4 ms·
Something is not clear to me. How do you know that a false positive is, in fact, a false positive? I presume there are some other pieces of information that pr
by lordnacho 11y ago
Something is not clear to me. How do you know that a false positive is, in fact, a false positive?
I presume there are some other pieces of information that provide the reference value.
So when someone gets a positive test, why not put them through the next test to see if they keep coming up positive?
What's the argument for not testing in the first place? It will panic people? If that's the case, why not make it random whether you're called to an extra test?
- DanBC 11y ago> Something is not clear to me. How do you know that a false positive is, in fact, a false positive? Bob has a full body MRI. THat returns some shadowy spots on his lungs. He's asked if he used to smoke (he did), and if he had any childhood illnesses (german measles, measles, chicken pox, etc) (he did). Are those spots early cancer? Or are they just scarring from the childhood illness? One way is to cut Bob open and have a look. Or jab him with a massive biopsy needle. Except, now you know that this one spot isn't cancer, it was a scar. How about all the rest? That one biopsy increased Bob's risk of harm. > What's the argument for not testing in the first place? Testing often doesn't give any useful information. It often gives wrong information - people who have a disease are told they don't have it, and people who don't have it are told they do have it. Many of these people then go on to have treatment. All treatment carries some risk of harm. You need to know if the benefits of testing (especially of testing a large, mostly healthy population) outweigh the risks of providing risky treatment to those people, some of who will not have the disease. You also have to see what happens if you just don't do anything. Prostate cancer is a good example here. Some people will die of it - they have an aggressive cancer that will kill them rapidly. But many people will die with it - they have a slow growing tumour, and they'll die in old age from something else. Treatment for prostate cancer is pretty harsh.
- Jtsummers 11y ago> What's the argument for not testing in the first place? The wider the pool of people being tested, the greater the chance of false positives, which is why screening guidelines generally limit the population to be screened. The more independent tests you do at once, each with its own chance of error, the larger the chance that at least one of those tests produces an incorrect result, said Rebecca Goldin, director of STATS.org and a professor of mathematical sciences at George Mason University. It's not just panic, but that if the false positive rate is too high then untargeted testing becomes overwhelming and counterproductive. There's a slim, very slim, chance that as a man I could develop breast cancer. But there's little value in testing me as it's not present in my family history (male or female). So let's say (in this example) that the false positive rate is only 1% in men and the true incidence rate is 0.01%, and we test 100 million men: 10,000 men have breast cancer, great we've detected it 999,900 men don't, but are subjected to more testing That more testing is likely costly, time consuming, and potentially invasive. This acts as a net drain on medical resources because we've just performed nearly 1 million unnecessary mammograms. They take technician and physician time. They take machine time. And so on. So false positive rates may be low, but when applied to a massive population the total number of false positives can overwhelm the system.