Home › Statistics › The false positive paradox: when a 99% accurate test is usually wrong
The false positive paradox: when a 99% accurate test is usually wrong
A "99% accurate" test for a rare condition can be wrong most of the times it says yes.
The puzzle
A condition affects 1 in 1,000 people. A test for it catches 99% of people who have it, and correctly clears 99% of people who don't. You test positive.
What's the chance you actually have it? Most people say about 99%. The real answer is about 9%. More than nine out of ten positive results are false alarms.
🧪 Test 1,000 people
The picture shows the expected numbers out of 1,000, rounded to whole people; the percentage is exact. These are illustrations, not figures for any real test.
Why it happens: the healthy outnumber the sick
Picture 100,000 people. With a 1-in-1,000 condition, 100 have it and 99,900 don't.
- The test catches 99 of the 100 who have it.
- It wrongly flags 1% of the 99,900 who don't: that's 999 false positives.
So about 1,098 people test positive, and only 99 of them have the condition: 9%. The test's 1% error rate is applied to a huge healthy group, and that swamps the small number of true cases.
This is the base rate: how common something is before you test for it. Ignoring it is called the base rate fallacy, and it's one of the most common reasoning mistakes, including among doctors, lawyers and juries.
Accuracy has two halves
- Sensitivity is how many real cases the test catches. High sensitivity means few missed cases.
- Specificity is how many people without the condition the test correctly clears. For rare conditions, this is the number that matters most, because every fraction of a percent becomes a crowd of false positives. Try moving it from 99% to 99.9% in the simulator.
Bayes' rule
P(has it | positive) = (sensitivity × prevalence) ÷ (sensitivity × prevalence + (1 − specificity) × (1 − prevalence))
Psychologist Gerd Gigerenzer showed that people get these questions right far more often when they're written as counts ("10 out of 1,000") rather than percentages, which is why the simulator shows a crowd of 1,000 people.
What it doesn't mean
It doesn't mean tests are useless. A positive result still raises the chance a lot (from 0.1% to 9% in the example), and a confirming second test can push it much higher. It's why screening programs follow a positive result with a different, more specific test. And if you have symptoms, you're no longer a random member of the population: your starting chance is higher, so a positive result means more. Always discuss real results with a doctor.
Questions
What is the false positive paradox?
When a condition is rare, even a very accurate test produces more false positives than true positives, so most positive results are wrong. It happens because the small error rate applies to the large number of people who don't have the condition.
What is the difference between sensitivity and specificity?
Sensitivity is the share of people with the condition who test positive. Specificity is the share of people without the condition who test negative.
Does a second test help?
Yes, if the second test is independent. Those who tested positive the first time are a group where the condition is far more common, so a second positive result is much more convincing.