In an outbreak alert system, what does positive predictive value measure?

• A. The probability that an alert truly indicates an outbreak given the baseline prevalence and test characteristics; higher PPV reduces false alarms.
• B. The probability that a lab test is positive for a disease when disease is present.
• C. The probability that an alert is generated for every case.
• D. The probability that an outbreak has occurred with certainty.

Answer: (A)

Positive predictive value is the probability that an alert truly indicates an outbreak given that an alert has occurred. In other words, it’s P(outbreak | alert). This measures how trustworthy a positive alert is, incorporating how common outbreaks are in the population (baseline prevalence) and how the alert system performs (its sensitivity and specificity). When outbreaks are rare, even a decent alert system can produce many false alarms, which lowers PPV. When outbreaks are more likely or the test distinguishes well between outbreak and non-outbreak conditions, PPV is higher, and a positive alert is more likely to reflect a real outbreak. This concept is different from sensitivity (the chance an alert appears when an outbreak is present) and from the idea of generating an alert for every case, or asserting certainty that an outbreak has occurred.