Epidemiology (III) Screening Basics

Question 1

What type of prevention is screening?

Answer 1

Secondary (early detection of disease)

Question 2

What is the difference between a screening test and a diagnostic test?

Answer 2

Screening tests - detect early disease or risk factors for disease in large numbers of apparently healthy individuals. (identification in asymptomatic patients) Diagnostic tests - establish the presence (or absence) of disease as a basis for treatment decisions in symptomatic or screen-positive individuals (confirmatory test). (confirmation in symptomatic patients)

Question 3

What part of the disease’s pathogenesis must be present for a screening to be useful?

Answer 3

A detectable preclinical phase

Question 4

What are some considerations to make before developing a screening tool?

Answer 4

Is there a long, recognizable preclinical phase?

Can we improve the prognosis with earlier detection?

Is the test valid and reliable?

Is it sufficiently cheap and noninvasive?

How harmful can the test be?

Question 5

What is external validity?

Answer 5

Generalizability of results to the general public

Question 6

What is internal validity in screening?

Answer 6

The test measures what it is designed to measure

Question 7

What is reliability in screening?

Answer 7

Consistency and replicability of the test

Question 8

What are two measures of a screening test’s validity?

Answer 8

Sensitivity and specificity

Question 9

What is screening sensitivity?

Answer 9

The ability of the test to correctly identify those with the disease

(the proportion of people with the disease who test positive)

Question 10

A test with a high sensitivity has a low what?

(Hint: it is measured by 1 - sensitivity)

Answer 10

A low false-negative rate

Question 11

For what type of test is a high sensitivity especially important?

Answer 11

Screening tests

Question 12

Why are false negatives damaging?

Answer 12

They can lure diseased patients into a false sense of security and delay disease detection.

Question 13

What is screening specificity?

Answer 13

The ability of the test to correctly identify those who do not have the disease.

(the proportion of people without the disease who test negative)

Question 14

A test with a high specificity has a low what?

(Hint: it is measured as 1 - specificity)

Answer 14

A low false-positive rate

Question 15

For what type of test is a high specificity especially important?

Answer 15

Confirmation tests

Question 16

Why are false positives damaging?

Answer 16

They may cause unnecessary anxiety and panic (think of a diagnosis with HIV or metastatic pancreatic cancer)

Question 17

How is sensitivity measured?

Answer 17

TP / (TP + FN)

(true positives divided by everyone that has the disease)

(how many of the diseased individuals did the test successfully identify?)

Question 18

How is specificity measured?

Answer 18

TN / (TN + FP)

(true negatives divided by all the healthy individuals)

(how many of the healthy individuals did the test successfully identify?)

Question 19

How can we identify which test should be the gold-standard diagnostic technique?

Answer 19

The test with the desired sensitivity and specificity is the gold-standard

Question 20

What is important about a test’s cutoff point for a disease case definition?

Answer 20

It determines the properties of the disease (high sensitivity or high specificity)

Question 21

What happens to specificity as sensitivity increases?

Answer 21

Specificity decreases (and vice versa)

Question 22

What questions about false-negatives and false-positives should be asked when selecting how sensitive and specific a test will be?

Answer 22

What is the consequence of an undetected case (can we wait to test again later, or does the disease progress rapidly)?

What is the consequence of calling a disease-free person a positive (will there be invasive tests, stress, high costs, or time wasted)?

Question 23

What does the ‘ROC’ in ROC curve stand for?

Answer 23

Receiver operating characteristic

Question 24

How is an ROC curve formed?

Answer 24

By plotting the true positive rate (sensitivity) against the false positive rate (1 - specificity)

(basically, probability of detection vs. probability of a false alarm)

Question 25

Where will a cutoff point do the least harm in the overlapping distributions of health and diseased individuals?

Answer 25

Question 26

How can the best cutoff point be determined from an ROC curve?

Answer 26

The point nearest the upper lefthand corner