Diagnostic study statistics

A test has a sensitivity of 95%. What does this mean?

• 95% of people who test positive actually have the disease

• 95% of people with the disease will test positive

• 95% of people without the disease will test negative

• 95% of all tested people have the disease

A test with high specificity is most useful for:

• Ruling out a disease

• Increasing false positives

• Ruling in a disease

• Screening large populations

What is the Positive Predictive Value (PPV)?

• The probability that a person has the disease if the test is positive

• The probability that a person does not have the disease if the test is negative

• The proportion of all true positives among all tested people

• The ability of a test to detect disease early

Which of the following affects the PPV the most?

• The test’s sensitivity

• The test’s specificity

• The disease prevalence

• The confidence interval

A test has a negative likelihood ratio (−LR) of 0.1. What does this indicate?

• A negative result does not change the probability of disease

• A negative result strongly decreases the probability of disease

• A negative result confirms the disease

• The test has poor specificity

Which combination is best for screening purposes?

• High specificity and high PPV

• High sensitivity and low cost

• High NPV and high likelihood ratio

• High prevalence and low sensitivity

Incorporation bias in diagnostic studies refers to:

• Using the same clinician to perform both the index and reference tests

• Selecting only healthy volunteers for the control group

• Failing to blind the test interpreters

• Including the index test as part of the reference standard

Which of the following best describes spectrum bias in diagnostic accuracy studies?

• When the index test is interpreted with knowledge of the reference standard

• When study participants do not represent the full range of disease severity seen in clinical practice

• When only positive test results are verified with the reference standard

• When the reference standard is imperfect and misclassifies disease status

What does a negative likelihood ratio (−LR) of 0.5 indicate?

• A negative result moderately decreases the probability of disease

• A negative result confirms the absence of disease

• A negative result has no effect on disease probability

• The test is highly specific

You're evaluating a new rapid point-of-care test for Influenza A in a study of 200 patients. PCR confirms 80 cases. Of these, 72 test positive and 8 negative on the rapid test. Among 120 PCR-negative patients, 92 test negative and 28 test positive on the rapid test. What is the sensitivity of the new test?

• 72%

• 90%

• 80%

• 88%