You are assuming errors would be random, which is probably false. A false positive would probably be more likely for someone with similar symptoms, than for som

You are assuming errors would be random, which is probably false. A false positive would probably be more likely for someone with similar symptoms, than for someone without any symptoms.