Hacker News Reader

Comment by DanBC | Hacker News Reader

False positives don't mean a month of visiting a doctor - they mean several months of anxiety inducing "watchful waiting" or invasive biopsies.

And false positives are not a rare event. False positives are very common.

On top of that the radiation (from CT scans) increase the risk of cancer.

Scans for healthy people are strongly discouraged.

There's a possible benefit if you have any of the genetic markers for lung cancer or are otherwise at risk of lung cancer - heavy lifelong smoker for example.

View on HN

Yep. My wife has yearly scans because of a tumour supressor gene mutation and last year they thought they spotted something on the ultrasound, so sent her for a CT.

CT showed that it was pretty bad, que months of stressing and worrying about what was going to happen.

Turns out it was a false positive, caused by an accessory spleen looking like a tumour. False positives do happen, and they can be pretty devastating - it took quite a while for her to get back to normal.

Would you care to share with us how you were able to determine that she had such a gene? Was it some sort of byproduct of a normal scan/test? Or was it some sort of specificly requested test you went somewhere to get tested for? Or?

I don't know about your country, but genetic screening / testing is something you need to ask for, not something that happens as a by-product of a CT scan.

If you worry that much about a false positive, I'm sorry for you, but you have no idea how much worse things may get. Many people don't get to complain about a false positive!

Stressing and worrying are both several orders of magnitude better than suffering and dying.

Source: My case wasn't a false positive.

In the context of full body scanning: false positives include spots on the lung.

So now you have someone who thinks they have might lung cancer, when in fact they probably have almost harmless scars from chicken pox.

But what do you do if you think you have lung cancer? Do you go and get biopsies? The stress from that result is significant and shouldn't be dismissed as something trivial.

People aren't worrying over a false positive. They don't know it's a false positive. In that moment their worry is just as genuine as yours is.

Of course you're right, but there's a misunderstanding there.

My point is people who went through the experience of having test results compatible with cancer and then discover it is just a false positive (which technically it's not even true, there's no cancer diagnosis before a biopsy, it just can't be excluded), well, these very people seem to speak out against screening, or adopting more extensive screening policies, because they had a "false positive", and "it's stressful".

I am pretty sure all the people whose lives were saved by screening find that pretty selfish.

I think you're misunderstanding something. This has nothing to do with selfishness -- you're only really considering the upside.

The harm/good ratio of screening depends crucially of the false positive/negative ratio vs. the actual prevalence of the condition or disease being screened for. (Plus, obviously the psychological/medical "cost" of a false negative/positive.) I would recommend the talk at [1]. This particular issue is brought up about midway through the video, but I recommend watching the whole thing.

[1] http://www.ted.com/talks/peter_donnelly_shows_how_stats_fool...

Haven't we been through this discussion recently, with regards to screening for breast cancer?

Harms of Screening Mammography: http://www.cancer.gov/cancertopics/pdq/screening/breast/heal...

> And false positives are not a rare event. False positives are very common.

Sure, but if you're scanning daily (or close to it), then you don't need to do anything immediately for a positive. You see a growth that looks like it might be cancer—then have the doctor watch it. If you're scanning often, then you'll see any changes. Does it grow? Then it's something to be concerned about. Does it stay the same? Then you don't need to worry.

With frequent enough scans, finding something isn't a cause for any alarm. You can watch how it changes, and decide if it's a concern based on that. Like skin moles. Yes, they could be a problem, but if you have one, you don't need to worry unless it starts changing. Just having one doesn't cause anxiety.

That's an insane amount of radiation, and would substantially increase your risk of fatal cancer. A single CT scan is roughly a years worth of normal radiation exposure, and then targeted to a single area of the body.

Right, I wouldn't want to do it with CT scans. But I think it's an area where a less dangerous, cheaper scanning method might save a lot of lives, even with our current techniques for treating cancer. Frequent scanning would mean people would find tumors when they're smaller, and only treat the ones which behave dangerously.

edit: I'm replacing my entire previous comment about the superiority of ct/pet-ct vs *mri scanning; it looks like some groups are doing work with MRIs that are comparable in accuracy and nearing clinical use. I still think daily whole body scans are a spectacular waste of human capital for relatively little benefit.

> I still think daily whole body scans are a spectacular waste of human capital for relatively little benefit.

I agree, and I'm not recommending we do that right now. But imagine if we had cheap, fast ways to do the scans. So much money is being pumped into cancer treatment right now, I'm just wondering if better cancer detection might be more beneficial. Our current treatment methods might get us a lot farther because we'd use them when the tumors are smaller, and we'd have a much better sense of how aggressive a tumor is.

And then the lawyers descend on all the false negatives. "Doctor, why did you miss this shadow on this photo? Two years later cancer was found there."

Reply on news.ycombinator.com