Comment by yummyfajitas | Hacker News Reader

Comment by yummyfajitas | Hacker News Reader

Parent Full thread

Interestingly, we already discovered a mechanism for drastically reducing the cost of healthcare back in 1986. It's a way of crowdsourcing the problem called high copays. Basically, you have to pay out of pocket for 90% of your health care up to a (high) cap.

It turns out that patients are very good at figuring out which health care will improve health and which won't - the high copay group had no statistically significant difference in health from the low copay group, and spent about 30% less money. What a crazy magic bullet, huh?

http://www.rand.org/health/projects/hie.html

We ran a directionally similar experiment in 2008, and got much the same result: low copayment causes people to consume a lot more medicine, but with no objectively measurable improvement in health. (Subjectively, people with insurance feel healthier even if they never go to the doctor.)

https://www.nber.org/oregon/

In both cases we ignored the result because we don't like it.

A one-sided presentation of the conclusions. From the RAND study:

Cost sharing in general had no adverse effects on participant health, but there were exceptions: free care led to improvements in hypertension, dental health, vision, and selected serious symptoms. These improvements were concentrated among the sickest and poorest patients.

Personally speaking, I'm very much in favor of concentrating benefits among the sickest and poorest patients so as to minimize avoidable suffering. The pursuit of economic equilibrium on healthcare is a chimera because the conditions for perfect competition do not apply; patients cannot choose when to be sick, and thus lack the discretion on when they enter or exit the market; nor are competing therapies or competing providers of the same therapy offering a commodity product that can be freely selected between, just as if you suffer a catastrophic injury you won't get a timeout to make choices about which emergency room offers the best bang for the buck.

Like any sane person I want to minimize the overall cost of healthcare, but I also want to minimize avoidable suffering. I would much rather pay a bit more in tax and have a system that was somewhat less fiscally efficient but more accessible to those in the greatest need. This isn't just a matter of altruism or anxiety about high healthcare costs; if you have a maximally efficient healthcare delivery system then it's likely to fail in the event of a black swan public health crisis because all the gains from efficiency have already been booked for profit.

The number of positive results found by the RAND experiment is completely in line with what you'd expect from false positives from multiple comparisons, particularly given that these results were all found in non-preregistered segments.

nor are competing therapies or competing providers of the same therapy offering a commodity product that can be freely selected between, just as if you suffer a catastrophic injury you won't get a timeout to make choices about which emergency room offers the best bang for the buck.

The vast majority of medicine is not emergency medicine. This idea that patients are unable to price shop and compare providers is completely debunked. India's free market health system represents 1/6 of the world and Indians do this every day.

Like any sane person I want to minimize the overall cost of healthcare, but I also want to minimize avoidable suffering.

As RAND and Oregon show, free and reduced cost healthcare do not minimize avoidable suffering (at least suffering caused by medical conditions) any better than high marginal cost plans.

This isn't just a matter of altruism or anxiety about high healthcare costs; if you have a maximally efficient healthcare delivery system then it's likely to fail in the event of a black swan public health crisis because all the gains from efficiency have already been booked for profit.

Can you provide an argument why a more efficient medical system - specifically one where patients don't visit the doctor for frivolous/moral hazard reasons - would be somehow susceptible to black swan events?

The vast majority of emergency medicine is not emergency medicine, but emergencies are among the most likely things to have a catastrophic impact at the individual level. I cannot possibly treat the entire complexity of healthcare in a single comment and neither can you. That's why I'm offering a limited specific counterexample to your general argument, to point out the existence of fundamental economic problems in healthcare that are often buried by economic jargon which misrepresents the behavior of the healthcare market.

As RAND and Oregon show, free and reduced cost healthcare do not minimize avoidable suffering (at least suffering caused by medical conditions) any better than high marginal cost plans.

Unless you're poor or chronically sick. You have made it abundantly clear time and again on HN that you do not believe in altruism but think that all economic decisions should be made on the basis of self-interest; I think that approach is bunk and so we are not going to agree on a common goal any time soon.

Can you provide an argument why a more efficient medical system - specifically one where patients don't visit the doctor for frivolous/moral hazard reasons - would be somehow susceptible to black swan events?

If you're maximally efficient there's nothing to cut when the shit hits the fan. I think critical systems should include some excess carrying capacity (ie waste) because it's faster to reallocate resources internally than to get additional resources under time pressure.

As a simple example (from which I am not attempting to generalize to the secotr as a whole and which I am not going to get into any nitpicky arguments about), there's a good argument that we have a bit of a doctor shortage these days: https://www.quora.com/Why-dont-we-train-more-doctors

It would be rather inefficient to have a public health system with too many doctors and nurse practitioners. They wouldn't be as busy and wouldn't make as much money as they could at equilibrium, and you can easily imagine various second-order effects of such an inefficiency, and the fact of a cumulative social cost.

But if there is a public-health crisis, would you rather meet it with too many doctors or too few?

The vast majority of emergency medicine is not emergency medicine, but emergencies are among the most likely things to have a catastrophic impact at the individual level.

And this is handled by the high copay, high cap model proposed by RAND, Singapore and myself. You still have not identified any "fundamental economic problems" or misrepresentations of the behavior of the healthcare market.

Unless you're poor or chronically sick.

False - the group studied in the Oregon experiment was poor. Oregon explicitly studied a medicaid expansion. RAND included poor folks too - in fact, one conclusion you can data mine from RAND (if you are so inclined) is that low cost medical care increased certain health problems for poor folks.

(Of course, that result is just as shaky as all the other data mined results.)

You have made it abundantly clear time and again on HN that you do not believe in altruism...

When did I make this clear? Are you confusing my opposition to harmful or wasteful altruism as opposition to all altruism? You are confusing my criticism of bad implementations with opposition to solving the problem.

Concretely: I have no major objection to effective altruism - perhaps something like Singapore's model. I object to wasteful feel-good measures that don't have any measurable benefits, like Obamacare's medicaid expansion.

As for things like maintaining an excess supply of medical pros to handle crises, if that's necessary I don't think it should be provided via random inefficiency. It should be explicitly planned in.

I just can't be bothered to go around this today since we're not going to agree anyway. On altruism, I was thinking of a specific comment you made a while back about being rationally focused on your interests, reproductive ability and descendants (qua biology rather than social attitude, but I think it's consistent with your expressed libertarian viewpoint. I don't remember enough to successfully look it up right now.

I think this encapsulates a key difference in our worldview:

As for things like maintaining an excess supply of medical pros to handle crises, if that's necessary I don't think it should be provided via random inefficiency. It should be explicitly planned in.

I used to think that, but now I favor a little general random inefficiency because we don't know what we don't know. My inclinations are utilitarian, but sadly my ability to predict the future is limited so I systematically distrust my own utility calculus.

I'm pretty sure I know the comment you are referring to: https://news.ycombinator.com/item?id=10180416

Altruism in that comment actually referred to the willingness of an individual to sacrifice personal reward for the cause of white power. I said that I'm not altruistic and am happy to make choices that harm white power in return for personal gain.

Also I'm not a libertarian, I'm closer to neoreactionary than to any other identifiable ideology.

You would predict that discouraging healthcare "consumption" through higher co-pays would cut down on office visits for minor problems, but not for major issues, because with major health issues you have no choice not to go.

How much is an office visit for runny nose? USD 50, USD 100? And how many frequent flyers are thee in the population, one out of ten, perhaps? It's just not plausible that it's the doctor-fetishists that cause US healthcare to be 4 times as expensive as the rest of the developed world.

I explicitly said a 30% decrease in costs, not a 75% decrease.

No, all that plan does is make people not go to the doctor when sick. They put stuff off until it is far too late, and they end up needing more expensive care that's much riskier.

If your narrative were true, medical consumption would go up and health would get worse. Neither of these happened.

Turns out medical consumption is well beyond the point of diminishing returns, and people are quite capable of determining what medical care is necessary.

Almost smells like the ACA, doesn't it?

Edit: My apologies for sounding a bit antagonistic. The ACA is currently making use of high copays for that reason and more. It also hopes to force more transparency as well as most people want more than a icd-10 code when they have to write a check for $4k.

Healthcare discussions can quickly devolve into flame wars. Please take extra care to comment civilly and substantially.

That's a valid point. I've updated it to be less flippant. Appreciate you bringing that to my attention as there is a lot of great discussion going on in this thread.

There are a wide variety of routine (STD tests, birth control pills) and completely predictable (pregnancy) services where the ACA explicitly prohibits copays.

I've been told Singapore's forced savings + cat coverage plans are reasonably close to this, however. Hardly surprising given how well Singapore tends to govern.

But your broader point (or perhaps my steelman of it) that high copays are not incompatible with socialized medicine is absolutely correct.

Reply on news.ycombinator.com