I'm on Medicare/Medi-Cal now, but when I was working, I had $125 in co-pays for monthly medications and often had to put it on a credit card. That's another result -- higher credit card debt, just to keep up with the co-pays and that becomes a vicious cycle.

If co-pays are good, shouldn't the amount of the co-pay be pegged to the income of the patient?

Three hours of pay, or something like that?

La la la, socialized medicine, all going to die, etc.

Co-pay has basically no effect on whether I go to the doc. The pain-in-the-a** factor of making an appointment, leaving work, sitting around in an office full of sick people is sufficient deterrent to my going to the doctor "frivolously".

Thanks for this post.

Sounds like RAND might've ignored "survivor bias".

I find Ezra's arguments unpersuasive because common sense and basic economics would tell you co-pays reduce doctor visits, but people would stay just as healthy. After all, instead of giving money to doctors for higher co-pays, people can use it to stay more fit by getting a membership in a fitness club or buying a thighmaster.
This would certainly be a better result because people should be healthy by working out instead of taking pills. The logical consequence of this reasoning would be that making people pay their bill entirely through co-pays is the best policy of all.

What is totemic is the fetishizing of markets.

I'm with Thalarctos on this. For most American's, especially those with lower incomes, a trip to the dr's office is no visit to Marcus Welby land.

Even when I had pretty good insurance, I found myself dealing with cheerful and very competent doctors who were very over-booked. A quick look at my charts, a few questions, a quick look at me and out the door all in less than 15 min. Of course I had spent 20 min in the waiting room nd another 10-15 min in the exam room waiting for the doc to show (with 1 or 2 interfaces with an equally hurried nurse).

Conditions that never encouraged me to visit the dr's except when very ill.

Kevin -- Please don't foster the notion that health care research is too expensive. Not doing research is far more expensive.

This is a matter of priorities in a society that pays enormous sums for military equipment that is then too expensive to actually use in battle or not appropriate for the battles we are fighting.

I know the "moral hazard" argument on co-pays. But I'm with Keith G. I live in Tucson AZ where there aren't enough doctors (Massachusetts it ain't), it is really hard to get an appointment, you wait months, and are lucky if you get 5 minutes of actual M.D. time. Our co-pay is 10 bucks. Sometimes my husband's doctor forgets to collect it and his office sends a bill. I am morally (hazardously) certain that the co-pays (which of course are different for every patient) cost more to administer than they bring in. And in mine and Keith G's situation they are completely irrelevant. The idea that I would be able to just casually drop into a doctor's office is completely absurd (don't let me bore you with the time the triage nurse gave me an app't a month out for symptoms of a possible detached retina...)

Also worth baring in mind that with the decrease of manufacturing and other unionized jobs, many, many more American's are hourly workers with no sick leave.

A visit to a doc. means a shift missed. That's probably $30-$50 gone from the next check with possibly little likelihood of making it up.

The plural of anecdote is not data. As much as we all like to talk about ourselves, our individual experiences with copays won't illuminate this discussion. I trust RAND will take an open review of this problem. Science has a good way of working out errors. Cheers to the skeptics everywhere in the scientific community who undermine sacred cows!

Abs Obs,

Certainly. Yet the plural of anecdote may illuminate the need for a new data set. The possibility of new conditions (above the potential flaw in the original study) may indicate the need or a fresh look.

If it turns out the RAND study was faulty, there's a whole lot of subsequent bloviating about healthcare that's going to turn out to be misguided too.

Misguided but not withdrawn. I hope you don't think this will cause anyone to publicly change their mind about policy based on new information.

The RAND study ignores the fact that the patient always has skin in the game. As I tell my doctor, he isn't going to die for me.

As for the argument that if people just excercised more all our problems would be solved. Back when I was running 1600 miles a year and my BMI was 23 I was already showing the first symptoms of metabolic syndrome that later developed into Type 2 diabetes.

Nobody's minds are going to be changed, because there's too much money at stake. Even if there turn out to be biases, poorly-reasoned opinion and suspect data will remain enshrined as orthodoxy. All hail the gods of the free market.

And the plural of anecdotes may not be data, but certainly countervailing theory supported by those anecdotes is a reason to question the original theory supported by the original data. The arguments that the direct cost of a copay is insignificant compared to lost wages and/or the hassle factor, and that nobody who is truly ill enough to go see a doctor ever let a copay stop them from going (are these free-market pundits ever ill?), are entirely plausible and consistent with general theories of economic behavior, and as such should be given weight in the discussion.

Has the state of liberal economic thinking actually decayed to the point where someone is required to defend a study saying that people will tend to use less of something if they have to pay for it themselves?

Has the state of conservative economic thinking actually decayed to the point where someone states without qualification that people will tend to use less of something if they have to pay for it themselves?

Has Herr Harry not heard of demand inelasticity?

Go back to your HS econ teacher and have this concept explained to you.

Certainly there are things other than out of pocket cost that impacts one’s decision to go to a physician.

My family was forced to switch to a high deductable plan. The family copay is $3000 before any insurance kicks in and then it is a while til the cap.. Surely all thoe who say deductables don't matter aren't talking about $3000. Knowing that a particular test will cost$1500 certain'y affects me even if I know the consequences could be deadly.

If demand elasticity applied to medical care to the degree you seem to think it does, then people would not stop getting the care they need just because costs were going up.

If making it free would not increase consumption, how can you justify the idea that paying for more of it would decrease consumption?

Everybody is talking about gasoline's "demand elasticity." But a lot of people are buying hybrids, too, same as they bought many more small cars after the 70s. Market forces work.

Everyone talks about medical care as though the only things Americans ever spend medical money on is absolutely necessary care. You might say everybody has a right to an appendix operation without having to worry about paying for it. But that's just the start of the universal care scenario. What length of hospital stay do they have a right to? Two days? A week? Single room? Who decides? Drugs name brand or generic? Follow-up care on some kind of schedule, or any time the nervous patient wants an appointment?

All the questions that an insurance company now fields are going to have to be fielded by somebody else. And at some point, the answer will have to be "no," just as it is for an insurance company. This discussion rarely gets to this point.

And under universal care, what's the surgeon getting paid? Enough to pay for the medical education? Who decides that? Or would the education have to be "free" too? Who decides who gets to go into medical school under those terms?

"We find that, with the exception of use by those who died (1 percent of those initial enrolled),there was no statistically significant difference in the rate of use between those who did and those who did not complete the Experiment." Newhouse, JP and the Insurance Experiment Group. 1993. Free for All? Lessons from the RAND Insurance Experiment. Harvard University Press: Cambridge, MA

It's not the expense of the research, it's the randomization into groups of insurance products. It would likely be hard to get that through a contemporary IRB. Plus, the issue of creating the insurance products.

Surely all thoe who say deductables don't matter aren't talking about $3000. Knowing that a particular test will cost$1500 certain'y affects me even if I know the consequences could be deadly.

And think about the weird incentives this situation produces. If you have a plan with a high deductible and turn out to really need (and can afford to have) an expensive test done, your incentive is to have the test done just after the calendar year begins, then have as many other speculative expensive tests done as you can, since you've already busted your deductible.

.... But that's just the start of the universal care scenario. What length of hospital stay do they have a right to...Who decides who gets to go into medical school... harry at 3:07 PM

tbrosz: All the questions that an insurance company now fields are going to have to be fielded by somebody else. And at some point, the answer will have to be "no," just as it is for an insurance company. This discussion rarely gets to this point.

The discussion, of course, frequently gets to this point. However, since it always goes on to conclude, "We know there will have to be treatment tradeoffs in UHC and we believe the system will still be vastly superior to the current 'you get the healthcare you can afford, and if you can't afford it, fuck you' system"--not the answer you were hoping for--you continually pretend we haven't discussed it at all.

Let's just skip ahead and take it as a given that you'll say the same silly thing in two weeks, and four, and six....on and on to infinity. You can use the time you save to try and figure out what you're going to do about your family's gigantic medical bills should you lose your present coverage. Presumably you'll be fine with your family members sacrificing their health, functionality and perhaps even their lives to the god of your free market--because health care's a privilege, not a right, right?

There's an important part of the study that I think is being lost sight of. The trivial part is that people with big co-pays (they were really substantial in the study, not just a token $10) do use less medical care. As the conservatives note above, this was to be expected.

The radical finding was that, except for the poorest this had no impact on the state of their health. And this has been one strong argument for simply charging people more for their health care as a solution to the health care crisis. But it turns out to be this conclusion that is affected by the flaw in the study. If, as seems entirely possible, nearly 7% of the people in the co-pay groups pulled out when they needed expensive medical care, this conclusion is thrown seriously in doubt. They were among the most medically needy, and we do not know what care they would have chosen to pay for or how that choice would have affected their health. While the conclusion that high co-pays reduce the use of care may stand, the more interesting conclusion that this does not affect the outcome has become doubtful. Indeed, just as one would expect decreased use of care from increased cost, one would also expect that less care would mean worse health. And maybe that's so.

harry: Everyone talks about medical care as though the only things Americans ever spend medical money on is absolutely necessary care. You might say everybody has a right to an appendix operation without having to worry about paying for it. But that's just the start of the universal care scenario. What length of hospital stay do they have a right to? Two days? A week? Single room? Who decides? Drugs name brand or generic? Follow-up care on some kind of schedule, or any time the nervous patient wants an appointment?

You know, every other industrialized nation has solved these problems to the satisfaction of its citizenry. We probably can too, don't you think? Or are Americans just more stupid than everyone else?

If demand elasticity applied to medical care to the degree you seem to think it does, then people would not stop getting the care they need just because costs were going up.

While you're having your HS econ teacher explain elasticity to you, have him/her also explain budget constraints.

If demand elasticity applied to medical care to the degree you seem to think it does, then people would not stop getting the care they need just because costs were going up.

I don't think so. Unless demand is perfectly inelastic with respect to price, any increase in price is going to mean some people stop getting care.

If demand is highly inelastic, you'd expect, despite some people dropping of the rolls as prices went up, increases in the total expenditures as unit costs rise, which seems to match reality well.

If making it free would not increase consumption, how can you justify the idea that paying for more of it would decrease consumption?

Price elasticity of demand for real goods often is not constant, but is different (among other things) at different price points. Its quite possible for demand to be relatively elastic with respect to price below a certain point and relatively inelastic above that point.

Everybody is talking about gasoline's "demand elasticity." But a lot of people are buying hybrids, too, same as they bought many more small cars after the 70s. Market forces work.

Since every action (including government regulation) that affects economic behavior does so largely through "market forces", that is, while true, almost entirely vacuous.

Everyone talks about medical care as though the only things Americans ever spend medical money on is absolutely necessary care.

No, "everyone" does not. If you would like to make specific criticism of specific arguments rather than empty-headed false generalizations, that would be a welcome change.

All the questions that an insurance company now fields are going to have to be fielded by somebody else.

That's hardly something that proponents of universal systems ignore.

And at some point, the answer will have to be "no," just as it is for an insurance company.This discussion rarely gets to this point.

Actually, discussing how these decisions will be made is usually the bulk of any discussion of a universal care plan. Again, instead of making empty-headed generalizations, please try to deal with reality.

And under universal care, what's the surgeon getting paid?

Money.

Enough to pay for the medical education?

Almost certainly, just like in every developed country with some form of universal care.

Who decides that?

Ultimately, absent a draft of surgeons, the surgeons themselves.

Or would the education have to be "free" too?

It wouldn't necessarily be a bad idea, but its most irrelevant to the topic of how healthcare is financed.

Who decides who gets to go into medical school under those terms?

Uh, the medical schools, just like they do now. Ability to pay may keep someone from choosing to apply to medical school, or make them unable to attend even though offered a slot, but it won't generally be a factor in admissions decisions now, as I understand.

shortstop,

Is there a reason that you think Harry is tbrosz?

al,"I find Ezra's arguments unpersuasive because common sense and basic economics would tell you co-pays reduce doctor visits, but people would stay just as healthy."

And common sense said for thousands of years that the earth is flat. Sometimes common sense is wrong, and that is why you need scientific studies

It doesn't really matter if he is or he isn't. tbrosz is the gold standard for that particular point of view. Plus, it marks her (and us!) out as Washington Monthly Old Farts, to even have this conversation.

Yancey: Yes. But do ask him yourself if you're in doubt.

David in NY: Indeed, just as one would expect decreased use of care from increased cost, one would also expect that less care would mean worse health. And maybe that's so.

I haven't looked at this in detail yet, but I'd be interested in seeing how this point relates to preventive care and its effect on forestalling or delaying serious and costly diseases. This is one area in which 30 years' research has made a particularly significant difference.

al,"I find Ezra's arguments unpersuasive because common sense and basic economics would tell you co-pays reduce doctor visits, but people would stay just as healthy."

Bobo, Al may have been right this time. This finding, that people "stay just as healthy," may have been flawed if the ones who didn't stay healthy dropped out.

But I'm puzzled by what lee quoted: "there was no statistically significant difference in the rate of use between those who did and those who did not complete the Experiment" Does that mean they followed the people who dropped out? Does this affect the argument? Or are they the people who wouldn't have gotten care, and would have suffered, had they been subjected to the high co-pays all along?

harry,

Are you really tbrosz risen from the dead?

Everybody is talking about gasoline's "demand elasticity." But a lot of people are buying hybrids, too, same as they bought many more small cars after the 70s. Market forces work.

Please FSM, make it stop. Yes, market forces work great, especially in areas where they work.

Medical care isn't like gasoline (or food, before you bring that up next). It isn't, and worshiping Mr Market won't make it so.

Well, he never died, Yance. He's been through quite a few handles these past months: elmendorf, maryjane, monkeybone, rnc, bart, ein, Dingleberry, etc., etc.

"maryjane" seems pretty appropriate to his political philosophy, somehow.

Dunno, "monkeybone" works better. It's especially apt for a "scientist" who aligns himself with the anti-science party.

Shall we go with Stoned Ape?

Hmmm... So you really didn't have a reason to think so, I take it, other than you thought it was tbrosz?

The Nynan paper upon which all of this speculation is based is derived from his guess about why people dropped out of the experiment in greater numbers on the experimental arm randomized to cost sharing. In other words, he has no idea why they dropped out, just a guess about what makes sense for people to stop. That isn't science, it's opinion.

"The explanation that makes the most sense is that the dropouts were participants who had just been diagnosed with an illness that would require a costly hospital procedure. Rather than paying what was likely to be the maximum cost-sharing amount, which could be as much as $1,000
then...many of the cost-sharing participants chose to drop out of the experiment."

Nyman JHPPL 32(5): 759. (2007)

You "take it" wrong, Yancey, as is your wont. Go back and read what I actually said.

Lee: In other words, he has no idea why they dropped out, just a guess about what makes sense for people to stop. That isn't science, it's opinion.

Yes that's correct to a point, as Kevin noted in his post: "it should be considered a tentative criticism." But it is a little more than a "guess;" it's a plausible hypothesis to be considered. If it's a hypothesis the study doesn't rule out, or even address, there's a genuine problem with the study. I'm not sure whether the study did deal with it, so I'm staying agnostic until I find out, but there's a genuine potential problem here, especially if people are using the study for deciding what forms of health insurance are most cost-effective.

BTW, Kevin, did you ever let us all know that Ezra Klein had done an update of his very useful "Health of Nations" at American Prospect?

As David notes, Nynan is generating a testable hypothesis.

Lee, if you have a more plausible hypothesis as to why only 0.4% of the free insurance group dropped out while a remarkable 6.7% of the co-pay group quit, we're all ears.

The key point here is that the 16-fold difference in drop-out rate suggests a major flaw in the study, irrespective of its cause.

Aaron Swartz commented on the RAND study earlier this year:

http://sciencethatmatters.com/archives/30

He noted:

The RAND study was by far the biggest study of this kind, but other studies find similar results. One analysis found that regions whose Medicare programs give out more money (when the underlying healthiness of the residents is held constant) see no increase in survival rates. A replication found the same results in VA hospitals. Cross-national comparisons find "the impact of public spending on health is ... both numerically small and statistically insignificant". Correlational studies find "Environmental variables are far more important than medical care." And there are more where that came from.

best line of the day: "The plural of anecdote is not data." --absent observer

But I'm puzzled by what lee quoted: "there was no statistically significant difference in the rate of use between those who did and those who did not complete the Experiment" Does that mean they followed the people who dropped out?

I would assume it referred to the rate of use while they remained in the experiment, since I'd expect that any measurement of "rate of use" post-drop-out would be subject to all kinds of different issues that would make it difficult to validly compare to those in the experiment, and responsible experimenters wouldn't make claims about it...

Wow. Make a comment and get the "we're all ears" snark...

Okay, here's the deal. I find it interesting that an obscure medical journal publishes a claim of a "fatal flaw" in one of the largest trials of health insurance ever conducted and they don't provide an opportunity for any one of the RAND researchers to respond. It's not like we can't find Joe Newhouse or Bob Brook.

The 1987 paper discusses attrition as well as the book by Newhouse, et al. Estimation of demand for care among those who dropped out cannot be estimated for what happend after they left. However, service use to the point of leaving the study did not differ significantly among those who stayed and those who didn't. In addition, people might have stayed with free care because they got free care. Whereas, people in the other group could have left because they changed jobs, got married/divorced, or moved to France. It's an unknown event. And, anyone can generate a testable hypothesis. That's the easy part. The HIE also protected people from becoming financially less well off due to participation in the experient. No experiment is without some attrition/enrollment and other flaws. And typically one controls for them as best as possible through analysis of data. Calculating a model for drop-out based on baseline measures as well as for the other data that is available (e.g., health status) isn't that hard. I suggest that some people demanding that I come up with other testable hypotheses might want to go read the papers.

Great to see people discussing this very important issue.

Shorter Lee: I ain't got shit.

Here's the best part:

Whereas, people in the other group could have left because they changed jobs, got married/divorced, or moved to France.

I left out the antecedent in this quote precisely because "other group" could refer to *either* group. Unless you can provide reasons for leaving the study which skew towards the co-payers (as Nynan did) and explains the 16-fold opt-out diff, then you ain't got shit.

Nynan, on the other hand, did provide such a reason.

And, anyone can generate a testable hypothesis. That's the easy part.

Yep, so terribly easy that you cannot manage to do it, and are instead crawling under your bed whining about snark.

I suggest that some people demanding that I come up with other testable hypotheses might want to go read the papers.

LMAO @ a rube who is demanding others do his research for him.

The HIE also protected people from becoming financially less well off due to participation in the experient.

That is not true. They were compensated for participation, but that compensation did not guarantee coverage of all their expenses.

But more importantly, compensation for enrollment in the study introduces yet another bias and marks yet another huge flaw in a study that was purporting to measure how people with limited resources demanded health care based upon differing monetary incentives.

I suggest you take your own advice and go look up the RAND study.

I have been a senior healthcare executive for forty years. The Rand study is not only flawed but also the entire market-based consumer incentives model.

Anyone who has ever reviewed real world medical claim experience of a significant number of covered lives knows that 20% or less of insureds generates 80% plus of medical claims. Utilization of health care services significantly disportionate to the norm is not driven by excessive and unnecessary primary and specialty care outpatient visits but by serious chronic and acute illnesses necessitating high-cost inpatient treatment. Consumers who need such treatment are not detered by co-pays and other market-based disincentives to healthcare utilization.

In fact, high co-pays and deductibles are more often than not "penny wise and pound fooliish," i.e. detering consumers from seeking preventive and primary care services that can mitigate the need for more expensive treatment.

High co-pays and deductibles are a stratgegy to shift healthcare costs, not reduce them.

Bob Hall: In fact, high co-pays and deductibles are more often than not "penny wise and pound fooliish," i.e. detering consumers from seeking preventive and primary care services that can mitigate the need for more expensive treatment.

This is what I was getting at in my post of 4:22. Thanks for weighing in on this.

Simple answer to whacko free-market worshipers: If the free market could have solved this problem, it would have done so by now. It's had more than a hundred years.

bleh : And the plural of anecdotes may not be data, but certainly countervailing theory supported by those anecdotes is a reason to question the original theory supported by the original data

You are saying "countervailing theory supported by [no data] is a reason to question [a] theory supported by [some] data". In the usual scientific way of using "theory," by the time it becomes a theory, it has a pile of data underneath it. Something with only anecdotal "evidence" behind it is more properly a hypothesis or a conjecture.

Hey Disputo,

I really am impressed with your anonymous comments on a blog. Why don't you send your stuff to Health Affairs? I'm sure they're waiting for your well reasoned commentary and remarks. Hiding under my bed? You're right, I don't have anything with which to respond to you. You win. I quit.

I now understand why I never waste time looking at comments on Kevin's blog.

Please spend some more of your valuable intellectual prowess on this important issue.

And Bob Hall ends the discussion with facts and logic, things which are like garlic to vampires and conservatives.

LMAO @ the anonymous poster bashing me for being anonymous (incorrect -- I'm pseudonymous).

My guess is that AH is back again testing the mod.

Bob Hall

Thank you for injecting insight and analysis into this discussion.

I learned what was bothering me about it, you crystallised it brilliantly.

FWIW the Canadian data supports you. Canadians argue about copay, but they never go there.

Since the 80/20 or even the 90/10 rule applies to healthcare expenditures, the key in healthcare is to get those with chronic conditions like diabetes to be managing them to the best medical protocol, from early on. That can save masses of expensive treatment later in the course of their lives, or at least delay it, and increase the term of their productive healthy lives.

Copay just seems to generate a blizzard of paperwork.

This is where tbrosz usually is by the way:

http://www.habitablezone.com/space/

The real question is how do they affect outcomes for poor people and the real study there should be of Medicaid and of managed care.

shortstop,

Well, I am waiting. How do you know Harry is tbrosz, along with all the other pseudonyms you have listed?

Wouldn't you get some tests done (colonoscopy, heart stress tests, etc.) if the copay wasn't $1500 or more?

I'm not going to tell you that, Yancey. You asked him, he declined to answer you (at least in this thread--don't know about email) and you can draw inferences from his odd failure to deny it.

Whether or not you believe me isn't of consequence; I'm not going to stop addressing him by his real name simply because you're not satisfied or pretending not to be.

I will take you partway there, Yancey. Note that if your industriousness matched your thirst for knowledge, you could have found this yourself.