Cost of False Positives

http://www.ncbi.nlm.nih.gov/pubmed/15598770

This study shows that with the increase in routine cancer tests, there is a significant increase in the number of people receiving false positive test results. The costs of those false positives includes both financial costs (over $1000 per person over the year following a false positive) and a huge emotional stress on both the individual and their family. I know these tests save a lot of lives, but I wonder if the reflexive testing of people who aren't in "at-risk" categories has more negative effects than positive effects. This study's sample group had a full 43% of participants receiving at least one false-positive test when tested for multiple types of cancer.

from Wikipedia:

"False negatives may provide a falsely reassuring message to patients and physicians that disease is absent, when it is actually present. This sometimes leads to inappropriate or inadequate treatment of both the patient and their disease. A common example is relying on cardiac stress tests to detect coronary atherosclerosis, even though cardiac stress tests are known to only detect limitations of coronary artery blood flow due to advanced stenosis.

False negatives produce serious and counter-intuitive problems, especially when the condition being searched for is common. If a test with a false negative rate of only 10%, is used to test a population with a true occurrence rate of 70%, many of the "negatives" detected by the test will be false. (See Bayes' theorem)

False positives can also produce serious and counter-intuitive problems when the condition being searched for is rare, as in screening. If a test has a false positive rate of one in ten thousand, but only one in a million samples (or people) is a true positive, most of the "positives" detected by that test will be false."

To put it simply, this kind of sucks. The best way to minimize this kind of error is by having a large sample size, but you really can't have a large sample when it is a small minority of people that actually have the disease. And even with a large sample, each individual that receives a result may be affected by it in a very significant way.

You could even extend this to vaccines. If the majority of people get vaccinated for a disease, but only a very small percentage of those people will get exposed to the actual disease and there is some risk from receiving the vaccine, is it worth it? I guess receiving a vaccine and getting the disease later can be compared to a "true positive," while receiving a vaccine and not getting the disease can be compared to a "false positive." Just something to think about...

Whenever I hear that something is 99.9% effective, in my head I just round that up to 100%. I found this article to be interesting because it just shows how magnified that .1% can be when tested over millions of people. Although false negatives are of course much worse when it comes to cancer, I could see how a false positive would be extremely emotionally scaring and costly. There was an episode of the television series “House” in which he was suing the hospital over a false positive. He had spent all of his money and went on a last vacation only to find out he really didn’t have cancer. Not that this is a real person, but it could very well be a true scenario.

See, this is why statistics scare me.

Personally, I think that any test is better than no test, unless its false positive/negative occurrence is so bad you're not actually saving lives anymore. After all, how are you supposed to make a test better and more accurate if you don't try it and gather data regarding its problems? I think as long as people are warned of the possibility of a false result, the issue isn't so bad.