Predictive Values and Prevalence

#Start with the test result

Positive predictive value is the proportion of people with a positive test result who actually have the defined condition. Negative predictive value is the proportion of people with a negative result who do not have it. These measures answer different questions from sensitivity and specificity.

Sensitivity and specificity start with disease status and describe how the test behaves. Predictive values start with the result and describe how often it matches disease status. They are particularly useful for understanding why a positive result may need confirmation and why a negative result may not settle every concern.

#Why prevalence changes the meaning

Prevalence is the proportion of a defined population that has a condition at a particular time or during a specified period. If sensitivity and specificity stay the same, a lower prevalence generally lowers positive predictive value. When few people have the condition, false positives can make up a substantial share of positive results.

Under the same assumption, lower prevalence generally raises negative predictive value. When the condition is more common, a positive result is more likely to represent disease, while a negative result is less reassuring. The test itself may be unchanged even though the meaning of its results shifts.

#Population figures and individual context

The starting likelihood for someone undergoing assessment is not always the same as prevalence in the wider population. Symptoms, exposures, medical history and earlier findings can change that likelihood. A specialist clinic may therefore see different predictive values from a screening programme using the same test.

Published predictive values should be interpreted with attention to the population studied and the testing pathway. Sensitivity and specificity can also vary between settings, adding uncertainty. Clinicians combine the result with the prior evidence rather than treating a positive result as certainty or a negative result as complete exclusion.

#Common misunderstandings

A positive result does not always mean the condition is present, and a negative result does not always rule it out. Predictive values describe how often those results match the condition’s actual presence or absence in a particular setting.

One common mistake is to confuse sensitivity with positive predictive value. Sensitivity asks how often a test detects the condition among people who have it. Positive predictive value asks how often people with a positive result actually have the condition. These are different questions.

Another misunderstanding is that a test’s predictive values are fixed. Even when sensitivity and specificity stay the same, predictive values change with prevalence. When a condition is uncommon, false positives can make up a substantial share of positive results.

A high negative predictive value also needs context: it may partly reflect that most people tested did not have the condition, rather than showing that the test rarely misses cases.

#Questions worth asking a clinician

• How do my symptoms, risk factors, and the condition’s prevalence affect my chance of having it before testing?
• If my test result is positive, how likely is it that I actually have the condition?
• If my test result is negative, how likely is it that I still have the condition?
• Does this test produce more false positives when the condition is uncommon in people like me?
• Given my chance of having the condition before testing, would a positive or negative result need confirmation with another test?