50 / 194
Pre-event/Post-event
• Test > 90% sensitivity
• Pre-event prevalence of smallpox is zero:
– If test has 95% specificity, 10 tests done per month, in 6 to 8
sites, every month there will be 3 or 4 false positives
• Post-event prevalence of smallpox is finite:
– If test has 95% specificity, 1000 tests done per week, every
week 50 results will be false positives
– If test has 99% sensitivity, 1000 tests done per week, 10
results will be false negatives
Test Parameters
• Sensitivity and specificity are independent of the
l
f di
preva ence o sease
• Predictive value positive and negative vary with
disease prevalence (Bayes’ Theorem)
• Implications for smallpox testing “pre-event” and
“post-event”: use of tests for decision making