ch 8 evaluating clinical evidence

What is sensitivity?

The probability that a person who HAS the disease will have a positive test; also called the true-positive rate.

What is the formula for sensitivity?

a / (a + c)

What does high sensitivity mean?

The test detects most people who actually have the disease, so there are few false negatives.

What does SnNout mean?

A highly SeNsitive test with a Negative result helps rule OUT disease.

What is specificity?

The probability that a person who does NOT have the disease will have a negative test; also called the true-negative rate.

What is the formula for specificity?

d / (b + d)

What does high specificity mean?

The test correctly identifies most people who do not have disease, so there are few false positives.

What does SpPin mean?

A highly SPecific test with a Positive result helps rule IN disease.

What is a true positive?

The patient has the disease AND the test is positive. Cell a.

What is a false positive?

The patient does NOT have the disease but the test is positive. Cell b.

What is a false negative?

The patient HAS the disease but the test is negative. Cell c.

What is a true negative?

The patient does NOT have the disease AND the test is negative. Cell d.

What is positive predictive value (PPV)?

The probability that a person with a POSITIVE test actually HAS the disease.

What is the formula for PPV?

a / (a + b)

What is negative predictive value (NPV)?

The probability that a person with a NEGATIVE test truly does NOT have the disease.

What is the formula for NPV?

d / (c + d)

What is the major difference between sensitivity/specificity and predictive values?

Sensitivity and specificity start with whether disease is present or absent; PPV and NPV start with the patient's test result.

What is disease prevalence?

The proportion of people in a population who have the disease.

Does prevalence affect sensitivity and specificity?

Sensitivity and specificity are generally properties of the test and do not change simply because prevalence changes.

Does prevalence affect predictive values?

YES. PPV and NPV can change substantially when disease prevalence changes.

What happens to PPV when disease prevalence decreases?

PPV decreases; a larger proportion of positive results may be false positives.

What happens to PPV when disease prevalence increases?

PPV increases because a positive result is more likely to represent true disease.

What generally happens to NPV when prevalence decreases?

NPV increases because a negative test is more likely to truly represent absence of disease.

What generally happens to NPV when prevalence increases?

NPV decreases.

In the textbook example with 10% prevalence and 90% sensitivity/specificity, what was the PPV?

50%.

In the textbook example with 1% prevalence and the same 90% sensitivity/specificity, what happened to PPV?

It fell to about 8.3%.

Why can screening low-prevalence populations produce many false positives?

Because very few people actually have the disease, so false positives can outnumber true positives even when the test has good sensitivity and specificity.

What is a likelihood ratio (LR)?

A measure of how much a diagnostic test result changes the probability that a patient has a disease.

What is pretest probability?

The probability that the patient has the disease BEFORE the diagnostic test result is known.

What is post-test probability?

The probability that the patient has the disease AFTER the diagnostic test result is known.

What is LR+?

The likelihood ratio for a positive test; it tells how strongly a positive result increases the probability of disease.

What is the formula for LR+?

Sensitivity / (1 - specificity)

What type of LR+ is most useful?

A HIGH LR+; the larger the value, the stronger the positive result helps rule IN disease.

What is LR-?

The likelihood ratio for a negative test; it tells how strongly a negative result decreases the probability of disease.

What is the formula for LR-?

(1 - sensitivity) / specificity

What type of LR- is most useful?

A LOW LR-, especially one close to 0; it helps rule OUT disease.

What does a likelihood ratio near 1 mean?

The test result changes the probability of disease very little.

What is the relationship between sensitivity and false negatives?

Higher sensitivity = fewer false negatives.

What is the relationship between specificity and false positives?

Higher specificity = fewer false positives.

If the goal is to avoid missing a serious disease, what test characteristic is especially important?

High sensitivity.

If the goal is to confirm/rule in a diagnosis after a positive result, what test characteristic is especially useful?

High specificity.

EXAM: Patient has disease + positive test = ?

TRUE POSITIVE.

EXAM: Patient has no disease + positive test = ?

FALSE POSITIVE.

EXAM: Patient has disease + negative test = ?

FALSE NEGATIVE.

EXAM: Patient has no disease + negative test = ?

TRUE NEGATIVE.

EXAM: Sensitive + Negative = ?

SnNout → rule OUT.

EXAM: Specific + Positive = ?

SpPin → rule IN.

EXAM: Which values change with prevalence?

PPV and NPV.

EXAM: Low prevalence has what effect on PPV?

Lowers PPV and increases the proportion of positive tests that are false positives.

EXAM: High LR+ means what?

A positive result strongly increases the likelihood of disease.