Diagnostic tests and statistics
Sensitivity, specificity and predictive values answer different questions; disease prevalence changes predictive value.
Understand
Sensitivity and specificity describe performance against a reference standard; positive and negative predictive values also depend on prevalence. Likelihood ratios help update probability.
Assess
Use a two-by-two table to check calculations and distinguish screening from diagnosis. Think about the consequences of false positives and false negatives.
Apply
Choose a test only when its result could change action. Do not claim that a negative imperfect test rules out a high-risk condition without considering pre-test probability.
Key distinctions
A test's sensitivity concerns the proportion of diseased people it detects; specificity concerns the proportion without disease it correctly identifies. Predictive values change with prevalence even if the test itself is unchanged. Screening in a low-prevalence population can produce many false positives. Before ordering a test, decide how positive and negative results would change care.
Think through a case
A positive screening test in a low-prevalence group may include many false positives. Work through a simple two-by-two table before interpreting it.
The distinction to remember
Do not claim sensitivity is the chance that a positive patient has disease; that is a predictive value.
Test your recall
Why can a positive test be less convincing in a low-prevalence group?
Show answer
False positives make up a greater share of positive results.