Week 4 - Screening principles

Question 1

What did the commission on chronic illness discuss on preventive measures define screening as:-

Answer 1

•1951
• “Presumptive identification of unrecognised disease or defect using tests, examinations and other which can be applied rapidly”
• Screening tests sort out apparently well persons from who and who not have disease
• Screening tests are NOT diagnostic
- Suspicious test results are referred onto professionals

Question 2

Definition of epidemiology:

Answer 2

• “the scientific study of factors affecting health and illness of populations”
• Uses range of studies from observational to experimental to reveal relationships of:
- Nutrition, biological agents, chemicals on outcomes of disease and wellness

Question 3

What are the 10 screening principles?

Answer 3

1. Condition should be important health problem to community
2. Acceptable and effective form of treatment available
3. Natural history of disease should be adequately understood
4. Should be recognised latent or early symptomatic stage
5. Should be suitable screening test for detecting latent of early stage
6. Treatment at pre symptomatic/borderline stage should favour influence of its course
7. Facilities for screening should be available
8. Agreed policy on whom to refer as patients available
9. Cost of case-finding, treatment + diagnosis needs to be economically balanced
10. Case-finding should be continuous, not a “once for all” project

Question 4

What are the two sub-divisions of screening and their nature?

Answer 4

• Mass screening: large scale screening of populations, no selection made
- covid 19
• Selective screening: screening of selected populations who are at higher risk
- blue colour vision in divers
• Multiphasic screening: application of two or more screening tests on large group
- vision + hearing screening
• Monophasic screening: application of single test

Question 5

Why do we screen?

Answer 5

• Short test which is easy to administer
• Inexpensive
• Accurate
• Not diagnostic

Question 6

What are validity and reliability of screening tests dependent on + define the 2 terms:-

Answer 6

• Sensitivity- % of positives that are truly positive/ false positive
• Specificity- % of negatives that are truly negative/false negatives

Question 7

Sensitivity definition and equation:-

Answer 7

• Sensitivity = a/a+b
• Sensitivity correctly determines individuals with defect

Question 8

Specificity definition and equation:-

Answer 8

• Specificity = d/b+d
• Correctly identifies those that do NOT have the defect

Question 9

Define validity:-

Answer 9

• The proportional differences between sensitivity and specificity in the test, which will both be below 100%

Question 10

Define reliability:-

Answer 10

• Refers to repeatability
• Reliability usually expressed as test/retest correlation co-efficient
• 1- results or second test are perfectly predictable from first test, 0 - impossible to predict result of second test from result of first
• 0.8 coefficients are seen as satisfactory, with 0.7 may be unacceptable

Question 11

What are the reasons as to why tests can give different results:-

Answer 11

• Variations in the test or test conditions
- Inconsistencies in the test procedure or examiner
- Inherent differences in the individuals being tested

Question 12

What are the two types of errors:-

Answer 12

• Random errors increase the standard deviation of the measurement, without affecting the mean of the measurement.
• Systematic errors, such as a zero error in the calibration of a Goldmann tonometer, will change the mean, but will leave the standard deviation of the measurement unaffected.

Question 13

What determines the effectiveness of a screening test?

Answer 13

• Reliable and valid tests.
• The reliability is expressed in terms of the test-retest correlation co-efficient,
• Validity is expressed in terms of the sensitivity and specificity of the test.
• A good test will have this information available, so that it can be decided whether the test is suitable as a screening test, or not.

Question 14

Important notes for cut off points:-

Answer 14

• As cut off point increases, sensitivity increases but specificity decreases, and vice versa
• Cut off point is most efficient where there is a dip between both distributions
- Will give fewest total mis-classifications

Question 15

What is the single detection theory?

Answer 15

• Theory Differentiates between INFORMATION bearing patters, signals/stimuli from the random patterns/noise
- Does this using:- Signal recovery which separates from the background of the signal/stimulus
- Signal recovery is therefore dependent on threshold levels
- Example: wartime personal able to detect abnormalities in background noise
• Applications include radar, psychophysics and psychology

Question 16

What is the receiver operating characteristic? (ROC)

Answer 16

• Plot that show’s performance of a yes or no test with different thresholds
• Curve is created using true positive vs false positive at various thresholds
• Sensitivity in science, recall in machines
• False positive is called fall-out calculated as 1- specificity
• Graph is plotted as specificity on X axis, Sensitivity on Y axis, with a “no information line” in the middle

Question 17

What is the significant of the no information line on an ROC curve:-

Answer 17

• Greater area between no info line, the better the test is at discriminating between two populations
- Using a parallel line, the last point on the curve to touch the line is the optimal cut off value

Question 18

Which 4 conditions are screened for in children?

Answer 18

• Neonatal conditions
- Retinopathy, retinoblastoma, cataracts
• Risk factors for amblyopia
- Cataract, retinoblastoma. Good to find out before 8 as improves prognisis: due to cortical plasticity beyond this age
• Refractive errors
- Anisometropia; most common cause of non-strabismic amblyopia (2-4yrs)
• Colour vision abnormalities
- as may affect learning and excludes individuals from jobs (electrician, train driver)

Question 19

Types if screening for children by age:-

Answer 19

• Newborn exam and 6-8week review
- red reflex test, corneal light reflect for squint, general inspection of eye.
• Preschool screening - 4/5yrs of age
• Screening in school-ages children