Screening Tests

Screening tests Natural history of disease prepathogenesis

pathogenesis X

 Stage of prepathogenesis when there

are no symptoms of the disease

 Stage of pathogenesis, when the

disease manifests itself by certain symptoms and signs which bring the patient to the physician

There are two types of tests: Screening tests & diagnostic tests Purpose of screening tests:

 To detect individuals with risk factors that

predispose to disease development, thus preventive interventions could be done (primary prevention)  To identify persons with early or asymptomatic ( latent disease) that can be effectively treated leading to better outcome (secondary prevention)

Examples  Overt carcinoma of the cervix is preceded by a stage when

cancer cells are localized called carcinoma in situ, if detected prevent the development of the disease

 Senile keratosis of the skin ( benign lesion) has a great

probability of transformation to malignancy, so its detection and removal is a preventive measure



Thus detection in the prepathogenic state or early stages of the disease requires the examination of symptomless or apparently healthy individuals which means

Purpose of diagnostic tests  It is to move the estimated probability of disease toward either

end of the probability scale. When the estimated probability of the disease is close to Zero, the disease can be ruled out, when the estimated likelihood of a disease is close to 100, the disease is confirmed

disease probability scale Zero probability

100% probability

(The test is used to measure the probability of an outcome(

Screening vs. diagnostic tests - Screening test is done in the prepathogenic state - Persons who are positive are more likely to have the disease, and who are negative are more likely not to have the disease - Applied to a pop. with low probability of the risk factor or disease prevalence relative to those with symptoms

- Diagnostic test is done in the pathogenic state -Persons may be positive or negative for the disease - Applied to a pop. with high probability of risk factor or disease prevalence relative to those with no symptoms

Criteria of the successful screening program The disease:  The disease should be common enough (high

prevalence) to warrant a search for its risk factors or latent stages, because screening for rare disease leads to unacceptable cost- benefit ratios

 The morbidity or mortality of the untreated condition

must be substantial (potential of screening)

 An effective preventive intervention or therapy should

exist and should have a more beneficial outcome when applied to the prepathogenic than the pathogenic state

Criteria for a successful screening program:

The disease: - Early detection and intervention must improve disease outcome - The natural history of the disease should be

understood, such that the detectable prepathogenic stage is known and identifiable - Facilities for diagnosis and appropriate treatments should be available for individuals who screen positive

The screening test  It must be suitable to the population and suitable for

routine application  Maintenance of test accuracy over time and freedom

from screening related adverse effects  A combination of the objective and subjective criteria

must be present

Criteria of screening tests 

objective criteria

 Operating characteristics

( sensitivity + specificity = validity)  Predictive value ( positive &

negative)  Cost effectiveness

 Subjective criteria  Individual and public

acceptability ( rapid, simple, painless, with no complications)  Financing ( inexpensive

)

Screening test will divide examined apparently healthy population

 Healthy

 With abnormality

Being rapid and simple, screening test is subjected to many errors, so a confirmatory test must be done to evaluate its results. This will be applied for both healthy and people with abnormality as revealed by the screening test: confirmatory test

Healthy  The test done to this group

will help to assure that none of them has the disease  By this test some of the healthy will be fined to have the disease (false negative), and others will be negative for it ( true negative)

With abnormality  The test done to this group

will help to assure that all of them really have the disease and none of them is negative for it  So, by this test some of the population proved before to have the disease will be negative for it (false positive) and others will be proved to be positive for it ( true positive)

Screening test +ve

Confirmatory test +ve -ve a b

-ve

c

d

Total

a+c

b+d

total a+b c+d a+b+c+d

 a) Individuals positive by the screening test are also positive by the confirmatory test (true positives)  b) Individuals positive by the screening test , but negative by the confirmatory test ( false positives)  c) Individuals negative by the screening test , but positive by the confirmatory test (false negatives)  d) Individuals negative by the screening test and negative by the confirmatory test ( true negatives)

The overall validity =

a+d a+b+c+d

zero validity

% 100% validity

X

 A 100% valid test is one that

 A Zero % valid screening

has no false positives (b) or false negatives (c) results  Which means that its results agree 100% with confirmatory test  So, this screening test could be considered a diagnostic test (screening test= diagnostic test)

test doesn’t give any true positives (a) or true negatives (d)  Which means that its results agree zero% with the confirmatory test ( this test must be rejected)

Sensitivity  Definition: It is the ability of the test to give positive results in

diseased population .It is the percentage of persons with the disease of interest who have +ve test results= a/a+c x100,(by relating the true positives by the screening test to all positives by the confirmatory test  The greater the sensitivity of the test, the more likely that the

test will detect persons with the disease of interest  Tests of high sensitivity are important to rule out the disease,

that is a –ve result would virtually exclude the possibility that the patient has the disease of interest

specificity  Definition : It is the ability of the test to give negative results in free

population. It is the percentage of persons without the disease of interest who have –ve test results = d/ d+b x 100, ( by relating the true negatives by the screening test to all negatives by the confirmatory tests  The greater the specificity, the more likely that persons without the

disease of interest will be excluded by the test. Very specific tests are used to confirm the presence of the disease  If the test is highly specific, a positive test result would strongly

implicate the disease of interest

The effect of false positives: - False positives tend to swamp true positives in population, because most diseases we test for are rare

 The effect of false negatives: - Delayed intervention - disregard of early signs and symptoms which may lead to delayed diagnosis

Risks of screening:  Labeling effect: individuals truly diagnosed to have the disease will be positive for the disease to the end.  Some of the individuals who were finally truly diagnosed

as not having the disease may be diagnosed as having the disease by the screening test (false positives), these may path through psychic trauma, spent unexpected money to reach the final diagnosis of being healthy

Relation between sensitivity and specificity: sensitivity and specificity are inversely related +ve -ve

total

+ve -ve

+ve

14

8

22

+ve

-ve

1

91

92

-ve

total

15

99

114

total

10 (a) 5 (c ) 15

12 (b) 87 (d) 99

total 22 92 114

Sensitivity=a/a+cx100=14/15x 100=93%

Sensitivity=a/a+cx100=10/15x 100=66.6%

Specificity= d/d+bx100= 91/99x100 = 92%

Specificity= d/d+bx100=87/99x100=87.8%

Predictive values  Whereas the operating criteria of a test are of major

help in selecting a screening test, the predictive value of a test is a major aid in interpretation of results.  The operating criteria of a test are descriptors of

accuracy of a test, the predictive values are estimations of disease probability

+ve predictive value:  It is the probability of the presence of the

disease given the test result is positive. It is the percentage of persons with +ve test results who actually have the disease of interest =(a/a+b)x100

-ve predictive value:  It is the probability of the absence of the

disease given the test result is negative. It is the percentage of persons with –ve test results who actually have not the disease of interest= (d/d+c)x100

Relation between predictive values and disease prevalence: screening in women with no breast mass

+ve -ve

screening in women with breast mass

total

+ve

-ve

total

+ve

14

8

22

+ve

113

15

128

-ve

1

91

92

-ve

8

181

189

total

15

99

114

total

121

196

317

Prevalence=15/114x100= 13% Sensitivity =14/15x100=93% Specificity=91/99x100=92% +pv=14/22x100=64% -pv=91/92x100=99%

Prevalence=121/317x100=38% Sensitivity=113/121x100=93% Specificity= 181/196x100=92% +pv= 113/128x100=88% -pv=181/189x100=96%

Sources of bias in the evaluation of a screening program:  Lead time bias  Length bias  Volunteer bias

Lead time bias:  Lead time: it is the interval between the diagnosis of

a disease at screening and the usual time of diagnosis by symptoms lead time I diagnosis by screening

I diagnosis by symptoms

Lead time bias:  Consider a condition where the natural history allows

for an earlier diagnosis, however survival does not improve despite identifying it earlier - The screening program here will  over-represent earlier diagnosed cases  survival will appear to increase (but in reality it is increased by exactly the amount of time their diagnosis was advanced by the screening program  Thus there is no benefit of screening from the survival standpoint

Lead time bias:  Assume survival is the time between screen and

death  Does not into account the lead time between diagnosis at screening and usual diagnosis survival=14 years

I diagnosis by screening in 1994

I death in 2008

Lead time bias survival=14 years

I

I true survival=10 years

lead time 4 years

I

I

I

diagnosis

usual time of

death

by screening in 1994

diagnosis via symptoms in 1998

in 2008

Length bias:  Most chronic diseases, especially cancers, do not

progress at the same rate in everyone  Any group of diseased people will include some in

whom the disease developed slowly and some in whom the disease developed rapidly  Screening will preferentially pick up slowly developing

disease (longer opportunity to be screened) which usually has a better prognosis

O

P

biological onset of disease

Y

disease detectable via screening

D

symptoms begin

death

Screening O

P O

O

P Y

O

Y P

D Y

D

P

Y

D

O O

D

P

P Y

Y

D D

Time

Volunteer bias:  This occurs when those accept to participate are

different from the others  Volunteers tend to have:

- better health - lower mortality - likely to adhere to prescribed medical regimens

Screening in breast cancer  There is an evidence of effectiveness that breast physical

 

 

ex.& breast mamography done annually to women above 50years will reduce mortality of breast cancer Risks : Each rad of radiation delivered to the breast increases woman’s chance to develop breast cancer by 6/1000.000/year. This increases for developing breast cancer by less than 1% each time False positive results occur in 1.5% of examination, which leads to biopsies that are unnecessary. cost effectiveness: is measured by comparing the reduction in mortality achieved with the financial costs of screening

*2/3 of the effectiveness of breast cancer screening program can be achieved by screening with physical examination only * Adding mammography increases the effectiveness more than 50% but increases the cost *The cost of increasing life expectancy by reducing mortality from cancer= 5000$ for physical ex. * Adding annual mamography gains additional year of life expectancy at a cost of 25000$

Conclusion 

Sensitivity will determine the number of diseased people that will be missed (false -ve) and this increases when sensitivity decreases.



Specificity will determine the number of well people wrongly included as diseased ( false positive) and this increases when specificity decreases.



A test with false positives gives false alarm and psychological trauma



A test with false negatives gives a false sense of security and the disease may progress to non-curable stage



High disease prevalence will increase the + pv and decrease the - pv