If there is a medical test that is accurate 99% of the time about a disease that occurs in 1 out of 10,000 people, then testing one million people would approximately yield the following results:
Healthy and test indicates no disease (true negative)
Healthy and test indicates disease (false positive)
Unhealthy and test indicates disease (true positive)
Unhealthy and test indicates no disease (false negative)
If a patient received a positive response from the test the odds are ~99.02% (9999/10098) that they are healthy and the test is incorrect even though the test is 99% accurate.
Discriminating between true-positive and false-positive clinical mastitis alerts from automatic milking systems
Jun 01, 2010; ABSTRACT Automatic milking systems (AMS) generate alert lists reporting cows likely to have clinical mastitis (CM). Dutch farmers...