Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
How to reduce healthcare costs and improve quality (histalk2.com)
12 points by pseale on Dec 2, 2009 | hide | past | favorite | 47 comments


I like almost all of those, but it fails to address issue #1 which is the employer-funded health care system (which itself is an artifact of salary restrictions that were part of WWII rationing). Right now most people get what appears to them to be unlimited health care at a fixed price and thus have no incentive to strive for cost-effectiveness.

Imagine your employer said "We made a deal with Safeway. We'll take $30/mo out of your paycheck, and you can go to Safeway and all of your groceries are free." You'd be eating Filet Mignon washed down with a bottle of '82 Margaux every night.

That's effectively the way our health care system works. It won't become truly efficient until the people consuming the service are paying for it too.


Single payer systems don't just write blank checks. The real issue with US healthcare is it's become a game where the players spend money to lower their percentage of the healthcare burden without lowering the total costs. The net effect of which is the average person pays a lot more for zero average gain. With a small number of people get a free ride and everyone doing a lot of extra paperwork.


I don't think this is the #1 issue with health care, Matt.

I think the #1 issue with health care is that virtually all market health care in the US is provided through insurance companies who collude with providers to set prices.

It's hard to convince me that our big problem is that people are paying for unnecessary care when we have no idea what care costs. Per-service cost numbers are arrived at via allocation schemes based on bulk purchase agreements between hospital chains and giant insurers.

Unnecessary care certainly is an issue, but how do you rank it when you don't know what an MRI, a hospital bed, an ambulance ride, or an appendectomy should actually cost?


A deductible (present in most systems I've used) minimizes this buffet style spending.


Somehow, I think there's a difference between eating filet mignon and being treated for cancer. I could eat a good steak every day, but I wouldn't want chemotherapy unless I was pretty sure I had cancer in the first place.

Most people are just trying to get the treatment necessary to keep them at a high level of function. Why should anyone be denied that? Almost no one goes into a hospital and demands thousands of dollars worth of unnecessary tests for diseases they couldn't possibly have-- and those who do are often mentally ill hypochondriacs, given inappropriate treatment because (surprise!) health insurers are generally shitty about MH issues.


There is actually not much of a difference. Filet Mignon makes you marginally happier, and cancer care (mostly) makes you live only marginally longer.

http://www.washingtonpost.com/wp-dyn/content/article/2009/11...

Incidentally, last time we did a study to measure it, we found that at least 30% of health care consumed by patients is unnecessary.

http://en.wikipedia.org/wiki/RAND_Health_Insurance_Experimen...

The conclusion of the study is that health care you are unwilling to pay for out of pocket doesn't help more than it harms (on average).

A specific example of care people don't need but will get: mammograms. They are pointless (benefits don't outweigh harms) for people under 50. I mentioned this to a few women I know, but all said they intend to continue getting them.

[wow, down to -1 in 20 minutes from citing an expert opinion and an experiment. I should go back to reddit.]


Any time you want to talk about cancer, you have to acknowledge the fact that it's actually over a hundred diseases. Cancer treatment is phenomenally effective for some cancers, for some classes of patients, and ineffective for others.

I can name eight people, off the top of my head, I've known who've had cancer. (I'm sure I could list more.) Three died, four are alive and well, and one was diagnosed last month. We're talking about a disease that would have killed all 8 a few decades ago.

Also, even if the improvements in cancer treatment are only interative (8 months instead of 3) that's still a major gain, from a research perspective. Look at HIV/AIDS; in 1985, it was an incurable and fatal disease. Due to a series iterative improvements, someone who contracts HIV today in the developed world will very likely be alive in 15 years, and by then, there may be a cure.


You guys can have a big long emotional conversation about cancer care, but cancer isn't where all the cost is in the system, and cancer treatment may not even be where all the cost is in cancer care (tests, many of which have been demonstrated to be ineffective, may be the major component instead).


People absolutely go to the hospital to seek treatment thy don't need. We see it in the US already - people flooding emergency rooms with the sniffles looking for antibiotics. That's entirely the reason triage exists, to seperate out the people looking for clinic care and the people with real emergencies.

You are correct about cancer and major surgeries - almost no one is going to elect to have these if they can avoid it. The only exception is obeseity cases where people would rather have a gastric bypass then have to suffer through diet and exercise.


Much of the ER flooding comes from two sources:

1. Uninsured people who go to the ER because they legally can't be turned away there.

2. Panic attacks. Untreated anxiety/panic disorders are a huge burden on the hospitals. Most repeat panic visitors have never had a specialist (a) tell them what they're experiencing, and that it's non-fatal (and that there's absolutely no benefit in going to the ER for a panic attack) and (b) prescribe them medicines that can actually help them. Any guess as to why this is the case? Because those people are uninsured.


The nationwide direct costs of arthritis are within 90% of the direct costs of cancer. Matt wasn't saying "withhold chemo".


You completely and totally missed the point.


In that case, so did I.

You compared free high-end foods to free healthcare, which frankly just seems bizarre.

Nobody wants to use healthcare. Even when I lived in the UK, and I could have visited the doctor for free every single day if I had chosen to, I did not. Neither did I ever hear of anyone ever doing that or anything close to it.

Having lived in both the US and the UK, I haven't changed my habits. I visit the doctor when I feel I really need to. The only difference is that in the US, an insurance bureaucrat then sends me a letter telling me they're not paying up because <insert random excuse here>.


And in the US, you get the privilege of paying for the paper on which the denial letter was printed... and, to make it better, the bureaucrat's salary.

If you buy private health insurance, you're literally paying the salary of the asshole who pores over people's medical records looking for reasons to deny care (e.g. undisclosed insomnia prescription from 1982 as a "pre-existing condition").


Just to play devil's advocate, here's a mind bender for you: we should be grateful to that asshole.

The insurance company is paying his salary because it costs them less than would the care that he manages to deny. If he didn't exist, the insurance company's costs would increase. This means that insurance premiums would have to increase. Whether you buy your insurance directly, or it's deducted from your paycheck, or your employer pays for it in full, that increased cost of insurance premiums would trickle down to you in the end.


Just to play devil's advocate, here's a mind bender for you: we should be grateful to that asshole.

Until you or I get cancer, that is.

Last I checked, the probability of you and I having at least one life-threatening health problem in the next 100 years was pretty close to 1.0.

What's actually happening, regarding that asshole, is that you're paying 95% premiums for 85% care. It gets worse than that, because the 15% being shaved off is at the high-cost end, which is precisely where you need your insurance.


You made an inappropriate analogy.

There are no inherent costs to the consumer in eating fine foods.

There are enormous inherent costs to medical care. Most patients want to be treated with minimal pain, risk, and time lost, so they can get back to their regular lives. Their incentive is already, in virtually all cases, in the cost-minimizing direction.


It's not inappropriate. The point is that when people perceive something as free, they intrinsically opt for the most expensive option. They also will get costly tests and procedures that have at best an extremely remote chance of being useful. The whole reason tort reform, which the original author did mention, is needed is because doctors know this and then prescribe expensive, unneeded tests simply to cover their asses. Between that perverse incentive (which he did address) and the patient's lack of fiscal accountability, enough people have an MRI done every time they sneeze to have a huge impact on health care prices.

When insurance companies instituted even very small patient copays on non-generics, usage rates of generics skyrocketed. Just having to pay $5 more (for a product that actually costs hundreds more, but is chemically identical) aligned people with their economic incentives and caused a massive shift in consumer behavior.

Brand name pharma companies fought back by giving vouchers to doctors to pass out to patients that covered their co-pays. Brand name medicine market share went right back up. Insurance companies have no effective way to combat vouchers and are forced instead to raise health care rates.

Also you're just plain wrong because you're applying rational thought to people who are in a panicked state. When someone has a health problem, they'll do anything and everything to fix it. We all value our own lives as effectively worth infinite dollars, which makes any treatment rational from our perspective. Society, on the other hand, cannot afford to be so generous.


This has been shown empirically to be false. Patients will consume medicine with no health benefits if it is free, but not if they must pay for it.

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


You did not miss the point.


1 idea to reduce healthcare costs and improve quality:

* Institute a national, single-payer health insurance program.


I think he's saying the same thing, but his proposed program is a high-deductable+HSA plan.


This will help how? There are few economies of scale to be gained by going national [1], and single-payer gives consumers no incentive to consume less.

[1] You save a bit of money by having the IRS collect payments, but the billing department is not a major component of healthcare costs.


It seems to be working rather well in a large number of other first-world countries (I'm speaking from experience - I'm British, and my wife is half-French and grew up in France. Now we live in the US where we enjoy just about everything except for the healthcare system).

"single-payer gives consumers no incentive to consume less"

Having been fortunate enough to live in a country with a single-payer system, I never saw the kind of over-consumption you're hinting at.


I live in Canada, and have to laugh at free market fundamentalists who claim that single payer monopsonies encourage over-consumption. When it's pointed out that Canadians actually see doctors less frequently than Americans, they turn around and conclude that Canadians can't get access to doctors.


Mouth-breathing teabagger reasoning: It's soshullism, therefore it must suck. Also, Canada's cold and Europe can't decide on a common language, so no point taking those people seriously.

I also find it odd that these people can focus on the comparably miniscule problem with the Canadian and British systems, while completely ignoring the glaring faults of ours.


I didn't say it couldn't work. I said that there are minimal (if any) benefits to a national health care system, as opposed to city, state or private systems. The economies of scale just aren't there.

As for overconsumption, the RAND experiment (as well as plenty of other statistical studies) shows that consumers will consume up to 40% more health care than they need if it is free (but not if they must pay out of pocket for it). Most likely your country either prevents this via other rationing mechanisms (which are politically infeasible in the US) or else you observed it but were unaware of it.


I have seen numbers anywhere between 15 to 30% as the overhead / billing cost of healthcare in the USA. It depends on what you call overhead, but pre approval for a procedure is not free. Neither is 3 or 4 letters back and forth between the healthcare provider and the insurance company.

PS: Many small doctors’ offices have 1/3 of their staff devoted to dealing with insurance and billing issues.


A pre-approval may cost $10-20. An unnecessary procedure may cost $1000. If you cut admin costs by $100 at the expense of $1000 in unnecessary procedures, you will lower the percentage of overhead quite nicely. You will also be wasting $900.

Is there any reason to believe the (highly competitive) health insurance market is far from the optimum overhead/healthcare ratio? If you know better, why don't you build a startup to exploit this and get rich?


Insurance companies don't directly pay for the total costs of pre-approval so they optimize to minimize personal costs independent from the total cost to the system.

In other words doctor's offices don't bill them for the time a sectary spends filling out their forms. Which means doctors include that overhead cost as part of the cost of procedures. Also while each insurance company's internal system is streamlined each doctor's office needs to deal with several companies’ which increases their overhead costs even more.

It's like a large game of Prisoner's dilemma where several people have already defected. http://en.wikipedia.org/wiki/Prisoners_dilemma

PS: A doctor making 200k/year spending 5 minutes on a pre aproval cost's someone more than 10$. The patent needs to show up which costs them even more money etc.


How large is the cost of a secretary filling out a form compared to an MRI? You are talking about adding a few dollars to the cost of an a procedure priced at $400 or more (I don't need preapproval below $400).

A doctor only needs to get involved in the case that a preapproval is not granted, i.e. when the procedure appears unnecessary.


It costs the fully loaded headcount cost of all admin staff at the doctor's office divided by the number of transactions the office undertakes in a year. It's still probably not that high, but it's not as simple as saying "filling out a form only costs $1.20 in billable time".


FYI, the actual cost of the average MRI is ~200$. Everything else is just padding.

Let's take an example of how these low costs work. In Denver, where I live, if you get an MRI of your neck region it's $1,200, and the doctor we visited in Japan says he gets $98 for an MRI. So how do you do that?

http://www.pbs.org/wgbh/pages/frontline/sickaroundtheworld/i...

Anyway, the idea it's just one person filling out a form is a mistake. It requires time from the patent, a doctor, the doctors admin staff, and the insurance company's staff.


Uh, isn't this --- respectfully --- ass backwards? The benefit of going national is that you greatly expand the risk pool. All I know about insurance I got from listening to Robert Schiller's financial markets class on iTunes, but, I'm pretty sure that's how insurance is supposed to work.


Work out the numbers and you'll see why it isn't. The variance in health care payouts over a pool of N people is proportional to sqrt(N), whereas the mean is proportional to N (see the central limit theorem).

For a family of N=1 to N=4, sqrt(N) is of the same order of magnitude as N. This is bad, since you can't reasonably predict how much you need to pay. This is why insurance makes sense for me (N=1) or a family of N=4.

Once N=10,000, sqrt(N) = 100, so the variance is only 1/100 of the mean. At N=100,000, it's only 3/1000 of the mean. Every factor of 10 increase in the insured pool lowers the standard deviation by a factor of 0.316=1/sqrt(10).


That's a good point.

But you've ignored adverse selection. Many of the people commenting here have no insurance coverage, because they're 22 and it is irrational for them to pay for coverage they're not mandated to have.

I see that if people were i.i.d., the difference between BCBS's risk pool and Medicare's risk pool would be irrelevant. But they aren't.


I don't need people to be i.i.d. Statistical independence along with some minor technical conditions are sufficient.

http://en.wikipedia.org/wiki/Lyapunov_condition

http://en.wikipedia.org/wiki/Lindeberg%27s_condition

Lumping in seniors with 22 year olds does not reduce the variance any more than lumping in 22 year olds with more 22 year olds does.

Statistical variance is not a major contributor to the cost of health insurance. You are simply wrong on this point.

Additionally, adverse selection has nothing to do with a variable risk pool. Adverse selection is a problem caused by asymmetrical information: if you know you are sick, you might lie to your insurance company about preexisting conditions to gain insurance. Adverse selection also has nothing to do with a variable risk pool.


You sure are better with math than I am. But you're not helping me understand how, given the fact that the majority of the uninsured in the US are below the age of 35, and the fact that people under the age of 35 have less reason to pay into the system than people over 35, and the fact that people under the age of 35 are similarly much less likely to incur costs, mandated guaranteed issue or government single-payer doesn't improve the risk pool.

You've clearly demonstrated that the problem with the current insurance scheme isn't a trivial stats 101 observation, but I feel like you're dodging the point because I've given you an excuse to do so by making that point clumsily.


I'm not sure what point you were making. You originally suggested that there were economies of scale due to the reduced variance of a national risk pool, which is true but so minor as to be irrelevant.

Regarding the lower average cost of people under 35, this is true. However, this is not an economy of scale - the cost to insure 1 million people aged 20-35 is very close to 100 times the cost to insure 10,000 people aged 20-35. No economy of scale there.

Adding cheaper people to the market will lower the average cost of the broader health insurance market. Millions of new netbook buyers lower the cost of laptops, since the average price of 1 million netbooks and 1 million macbooks is lower than the average price of 1 million macbooks. This isn't an economy of scale, it's simply a change in market composition. A change in market composition (in vacuo) doesn't lower costs to any individual. Millions of new netbook buyers didn't make my macbook cost any less.

It will also raise prices in any individual market segment due to increased demand: if demand for netbooks/health insurance for 20-35 year olds goes up, prices will too.


One second-order benefit of universal healthcare is that countries with such plans view health problems as a social cost, so there's more regulation of food production, industrial practices, et cetera. Consequently, Western Europe is a safer and environmentally cleaner place, in many ways, to live.


These 4 would lower costs the most:

----

Tort reform. Cap every state as has been done in California and Texas.

Eliminate doctors’ malpractice costs for patients who demand free care. If the government insists that ED docs see every patient (through EMTALA), they are de facto government employees for those patients and should receive government indemnification.

Give needles to addicts, along with access to treatment.

Strongly consider legalizing and decriminalizing drugs.

----

I would also add:

Make it legal to see anyone for healthcare. If I want to have my veterinarian cousin administer 5 stitches, then I should be allowed to.

Make it legal to OPTIONALLY sign away your right to sue.


If you make it legal to optionally sign away your right to sue, then everyone will lose their right to sue, except for the few adverse venues of last resort (public-funded hospital emergency rooms, perhaps) that won't be able to decline services.

The same thing happens with the "let people shop interstate for insurance" idea. If you look just a couple moves ahead into that game, you see that that proposal is just cover for "adopt the lowest common denominator of insurance regulation nationwide", because every rational insurance company will simply move to Alaska or Wyoming or wherever regulations are least onerous.

The exact same thing happens in banking, which is why Charlotte is a US banking hub.


The probability of the event happening * the loss = the premium of the surgery. That is it. If you would rather pay 1000 dollars for a surgery than 3000 then that is your business. I certainly wouldn't sign away my right to sue when I'm under back surgery.


> Tort reform. Cap every state as has been done in California and Texas.

Most studies show this claim to be specious at best. Tort reform has almost no impact on quality of care or cost of that care. The only real impact of tort reform in the realm of medical malpractice is to transfer money from the pockets of wealthy litigators to the pockets of wealthy insurance executives.


Socialize the judicial portion with flat rates for doctor/hospital boo-boos and a government panel to oversee and "punish". 3 strikes and you loose your license, back to school. No lawyers, reduced insurance, speeder payment processes. = Reduced rates for us.


Some good ideas, but mostly revolting. He's basically trying to preserve a system where the wealthy get better treatment than the rest. (See items #3, 9, 14, and 15.)

Realize that screening doesn’t save money for society.

Um, link? Although current screening practices may be overly aggressive-- I have no idea-- this is different from saying that screening is an unconditional money-loser, which is what the OP strongly implies.

Notice this, as well: Dan Field is a physician with The Permanente Medical Group.

So, at least he discloses the pocket he's in.


Yeah, the bit about name-brand drugs being only for those who an afford them out of pocket is especially bad. Drug patents are 20 years, so the peasants would be getting 20-year-old cancer treatments, while the select few would be getting actual modern medicine. Besides the ethics, though, what would the cost of modern medicines be for those few thousand who can actually afford them? Today, plenty of middle-class people get modern drugs because their insurance pays for it, so the cost of research, clinical trials, manufacturing, etc is distributed among a large population. What would the price of a pill be if only the richest of the rich could afford treatment?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: