Fractionalized Ownership of Medical Diagnostic Devices?

Hey, I’ve been thinking about the whole health insurance reform issue, and I don’t really understand the politics of it very well, so I tried to think of what policies I would put in place if I was a health insurance company who wanted to do more than just make money, but in addition help people (and possibly implemented using social networks).

I came across a model of medical device usage and payment that was analogous to the computer timesharing model. Let’s focus exclusively on cancer diagnostic equipment, and MRIs specifically, because they are much safer than CAT scans o X-ray machines.

I have a strong belief, although this may be unfounded, that if the frequency of cancer scans increased beyond a certain threshold over the entire population of adults over a certain age, that a large enough proportion of cancers would be found at an early enough stage for treatment to be effective. For instance, if every person over forty years of age were able to be tested for the twelve or so most common cancers every six months, and these tests were extremely sensitive yet excluded false positives, then these twelve cancers, if they had developed, would be detected by their sixth month of progression, at most, and on average by three months, if it was detectable at such a state. Presently, such a comprehensive diagnostic regimen wouldn’t be carried by any insurance company that was looking to stay in business, and without insurance, would be so expensive that only very wealthy people could afford it.

But computers were once used only by large corporations and the government. What are computers? They are machines. I believe that MRI diagnostic devices are similar to computers in that they are both machines, which by definition can be mass produced. Of course, just because something can be mass produced doesn’t mean that it will be. So far, MRI machines have a small market; namely, hospitals, medical school facilities, radiologist private practices, etc, just as computers once had a similarly limited market.

What if this present state of affairs regarding MRI machines can be changed, so that mass production can be initiated? I think it can, and that the key to doing so is asking the question of who owns the machine?

MRI machines are too expensive for the average person to purchase, and so are purchased by institutions and physicians, who either lease them from a manufacturer or intermediary, or purchase them directly, obtaining full ownership. Regardless of which option is taken, the institution or physician who offers access to the MRI’s use value has full effective control of the machine. In a world where few control the capital good that is so effective at diagnosing cancer, their usage per scan is going to be very expensive.

I figured it would be better for individual workers in a company, say, or even a group of people who live in the same neighborhood, to agree to come together to buy a single machine, or two or three (and perhaps at a group discount if more than one machine is purchased), and agree to split evenly the costs of the machines purchased, each individual also taking that even share of ownership in the machines so purchased. What they get from this deal is lifetime usage of the machine for themselves and anyone they so choose to allow to use it for free, within their fraction of the machine’s time/use value,and an opportunity to earn income from the machine by selling units of their own time allotment to others, either for a market price, or at a mandated discount price that would make cancer scans substantially more affordable for the average person. Since the number of people in the pool who group-bought the machine(s) is large, each person has to put in only a small fraction of the total cost, which would be exorbitantly high otherwise, allowing them to have a reasonable chance at not only having access to cancer scans for free in perpetuity (as long as the machine “lasts,” which can be 20 years, although they may become outdated and no longer cutting-edge), but also in recouping their initial purchase by selling a portion of their time to third parties (at a mandated discount price).

This takes advantage of the fact that the MRI machine, over a period of a year, can be used by an extremely large number of people; its use-value is directly proportional to the number of diagnostic scans run per unit time. So one person, or even a relatively small group of people, cannot make full use of an MRI machine’s use value, if they are using it only to scan themselves for cancers. For much of the time, the machines will be sitting around, not being used. But the present state of affairs is that people in general use the machines, but don’t have (even a) fractional ownership of them. I’m not a Marxist by any means, but wouldn’t Marx’s idea of collective ownership of the means of (cancer diagnosis, not really “production,”), implemented using free markets and voluntary participation, through either pre-existing institutions and large groups, such as workplaces, schools, that are already recognized by insurance companies as qualifying for group discounts, and self-selected organizations that were made through say, social network websites ( anyone?), or simply people who live next to each other on the same street and couple blocks down, result in the first steps toward mass-production, and eventually a long-run possibility of expensive medical devices being reduced in price by an order of magnitude or so, bypassing the insurance companies and physicians, so that a group of people have health clinics with machines that you walk in at an appointed time, with algorithms optimizing the time, with the machines owned collectively by the people…not unlike a computer timesharing place or even a Kinko’s, where you go in and get what you need “scanned,” and a computer algorithm analyzes the image, detects or doesn’t detect the cancer with a 99.99% success rate/no false positive rate, with the result being that the number of people who die of cancer in America drops down to the number who die of smallpox nowadays, since everyone pretty much finds their cancer, if it develops, early enough so that it’s almost a done deal that it will be treated without remission?

If a cure for cancer is discovered, that would be phenomenal. But I think that a quick and dirty solution for reducing the number of cancer deaths, although perhaps not the number of incidence of cancers, per year, to a tiny fraction of what it is today involves giving power to the people by having them obtain fractional ownership of the machines. When we get over the cult of the expert, namely that only doctors can be trusted to run these machines, and that the people cannot do this themselves safely, we will take our health into our own hands, and live happily (for a longer time than we are doing now).

Do you think social networks can be used to organize groups of people to purchase medical devices, who then own these devices and are both users and sellers of time on the machine, and that this will eventually lead to cancer scans being so accessible that the actual number of cancer deaths can be reduced significantly, even if the cancer incidence rate does not? Is this idea legal, or is purchasing medical devices and operating them limited to specific groups of people and institutions?


I think a big issue with this is that you still need a physician to interpret results or possibly operate the machine at all. Suppose you have an MRI that is owned by the Park Valley Neighborhood Collective. You call Dr. John O’Connor and tell him you have 10 results you want him to interpret. He quotes a rate that is marginally lower than what he would charge had everyone went to his office and used the MRI at the Medical Office of O’Connor, Schmidt, Perkins, and Johnson.

There’s also the matter of upkeep. Doctors, would you trust an MRI, CAT scanner, or other large ticket medical device that you were told was just sitting in the basement of the First Street Baptist Church for the past two years and taken care of by deacons with no medical knowledge?

Hmm, it could be made so that each group that creates a collective has to have a physician in it (or more reasonably, a minimum physician to collective group # of people ratio), which is probably not too much of a problem, who oversees the upkeep and interprets the result, and receives a share of each scan performed? One could perhaps develop computer algorithms to aid the process, as a computer algorithm that works for such a task is an application of “machine vision.” The good thing about algorithms is that if they are made right, they get the right answer almost every time, so that the only form of error you have to worry about is systematic error, more so than random error. Of course, this would mean cross-examining and verifying the algorithm with human doctors.

Unfortunately, your ideas about early prevention and treatment of cancer by massive screening are not borne out by the evidence, and seem to require magical screening technology which doesn’t exist. For example, it’s highly questionable whether lung CT screening scans of high risk individuals like smokers does much to reduce lung cancer mortality, let alone MRI pan-scans for the entire population every 6 months.

If you ask any person who finds that they have cancer either when symptoms first manifest, which is already often at at a medium or late stage, and has only six months left to live or has a prognosis of 50% or less chance of recovery and remission, if they wish they could have done a scan for that cancer earlier on, they would irrefutably say yes.

It would be so painful and so obvious, and one wouldn’t want to ask this question because there’s nothing the person can do about it anymore.

These screenings involve technology that is far from magical. They exist, and the reason why it might not be borne out by the evidence isn’t because these diagnostic tools don’t work, it’s because they’re not done enough, imho.

It’s like saying, “we’ve vaccinated some percentage of the people and they’re still dying of these infectious diseases…where’s the evidence?” That’s why we have universal vaccination. An idea that is partially implemented that is marginally effective may become much more so if it is implemented thoroughly. Of course, cancer and infectious disease are hardly comparable in their causes and treatments, and there is a different strategy of “fighting” these two classes of diseases.

But as long as we don’t have a cure, we have to look for the next best thing, and for the sake of humanity and our fellow citizens, do something about it. We’re already fighting a metaphorical war on cancer, and spending billions of dollars.

How is it highly questionable that lung CT screening scans of high risk individuals does much to reduce lung cancer? I agree with you that the results might not be showing up. But just because a method doesn’t work at a certain level of thoroughness doesn’t mean that it won’t when it’s implemented more frequently. I think we’re at the point where everyone over 50 is a high risk individual. If you had a hundred million dollars, and wanted to do everything you could to reduce your mortality from cancer, there are many devices and tools that you could use to do something about it. The problem is, not everyone can afford to do this, and like all other aspects of life, resources are scarce, especially for people who are not wealthy.

My vision is that people will be able to, one day, know that if they access these widely available services, then they don’t have to worry about cancer taking their lives early. It’s a public health issue, and if the government won’t do something, it’s up to us to organize our own resources to create a substitute that will do something.

I feel that we’re at a tipping point where the cost of healthcare will first increase steadily, and then decline precipitously, with the quality and availability of healthcare not being sacrificed. If the technology wasn’t there, then I would be far more pessimistic, but it’s there, I believe. It’s only a question of finding optimal ways to connect the technology with the needs of people.

Your faith in the diagnostic accuracy of MRI is wildly over optimistic. I’m not even going to get into your idea that you can get a computer algorithm to interpret the images - don’t you think that if that were currently achievable hospitals would be saving money on expensive radiology doctors?

Coming back to MRI - I’m going to make up some numbers which are fake, but in the ball park of real world numbers. You take 10,000 apparently healthy people and ten of them have a currently undetected cancer. You MRI them. The MRI is 90% sensitive and 90% specific for detecting “cancer” (these numbers are better than a real MRI).

Of your 10,000 people, 1000 have a false positive scan detecting some kind of abnormality that needs further testing. CT scans, endoscopies, blood tests, biopsies and possibly surgery. Not only does this put 1000 people (1 in 20 of your starting population) through a huge amount of stress, but all of these processes have their own risks. CT scans increase the risk of cancer. Colonoscopy has around 1% chance of causing bowel perforation, which can cause a fatal infection. Prostate cancer biopsies have around a 2% chance of causing sepsis, which can result in death. Surgery that goes wrong can be fatal. So of your 1000 people, it’s realistic that 20 of them come to some kind of actual harm because of your screening programme. And these are people who were healthy before they started this.

Now lets take your 10 people with an undetected cancer. Your MRI is 90% sensitive, so one of them gets missed. That person may later get symptoms of cancer, but they feel reassured that they had a clear MRI only months ago so they don’t go to their doctor as quickly as they would have done otherwise. The other 9 cancers get detected, so at this stage you’re almost even - you have harmed 10 people, falsely reassured one person and potentially helped 9.

But how many of those patients actually have a reduced chance of dying? DCIS (ductal carcinoma in situ) is a pre-cancerous lesion of the breast which is around 20% of lesions detected on mammography, and which only develops into cancer in 60% of cases. But most women diagnosed with DCIS undergo surgery and chemotherapy. So if two of your 9 are women with DCIS, they are exposed to the risks of surgery and chemo, but one of them would never have been unwell or known she had the DCIS without screening. Some cancers such as the form of lymphoma my dad has don’t actually have improved odds of recovery through early detection. Some of those patients might have presented with symptoms within a month or so without the screening programme. Some of them will die of their cancer no matter how much treatment they receive, so early detection won’t help all those 8 people.

Shmendrik didn’t say it was questionable whether CT scanning didn’t reduce lung cancer, he said it was questionable whether it reduces lung cancer mortality. So you can use CT to detect lung cancers earlier, but possibly the same number of people die of it anyway.

So if one of your 8 cancers is lung cancer, that person doesn’t necessarily have an improved chance of survival and you haven’t really helped them.
So your helpful screening programme has in fact harmed 21 people (I’m counting the woman who had unnecessary surgery as harmed), caused 1000 people weeks or months of terrible stress fearing they might have cancer, and maybe helped 7 people. And this is with MRI sensitivity and specificity that is better than currently achievable in the real world.

This is a very emotive argument. However if you’re going to actually be responsible for a screening programme you need to actually rigorously go through the numbers to make sure you are not going to cause more harm than good. More scans does not necessarily mean better as every scan carries a risk of harming the patient.

I couldn’t disagree more. I think the cost of healthcare is only going to keep rising faster and faster as more different and more expensive treatments become available. In many ways this is great, as more diseases become treatable, but the longer we live, the higher the costs are going to become.

Who’s reading all these scans? I’m not. I’ve got to go in on the weekend enough as it is.

On a more serious note, it isn’t just like looking under the hood of a car. You have to tailor the scan to the question being asked. A scan for the heart is not optimized to find breast cancer.

Widespread use of CT scans as a preventative (non-diagnostic) measure would probably cause more cancer than it would prevent. People get concerned about how much radiation you can get from a single chest X-ray, but a CT scan gives your body anywhere from 50 to 1000 times more radiation.

As for MRIs, as rekkah said, there is always the danger of false-positives causing more harm to patients due to unnecessary follow-up testing, emotional stress, harm from biopsies, and so on. Just look at all the debate over the “non-harmful” PSA blood tests used in prostate cancer screening, which really seem to be not helpful in screening and leads to overtreatment and inaccurate over-diagnosis, and this can cause harm to the patients.

Just was at a conference in which they cited a new study published in the NEJM about the use of low dose CT scanning of the lungs in asymptomatic people with a more than 30 pack year history of smoking, and they did find it helpful in reducing mortality in that patient population by detecting lung cancers early.

Here is the article:

No one has mentioned the Stark law which precludes physicians from self referral, which would be the case if they owned a MRI machine and kept referring their patients for MRIs on said machine.

While I was “skimming” posts, my sometimes foggy mind took this as “time shared medical devices” (like an artificial hip or oxygen generator) lol

That’s pretty interesting. I anticipate that these tests will become better and better until the issues of false positives and the resulting emotional financial and emotional stress will decrease to a point where they will become worth it. My main point is that even if the tests became really accurate, both in terms of high percentage of “true positives” and low percentage of false negatives, that wouldn’t necessarily mean that the price per scan would go down, as medical costs have increased faster than the rate of inflation even as medical knowledge, technology, and access to information have increased, in the last few decades. Since quantity demanded is inversely related to price, assuming scans are normal goods, the benefits of improving technology might not spread out very quickly throughout the general public. (It’s perhaps an example of what Bill Gates said about people in Africa who are still dying of diseases that can be treated; the reason why resources are not allocated by the market for this purpose is because the poor cannot afford to pay for them, so they are not “counted” as demand.)

I’m suggesting that the way to solve this problem is by what Gates called creative capitalism, the first steps being taken by social entrepreneurs, and being followed up by large scale funding by governments and philanthropists.

It’s my opinion that within the next decade, we will have the technology to make sure that people for whom money is not a concern will be able to greatly reduce their risk of dying of cancer, assuming that they take advantage of the best preventive and early-detection tools that money can buy. And if such tools can be taken advantage of by the very wealthy, it is a given that they can, at least in principle, benefit the middle class and poor as well, the economic problem (high prices) being the “only,” but obviously very daunting barrier.