
• High NPV
• Appropriate follow-up for positive screenees must
be in place

28
 Successful screening program
• For any screening program to be successful at
reducing mortality and morbidity, a substantial
proportion of cases must be detected during the
preclinical phase with enough time for treatment
to be more effective than it would have been if
treatment was given at a later time point.

29
 WHO — Principles of Screening
Guidelines were published in 1968, but are still
applicable today.

1. The condition should be an important health

problem.
2. There should be a treatment for the
condition.
3. Facilities for diagnosis and treatment should be
available.
4. There should be a latent stage of the disease.
5. There should be a test or examination for the
condition.
30
 WHO — Principles of Screening
6. The test should be acceptable to the population.
7. The natural history of the disease should be
adequately understood.
8. There should be an agreed policy on who to
treat.
9. The total cost of finding a case should be
economically balanced in relation to medical
expenditure as a whole.
10. Case-finding should be a continuous process, not
just a "once and for all" project.

31
Enhancement of screening program
• Apply screening among high risk population
– Net effect: increasing prevalence increases
PPV
• Targeting high risk group of individuals can
greatly improve feasibility of screening program
if disease is very rare
– Detecting most of the cases that would be
detected by screening the entire population
– Overall cost of screening is reduced

32
 Bias in screening program
evaluation
• Selection bias
– Volunteer effect
• Healthy people more likely to volunteer for
tests
• Lead-time bias
• Length bias

33
 Lead time bias
An woman had biologic onset of breast
cancer in 2000. Because the disease
was subclinical at that time, she had no
symptoms. In 2008, she felt a lump in
her breast which precipitated a visit to
her physician, who made the diagnosis.
The patient then underwent a
mastectomy. In 2010, she died of
metastatic cancer.

In scenario B, there was a screening

programme and in 2005 she was
identified, She had surgery in 2005 but
died in 2010. Apparently there was 3
year extra survival, but in reality no
extension of length of life.
This apparent (but not reality) increase
of survival is lead time bias

34
 Length time bias

Because more aggressive diseases are asymptomatic for a shorter period,

screening is more likely to detect slower progressing diseases, such as
slow-growing tumours, which have a better prognosis, including longer
survival. Screening may thus falsely appear to improve survival.
35
 Reliability
• Can the results obtained be replicated if the test is repeated
• The factors that contribute to the variation between test
results are
– intrasubject variation (variation within individual subjects),
– intraobserver variation (variation in the reading of test results by
the same reader),
– interobserver variation (variation between those reading the test
results).
 Percent agreement and kappa statistics
• Percent agreement between two
observers is often of value in assessing
the quality of their observations
• The percent agreement between two
observers does not entirely depend on
the quality of their training and practice
• Ignore the agreement by chance alone

a kappa:
>0.75 represents excellent agreement
<0.40 represents poor agreement,
0.40 - 0.75 represents intermediate to good agreement.
Example Kappa
Example Kappa

Level of agreement?

excellent!
 Adverse effects of screening
1. Stress and anxiety caused by a false positive
screening result.
2. Unnecessary investigation and treatment of false
positive results.
3. Prolonging knowledge of an illness if nothing can
be done about it.
4. A false sense of security caused by false
negatives, which may even delay final diagnosis.
5. Overuse/waste of medical resources.
6. Unnecessary and uncomfortable procedures
looking for a disease that is unlikely.

40