Hacker News Reader

Comment by tompccs | Hacker News Reader

The answer to the headline is astoundingly obvious yet the article doesn't mention it once. The reason is regulation: you can't use a medical device that you've hacked together unless it has been cleared by the relevant authorities as being safe. Sometimes it's a matter of simply getting a cheaper alternative through regulatory approval, but other times its just not worth the cost to prove a cheaper alternative is as safe as the more expensive one. Very annoying article about an interesting subject.

View on HN

The agonizing part is that having incredibly expensive medical tools isn't safe either - it leads patients to avoid medical treatment because of the cost.

It's not meant to improve patient outcomes, it's meant to avoid liability.

I wish people didn't phrase things this way. There's no single "meaning" to all of this. It's a bunch of different people with different priorities making different cost-benefit tradeoffs.

I mean, what would you do if a loved one died at least in part because someone was trying to save $1,000 on their treatment? How would you feel? What if you were permanently disabled or put in months of agonizing pain, and there were more expensive options that would have had a lower chance of doing that? Or just might have had a lower chance of doing that?

From the 10 miles up view, it's easy to say, "Holy shit, the US has such a high health care cost, we could get 97% of the same care at 50% of the cost," and that seems like a great idea. When you're personally looking at tragedy to save an amount of money that is not that significant for your specific medical case (the waste comes when it's applied to 50 more people who don't really benefit from it), the calculus looks very different.

It may still be the right decision. But can we be compassionate about the victims of that decision?

Then layer in that it is not in fact uncontroversially the case that this is what drives the costs of the US healthcare system. And layer in that there is unavoidable uncertainty about when a higher-cost procedure or device or medicine is really higher quality and when it's not, because we have well-founded limitations on how many experiments we can do on people.

I don't think it serves anyone to take such a moralizing tone. Liability exists for a reason. These cost-benefit decisions are tough ones. Information is not perfect.

97% of the same care at 50% of the cost could be rolled into further R&D which improves outcome faster in the long run.

Maybe! I mean, that's definitely one way it could go.

Or maybe the extra money wouldn't get spent on medical R&D. Or maybe more money spent on medical R&D is reaching a cost/benefit asymptote where there's not much further gain. Or maybe a lot of things.

And it's still cold comfort to be told that your or a loved one's inferior care today may (or may not!) be of benefit to a stranger 10 years from now.

I want to be clear: I'm not arguing for the current status quo. I agree it has big downsides. I'm asking for people to acknowledge that there are hard problems here that aren't trivially fixed, and even to the extent that some problems are trivially fixed, they still produce some downsides.

> what would you do if a loved one died at least in part because someone was trying to save $1,000 on their treatment?

Accept that said loved one wasn't the type to spend $1,000 on something they didn't think was necessary, made the wrong decision, and lost. Same thing as if they were driving around in an old poor-crash-safety car and got T-boned.

But embedded in your comment is an assumption that this $1,000 wasn't being paid by the patient, rather by some ambient health insurance company looking out solely for their bottom line. This double principle agent problem is a large part of the train wreck! A market cannot function without pricing information. So either we need to vastly simplify how cost sharing works and change the culture so doctors supply prices alongside treatment options, or we need to stop clinging onto the fallacy of having a functioning market.

Here's an example of the principle agent problem at work. Right now I'm actually trying to price an expensive medication. The name-brand costs 50% more than the generic. Due to the way (privatized) Medicare copays work, the name brand is significantly less out of pocket for the first several months! After a whole year, the name-brand works out to be about the same. So to save money, I will be going with the name brand and driving up the general plan costs.

Most of the time those are the same things.

The problem is ethical: How many hours of peoples lives must I improve for every person I kill?

If for example, you had a guaranteed treatment for obesity, but it killed 1 out of every 100 people it was used on, is it worth it? More than a few morbidly obese people would take those odds.

I was at CES many years ago, and a doctor had developed an asthma monitor for his daughter. This monitor would alert him and the school nurse immediately, and possibly 911 immediately if the episode was bad enough.

He couldn't get anybody to touch it because, at some point, it will miss an asthma event or the network will fail to deliver the event. And that child will die. And whoever sold the device will lose a multi-million dollar lawsuit.

As soon as some company with really deep pockets figures out that "indemnification" would allow them to generate scores of medical devices that nobody else can take on, we're going to witness an instant monopoly.

You mischaracterize the article. It mentions the FDA, and perhaps more importantly other disincentives.

Hospitals desire more expensive solutions because they get paid more for them. That is a second, enormous barrier.

The incentives are clear here. The costs are high, between regulations, marketing, out-maneuvering rent-seeking incumbents, and cultural issues. The reward is low. No one will become a billionaire for doing this work.

Yep.

And then you have to breach a very ugly topic most people don't like to discuss.

When does regulation kill more people by not having the $thing versus having the $thing? And how many people would suffer complication/dying with the $thing?

It's a very delicate balance, especially when you consider capitalist forces that encourage "move fast and break ~things~ people". The FDA is the chopping block for that... But also FDA overaction also kills people.

Right. The whole FDA system is set on the premise that inaction can cause no harm. This causes a lot of harm.

For every suction pump and super glue getting held up by regulation, there is also a witch doctor or snake oil. People love to talk about the clever hacks in some poor country, but not the high cost in QALY of superstitious or scam practices.

Since no one is clamoring to wholseale trade USA healthcare for India's, there are obviously done tradeoffs.

"Though FDA can trace its origins back to the creation of the Agricultural Division in the Patent Office in 1848, its origins as a federal consumer protection agency began with the passage of the 1906 Pure Food and Drugs Act. This law was the culmination of about 100 bills over a quarter-century that aimed to rein in long-standing, serious abuses in the consumer product marketplace."

I have no personal doubt that the FDA has saved millions of lives in the US and internationally, and the balance of harm/good they have caused would would make the harm side look infinitesimally small.

Also... I be interested in what cases you know that the FDA caused the deaths of people by overaction. There are people out there with axes to grind who thought they were going to get rich with some drug or device, who couldn't subsequently prove the safety and/or efficacy of their product.

The FDA's drug approval criteria are extremely strict. Ask anyone who works in pharma and they can probably name at least a couple of drugs which showed great promise but got axed by the FDA on dubious grounds.

A classic example is Aspirin. We know from over a hundred years of experience that Aspirin is a widely useful and relatively safe drug, but if it were discovered today the FDA would never allow it to market, let alone approve it for OTC sales. Aspirin's mechanism of action is far too broad and the gastrointestinal side effects alone would be considered grounds for rejection.

Perhaps if you dropped a few more specific examples would could talk about their issues instead of in generalities, instead of Aspirin? I was going to reply in my own words, but will instead yield to this retired FDA compliance officer: https://www.quora.com/Would-aspirin-be-approved-by-the-FDA-t...

So, the food regulation side of the FDA's activities have certainly saved millions of lives and probably swamp what the FDA does in terms of drug regulations. But generally new drugs tend to be rolled out gradually, doctors hear about disasters, and I'd be surprised if the FDA saved more than a thousand lives a year at most. On the other hand a bad call in disallowing new medicines can have very bad consequences. The FDA only allowed the use of beta blockers in the US about a decade after they came into use in Europe which probably cost something on the order of 100,000 excess deaths. Those early beta blockers did have serious problems and the FDA's decision also probably saved 1,000 people from liver failure after taking those drugs but I don't think that justified withholding the category until a liver-safe version was made. These days the FDA is less strict than it was in the 1960s and has certainly improved. But we can still do better to move more in the direction of European style regulation where they have a dozen competing EpiPen equivalents.

You had to go back 60 years. You didn't even mention Fast Track. You are ignoring device, test, and implant categories.

I didn't mentioned Fast Track specifically but it was one of the things I was thinking of when I said the FDA had "certainly improved." A large part of the credit for that should go to AIDS activists but Reaganite Republicans also deserve a lot of credit.

I'm not sure the FDA of the 1960s/70s was a net positive in its drug regulating role on human health. I'm almost certain the FDA of the 2010s is a good thing on net but I'm pretty sure it's bad on the margin, compared to the EMA which I think gets things more or less right.

I'm not particularly conversant in how FDA regulatory behavior is different across product categories but it sure looks like it's the FDA's role as a device regulator that causes the differences in the market for epinephrine injectors in the US versus EU.

What are your thoughts on this article? (which also cites a bunch of other articles on the first line)

https://slatestarcodex.com/2016/08/29/reverse-voxsplaining-d...

This can only be known after the fact.

To complain about "this obvious treatment that obviously works but regulations delayed for 10 years", is as valid as to complain about "this obviously flawed treatment that kills one person out of ten people and killed my husband, because obviously not enough testing was done".

The error in both complains is to assume that the answer was in anyway obvious. It was not. It almost never is.

I agree with you, totally. I just wanted to emphasize that point.

"Other time its just not worth the cost to prove a cheaper alternative is as safe as the more expensive one"

Sounds like a fatal flaw in the system which benefits manufacturers and insurance companies at the expense of consumers.

How do "more expensive" options benefit the insurance companies ?

Oh, they do.

https://www.npr.org/sections/health-shots/2018/05/25/6136857...

What about large companies that underwrite their own health insurance?

Not if you manufacture the cheaper alternative.

Reply on news.ycombinator.com