[steveBK123]

Same elitist attitude I saw on a recent NYT piece about paid full-body MRIs. "People might find stuff that isn't cancerous and freak out".

OK well, it might also find early stage cancers that show no symptoms until past the point of no return!

MRIs have no side effects aside from the high cost. Even their high cost is reasonably affordable if only done every 5-10 years. As long as doctors & patients make rational follow up decisions with the results, it's a net benefit to be able to get these scans every few years to catch early, slow moving, hard to detect cancers.

There are a wide range of cancers there really are no routine screenings for. Yes we screen for what.. breast, colon, prostate, skin.. But what of liver, kidney, thyroid, pancreas, and various others?

We had a close friend discover they had stage 2 cancer found during a CT scan after a routine medical procedure went awry. They were told that had the slip-up not occurred, they would have probably lived another 5-10 years, and not fallen ill with any symptoms until stage 4.

I don't understand the mindset that we should just pretend the tools aren't available to detect things earlier.

[hgomersall]

It's more complicated than that. Misdiagnosis comes with a high cost. What we emphatically do not have is a way to reliably confirm or stage without additional risky interventions, and that's before we even start discussing the mental health implications of misdiagnosis. So as a doctor, it's not about withholding information for the benefit of the patient, it's being aware that for many cancers, in aggregate, they may very well end up doing more harm than good if they screen for it.

[mathieutd]

I don't understand this logic at all. How can more information be bad? If you see a mass that looks very likely to not be cancer, for which the cost of further investigation is higher than the likely benefit, then the rational patient will agree not to investigate further. I don't see how more information can be bad unless you assume that the patient is an idiot or irrational.

[fidotron]

Exactly. You could even have the option of contributing your MRI scans to academic research so that future identification of cancer improves, and set up a happy feedback loop.

The scary truth is modern western medicine is primarily optimized to extract revenue while reducing spending and improving patient outcomes is merely a side effect of that process. Even in places such as the UK NHS it's all about not finding out things we don't want to know so we don't have to spend money dealing with it.

This is why I look forward to when we can replace doctors (not nurses) with AI.

[magicalist]

> The scary truth is modern western medicine is primarily optimized to extract revenue

MRI scans are a fantastic source of revenue, as are treatments for things that don't actually need to get treated. Reducing those things are actually doing the opposite of the motivation you're claiming.

[fidotron]

Those raise revenue. Actual spending involves effort which cannot be easily industrialized, and is only done to maintain the prestige of the industry.

This is why getting tested for something which results in endless prescriptions is done enthusiastically while a test for something which might find something which requires them doing actual work provokes the sort of self serving concerns expressed elsewhere.

[magicalist]

> This is why getting tested for something which results in endless prescriptions is done enthusiastically while a test for something which might find something which requires them doing actual work provokes the sort of self serving concerns expressed elsewhere.

You'll have to be more specific because right now this is just handwaving. What kind of "actual work" are you referring to?

It's generally actual medical researchers, who will neither get revenue or have to do actual work, who are objecting to excessive testing without patient outcome benefits.

[steveBK123]

Yes. This to me is the same line of thinking as "in a meta study, wearing a helmet makes bikers more reckless and prone to injury so it's actually safer to be helmetless". No, actually it's safest to wear a helmet AND not become reckless.

So similar approach here - its safer to get the imaging AND remain rational in evaluating results & next steps.

[hgomersall]

Because further investigation is dangerous. So you see a mass which has an a posteriori probability of being cancer of 1%, but the investigation causes serious complications in 2% of cases, then the decision to investigate is not clear cut. The additional information has not only not helped but has led to additional stress.

[steveBK123]

Not all further investigation needs to be surgical. A mass that is found can be observed in decreasingly frequent ultrasounds or some other imaging and surgically investigated/removed only if found to be growing or passed a concerning size threshold.

A doctor jumping straight to invasive procedures seems to be a mix of poor risk management and rarity of this type of medical imaging.

My doctor for example, pointed out that actually in some East Asian countries, there are routine annual imaging tests done that pick up some of the types of cancer we do no screening for.

To me the reason we don't in US is simply how medical care is paid for - employer provided insurance, and some actuarial calculation that on the insured pool they'd spend more money on imaging than they'd save on high cost stage 4 cancer care. Personally I'm happy to advocate more for myself, even if it costs money.

[hgomersall]

Do you have any experience of cancer staging or medical imaging? It certainly sounds like you don't.

[bell-cot]

Both human nature and the legal system can be very hostile to "we didn't investigate anomaly A, B, C, D, E, ... in the patient's scans and test results, because none of them seemed likely to be worth the costs of doing so".

[KaiserPro]

Full body scans have quite some history. If they were effective at routine preventive detection, the NHS would deploy them in a flash, because its cheaper right?

The problem is the false positive rate is >> catching unknown bad things.

its the same with breast cancer in the UK there is a 3.1% false positive rate. https://digital.nhs.uk/data-and-information/publications/sta... which requies follow ups. Now as breast surgery is reasonably uncomplicated (source: wife did breast surgery in training) its not _much_ of a risk and is worth it.

However, if its something in the liver, brain or spinal column, the risk its pretty high. This leads to a higher chance of injury from surgery vs stopping something unknown.

This is why something that actually identifies cancer cells is much better than looking for smears on an image. Unless you have monthly MRI scans, from birth, you are going to get nasty side effects from invasive investigations.

EDIT: also most people don't really understand the difference between CAT and MRI scans. CAT scans are much cheaper, faster, and better at finding cancer (you can use dies and junk). given the difference in cost, time and comfort, a lot of people will choose a CAT scan instead. However regular CAT scans will give you a much higher risk of cancer. Something the kardashians pushing whole body scans will neglect to tell you.

[steveBK123]

Thanks for the edit and even the NYT article obscures this.

My point on MRI is that they do not themselves have side effects.

If you have a doctor that immediately sends you for a bunch of CAT scans and/or cuts you open, then obviously there are side effects.

And why would you immediately jump to either? If it's the first ever scan, and you see something unusual it could be monitored by a 6month/1year ultrasound and/or MRI followup.

Unfortunately it doesn't seem like any of the direct cancer detection solutions are there yet either. There's a recent startup that claims something like 5% detection of stage 1 / 10% detection stage 2 / 25% detection stage 3 / etc on a set of cancers, but they also just accidentally mass-mailed a bunch of negative patients that they have cancer.

You also express a false dichotomy - single MRI bad, but monthly MRIs for life good?

A sober reading of annual MRI/Ultrasound type tests without knee-jerk invasive followups when you are 30+ seem like a reasonable risk weighted solution in contrast, doesn't it?

[KaiserPro]

> You also express a false dichotomy - single MRI bad, but monthly MRIs for life good?

yeah, this is badly expressed on my part. I was trying to get across that a single whole body scan without context (ie it hurts here, or it bleeds there or we suspect x) is difficult to interpret. think of it as a day's unstructured logs. Regular scans allows you to build up a picture of whats changing, and whats normal for you.

> A sober reading of annual MRI/Ultrasound type tests without knee-jerk invasive followups when you are 30+ seem like a reasonable risk weighted solution in contrast, doesn't it?

I think routine targeted scanning is something that is worthwhile. The UK does a number of them, and they were normally based on evidence of outcome. Prostate/breast/cervical etc etc. I personally think the future of public health is something akin to getting each personal a vitals dashboard.

But, I'm not sure regular MRIs will give us that. if the evidence changes though, then it should be reassessed.

[DrJaws]

that have a reason, as it's been demostrated by a lot of metastudies you can find on cochrane that there is usually much more worst outcomes and long term effects on the broad of the population when misdiagnosed by overdiagnosing than just simply saving an extra 0.01% (not real number)

the same reason of why for example now there is an advocacy to end yearly mammograms on older woman, because the number of them saved by that practice is inferior to the ones that are misdiagnosed and then put under other unnecesary medical practices that end up hurting more by unnecesary practices on a lot of them that would have never developed a cancer or under pressure to the ones that no one will be able to save no matter how sooner they got the diagnostic.

infinite constant and unnecesary medical tests is not the way for now, maybe in the future, but not now.

[steveBK123]

Isn't this more a product of relative rarity of this type of imaging & average doctor not knowing how to react properly other than escalation?

Not ever spec on an image should mean cutting someone open or blasting with radiation.