The follow-up study on Medicaid coverage in Oregon

Here is some overview coverage from Annie Lowrey, an important issue of course with some striking results.  Here is coverage from Sarah Kliff.  Here is commentary from Justin Wolfers, and here.  After the R&R saga, I say it’s time for someone to stand up and admit “We have some egg on our face with this one.”

Addendum: Reading more carefully through the quotations from Finkelstein and Holahan in the Lowrey piece, I find it amazing, and I suppose even embarrassing, what those commentators are claiming as a positive result.  Of course it is worth comparing the program to simply giving people the cash.


My biggest issue- what is follow up time period? Paper claims spending on preventative medicine increased- maybe that only shows up as improved health over 5 , 10, 20 year period?

I believe it was two years. However, I believe they chose to track health variables that are the most open to the influence of medical intervention.

So is there an un-gated version of the paper? (If not, it's absurd...this was all done with taxpayer money!)

I'd like to see for myself what the confidence intervals look like. Is Wolfers right?

The question is: Who has the burden of proof? And what if we continue to get "absence of evidence?" Can we roll it back? It always seems that the cause du jour gets cut the most slack.

The question is: Who has the burden of proof?

In any sensible polity, on the people asking you to pay for it.

Am I misunderstanding Wolfers' second tweet? It seems like he is placing the burden on critics of a policy to perform the logically impossible task of proving the negative (prove with evidence that no evidence supporting the policy does or can exist). A better method would start by asking a policy's proponents--do you have any evidence that your idea works?

One side starts with the presumption that before we spend money we should know if we expect to get value for it.

The other side starts with the presumption that we should all have the same things.

This is because universal access to health care is a tenet of the progressive creed, a dogma or postulate that is core to the identity of being a member of the progressive group. It is like arguing with creationists. There is no amount of evidene that one could offer that would change their point of view.

It seems he's also falling into the aphorism trap of saying "absence of evidence is not evidence of absence."

But that is wrong. If you look for something where it ought to be and cannot find it, that's evidence it doesn't exist.

NB: It is true that "absence of proof is not proof of absence." Evidence is not synonymous with proof.

Am I the only one not understanding what is supposedly in contention here? What is the supposed egg? I see a paper that has multiple conclusions, and articles that summarize it slightly differently.

The study concludes that expanding access to medicaid increased expenditure on "health care" but did not improve health (except for lowering the incidence of depression). Does that sound like an effective program?

Paging Robin Hanson... Healthcare doesn't make people any healthier, just makes them feel cared-for. So stress and depression dropped, but health outcomes are no better.

Stress and depression are health outcomes, and Hansonian analysis is for grads who think trite contrarianism is a cool way to troll their Facebook friends.

The key points are:

1) happiness is difficult to measure scientifically. One should not just accept a happiness statistic blindly. One should have some skepticism about how strong the effect actually is. How controlled was the context in which happiness was measured, etc. Was there a prompt in the question or the context that influenced the answer in one direction or another etc.

2) even if accurate, the increase in happiness may be just a status effect. They won something others like them didn't. If so, then universal coverage would actually diminish that result.

3) the choices available are not merely two, doing nothing and expanding Medicaid. Other mechanisms might increase happiness and lower stress better at the same cost. Other mechanisms might increase happiness and stress the same amount at lesser cost.

They remind a group of poor people who won a lottery that they won the lottery, and remind the losers, also poor, that they lost, screen for depression, and the significant result is that those who were reminded they won the lottery are slightly less likely to be depressed than those reminded they didn't? Why are we attributing that to the access to health care and not the prompt?

You make light, and bring up something they ought to correct for. (And in some areas of social science, your theory is taken for granted. Look up "stereotype threat.")

But in all seriousness, having your health care paid for really ought to reduce your stress. This is probably a very real phenomenon. Probably not worth the money it cost, but it's still real and we shouldn't laugh it away.

Whoa! "Universal" health care doesn't prevent poor people from consuming metric tons of sugar. This is so embarrassing.

Sadly, Medicaid enrollment may increase the amount of sugar people consume, as they will have more money to spend on food. It is also possible that now that their health problems are paid for by someone else, Medicaid enrollees will be less cautious in what they consume. This last point seems doubtful to me, however, given that insulin injections and limb amputations are inferior substitutes for a properly functioning metabolic system and set of feet. Most people stuffing themselves full of garbage lack sufficient self-control to improve their health even when it is in their interest to do so, as it almost always is.

Sorry - did you just say poverty improves health outcomes?

We were poorer, but more fit 50 yrs ago.

citation ? ( i doubt it)

tt, obesity numbers?

Also, I recently saw a study where there was some issue that resulted in a lack of availability of food (in some other country), causing caloric consumption to go way down. Health outcomes skyrocketed, obesity cratered, heart attacks and strokes dropped. When the temporary issue abated, calorie consumption went right back up and health outcomes cratered again. I wish Ii could remember what it was, it was recent.

Cliff, I believe it was Cuba in the 90's:

life expectancy is higher now

Are the empirical estimates really that different than what you would have expected going in? It's not clear to me what the story is here.

Don't we need to look confidence intervals and at other previous research on how large we would reasonably expect treatment effects to be? Are these results surprising to pro-ACA health researchers? How surprising?

(Of course the popular press coverage will be bad, and you maybe can't blame Finkelstein and Holahan for what the reporter decided to quote. Even if she is married to Ezra Klein :-).)

You probably won't get a better set up than this, a lottery based Medicaid population. The only significant missing data is just the longer term effects. A five year, ten year study etc. However, I'm assuming that the entire population will be covered soon due to Obamacare.

However, the results of this study will probably be the center point of many a healthcare debate going forward.

I've got to admit the Oregon program sounded better as reported a few months ago.

Apparently from reading articles on it written back in 2011 the first report did not have any blood tests and just relied on self-reported health status.

Ok, I assume the egg on the face is in reference to the authors of the paper? Their sample size isn't big enough?

Not sure what Tyler Cowen is so animated about.

Please explain.


ok i see the addendum (which I didn't see earlier). So my next question is why is the NYTs so excited about this? I would think the NYTs would bury any article that says Medicaid doesn't help.

Maybe the financial security counters this but not so sure...

You are right. If you check out Anne Lowrey's front-page article in the NYT (as opposed to her Economix piece), you'll find that the headline does indeed bury the inconvenient finding that Medicaid does not help. The headline simply reprises the findings already reported one year ago that Medicaid increases use of medical services. Only later in the article do you get the bad news.

And Lowrey certainly downplays the bad news in her Economix post on the subject:

In my paper it was below the fold and in the business section. I think that qualifies as "buried".

Well, the study did show that the program led to less frowning.

Finkelstein really does come across as desperate (and statistically naïve to a degree that I am quite sure she is not). "Well, we measured all these things, there must be some statically significant positive correlation in there somewhere. Look: it decreased depression!"

My first reading of this was Straussian (though maybe I'm being too generous). The results of the study seem so clear, and the defense of Medicaid so ridiculous (at least on the basis of this study), that I have to believe that this defense is intended to come across as ridiculous. If Finkelstein were to explicitly come out against Medicaid, it would be interpreted on the left as an attack against them, at which point reason would shut down (I'm not claiming anything different for the right). By presenting the results in this way, it may get people on the left thinking: "If this is the best someone on our side can do...."

Wolfers did at least say something true, albeit clearly diversionary. The lack of a statistically significant improvement in health outcomes could indeed be because the size of the improvement is too small to pick up with this sample size. The fact that the study did pick up a statistically significant increase in utilization does show that the fractional increase in utilization is larger than the fractional increase in health. Another way to handle this is to express the result as an upper bound (at 95% CL) on the size of "health increase per Medicaid dollar spent". Then, before the next study is done, Wolfers should answer the question: "below what level of health increase per Medicaid dollar spent should we not do the ACA Medicaid expansion?"

10,000 is a large sample. And the thing about statistically significance: a researcher can just sample until she gets the confidence level she wants. What size sample does Wolfers want? 20,000? 100,000?

Not sure why Wolfers thinks 10,000 is a small sample size when he was just trumpeting that his study "couldn't find ANY" unhappy millionaires in a sample size of 8. Surely I'm missing something.

There are some aspects you haven't considered.

I am retired, with Medicare Part A but currently refusing participation in Medicare Part B, which would cost me some outrageous $120 per month. Now I find out that, in my home state of Colorado, because of my income below 133% of poverty level, I qualify for Medicaid, which will pay my $120 per month and, in addition, my co-pays for Medicare treatment. I will not have to putrify standing in the Medicaid queue, in spite of the fact that my wealth, mostly in real estate, approaches $500,000.

My girlfriend earns a bundle and is also worth $500,000, and I'd be a fool to marry her.

Why should I qualify for free medical care?

Jimbino, I'm preeeetty sure you don't qualify for Medicaid if you have $500,000 in real estate assets. Try signing up -- they actually check for that.

Is this true, still? It is hard for me to find the actual answer. All I see is that the Medicaid expansion focuses on income, not assets.

Medicare spending in Oregon is 11,000 per year,
but the wash post said those covered by medicare only consomed an additional 1,100 in medical care. Can someone check the ungtated source for me?

Table 5 says controls spent estimated $3,257, and treatment group spent $1,172 more than that. "Annual spending was calculated by multiplying the numbers of prescription drugs, office visits, visits to the emergency department, and hospital admissions by the estimated cost of each. See the Supplementary Appendix for details."

Maybe it's marginal versus average? Although this curve seems to bend the opposite direction of what I'd naively expect.

Health care is for sick people. Wellness means eat a balanced diet, exercise, don't smoke and drink in moderation if at all. Is there some great mystery to this?

There is a great deal of explanatory power in this humble comment. Note that not only did the Oregon Medicaid treatment not improve outcomes, it "did not seem to have an effect on use of emergency departments" ( (There goes one important channel for cost savings we were promised from the ACA!) The simplest explanation for all these facts is that, aside from good luck, the most important factors for good health are the same as the most important factors for not being poor: good planning capabilities, good impulse control, good critical thinking skills coupled with good basic knowledge.

So good inheritance generally.

Austin Frakt and Aaron Carroll do a better job reviewing it than anyone else I have seen:

I agree, and their summary is fairly even-handed for a couple of wonks who clearly believe Medicaid is a helpful program for the poor.

One note on the statistics: Frakt and Carroll highlight the fact that the point estimates for the many of the health outcomes actually reveal slightly better outcomes for Medicaid patients (cholesterol, blood pressure, glycemic levels for diabetes). Are these effects statistically significant at the arbitrarily chosen p value of .05? No, but the authors do not provide a power calculation, so who knows how large the difference in outcomes would need to be to yield a "significant" result given the sample size.

And I think the most fundamental point they make is that health insurance, which expands access to health care, is obviously useful to people. There is a reason people buy it, a reason a worker will sometimes endure a really crappy job that nonetheless has a good health insurance plan. We never said there needed to be a study of private health insurance versus no insurance, because it is clear that access to routine care, protection from catastrophic health events and financial protection is something pretty much everyone values.

1/sqrt(N) is usually a pretty good back-of-the-envelope estimate for power. With a 10,000 person sample, you should expect to pick up effects as small as 1%.

While I certainly don't disagree with your point that many people value health insurance highly, I don't agree that carries more weight in the question of what we should provide to the indigent than actual outcome studies. People value nice vacations, psychotherapy, cool-looking smart-phones, and acupuncture. But welfare is not in the business of providing poor people with what everyone values; it is in the business of providing them a minimum level of subsistence for the minimum possible cost to the rest of us, and for that the question of how much objective (not subjective) benefit these welfare recipients got out of this particular treatment is highly relevant.

10,000 is a relatively large sample overall, but only a fraction of that sample actually had the outcomes measured. Only some people began with diabetes, high cholesterol, etc., limiting the pool of people who could be measured. It is telling that that the outcome that the highest proportion of participants stuffed from, depression, was also the one where Medicaid had a significant impact. Posted with the positive point estimates for most outcomes, I think it implies that sample size may have something to do with it.

How many Mercatus Center employees choose to forego health insurance, or have since becoming adults?

How many would rather save the money than have their families covered?

Libertarianism: "the poor should not have what we would never go without."

Show me all the libertarians who think the poor shouldn't get health insurance. I think everyone should buy catastrophic health insurance to guard against potentially, well, catastrophic medical events. However I do think that "insurance" which pays for damn near all medical expenses is ludicrous.

And you are living in the only industrial country where you can actually enjoy that perspective.

Of course, the rest of that world, which has lived with essentially universal health care for ca. two generations thinks that this a perspective not worth even laughing about when dismissing it.

Oh no, they laugh about it as they dismiss it..

Well, along with a real avoidance of travelling to the U.S. Or even more concretely, to working in the U.S., unless their employer guarantees they will not lose the sort of health coverage which is routine at home. I know Germans who have turned down work in the U.S. (Mercedes and associated supplier companies) for fear of the American health care system. Germans who had no problem working in China or Brazil, in comparison.

prior_approval: You are either making that up or your compatriots are exceedingly ill-informed. I have worked in the US, Germany, and China. While China has what it calls "universal health care" clinics, the lines are so long and the quality of care so poor by Western standards that no western expats use them. The go to expensive, western-level clinics paid for by their employers. So in that sense your access to good health care when working in China is just as precarious as in the US. Of course, in a broader sense, it is still worse. In the US you can seek out an outstanding specialist for any treatment beyond the expertise of your local provider; in China you would likely need to leave the country. And of course, anyone who, when presented with the hypothetical "if you were to have no medical insurance and have a heart attack on a random street, would you prefer to be in China or the US", answers "China" is either stupid, dishonest, or just plain doesn't exist.

Europeans have absurdly paranoid beliefs about the American health system, just as Americans do about European health systems. I once had a doctor in England remark that of course, I was lucky that the NHS was free because I'd never be able to afford treatment at home. I had strep throat. And I know other expats who've said the same. Anyone who works for Mercedes and is afraid of going to the US because of our health system is an ill-informed hysteric, not someone we should look to on our health policy.

Europeans have absurdly paranoid beliefs about the American health system, just as Americans do about European health systems\

Yes. I'm pretty sure that "absurdly paranoid" is the only way to describe someone who thinks that an American company would go through the trouble of searching for job candidates in foreign countries, pay to handle their immigration paperwork and relocation, but not spend five figures (tax-free!) on their gold-plated health plan that would be part-and-parcel of the compensation package for anyone worth going through that much trouble in the first place.

Even Americans who think the US should move to the Euro-model think that the rich here have very nice health care.

A few years ago the Canadian prime minister facing re-election when asked about the waiting times for treatment suggested that if you got impatient you could go the US to get treatment.

What you are doing is comparing the tax paying employed middle class people in places like Canada to the poor in the US. I pay taxes and fees for medical services here, and if I was in the US I would be doing similar. The medical services available to me would be vastly better and more timely than what we face under the universal system.

Well, let me repeat a link -

'The gap in life expectancy between Canada and the U.S. continues to widen; Canadians now live three years longer than Americans.


Canada has earned a steady “B” on this indicator for nearly five decades. Japan’s impressive increase in life expectancy has set a high bar for attaining an “A” grade on average in this decade, a standard that only Switzerland has met. Moreover, Japan, which had among the lowest life expectancies in the 1960s, has been a steady “A” performer ever since. Of the five top-ranked countries in the 1970s, only Japan has been able to consistently maintain its “A.” Switzerland dropped to a “B” in the 1990s, but regained its “A” ranking in the 2000s.

The Netherlands, Norway, and Sweden, on the other hand, have all lost their top-ranking status.

The U.S. has also seen a decline in its relative performance. It earned a “C” in the 1970s and 1980s and dropped to a “D” in the 1990s, where it has remained.

Countries that have jumped from being “D” to “C” grade performers since the 1990s include Austria, Finland, Germany, Ireland, and the United Kingdom.'

In other words, only one health care system has shown itself consistently able to underperform all the other ones when using life expectancy as a measure. And it isn't Canada's.

But life expectancy is a pretty horrible measure for a developed nation, right? What results do you get when adjusting for demographics?

"Well, let me repeat a link -"
It is what you do best.

Of course, the very fact that your link is to, whatever the hell that is, shows that you have a conclusion, and look for a link to support it. What exactly does it mean to get a C or a D according to the Canadian Conference Board? Who the hell knows! The important part is, the US scored low.

I'm not sure what the point here is. People who have more assets will generally spend more to protect them. I think that's the best understanding of the benefit of health insurance.

And to somewhat repeat myself - the rest of that world, which has lived with essentially universal health care for ca. two generations, does not share that 'best understanding of the benefit of health insurance.'

Those societies with essentially universal health care can just point to empirical data to show the benefits of such a system. As a concrete example - the benefit of health care for all children leads to a benefit that is not tied to any individual at all.

Mine was a reply to Chet. Obviously it's possible to imagine quite different systems than the one we have here, including ones in which the state provides the care universally and thus there is no insurance, as there's nothing to insure against. In the US, health insurance prepays for health consumption and protects against catastrophic loss. As we see in this study, the consumption of many treatments may not have significant effect on health outcomes and thus may have little financial value. Nonetheless, even the poor like the true insurance aspect, and we'd expect that those wealthier, such as Mercatus employees, would like it even more.

Is there a significant difference in the levels of research and development coming out of countries with universal health care vs those who dont (the US mostly).

I have no proof of this but ive always felt that the US essentially pays for the world's medical R&D. And by the US, i mean the citizens of the US with health insurance.

This sounds more like the critique used on Michael Moore for not providing health care for his staff after arguing that everyone should have it.

The U.S has the most responsive healthcare system in the world. While it may have problems, I do not see libertarians arguing that the poor should "not have" any. If you paid attention to the most staunch libertarian rhetoric, they're saying (in summary) that the poor (and everyone else) would be better served with a health care system that is not laden with excessive government involvement and red tape. They argue that costs could reduce, charitable services could increase, etc. and that everyone could have access to affordable health care via the market.

Its ok to disagree with them, but understand their POV first. Taken your opponent out of context does not help your argument.

I missed something here. TC mentions R&R; did the Medicaid study screw up their data somehow and get caught in the error?

No, I think the point is that progressives' evaluation of a study appears to be directly proportional to whether it confirms their ex ante bias.

'After the R&R saga, I say it’s time for someone to stand up and admit “We have some egg on our face with this one.”'

How is that Mercatus Center project to make its publications accurately reflect current data going, by the way?

Or instead of actually doing the work of making available accurate information for policy consideration, will the center simply decide that egg isn't really that big of a deal in an academic setting where donors are interested in conclusions.

I assume that Wolfers is just recycling some comments he made about R & R?

Mood affiliation abounds!

Agreed. Any comment saying that this "proves" or "disproves" the benefits of Obamacare is hogwash. I'd say the only thing this study shows is that the benefits of health insurance (universal or not) are primarily financial, not health-related, and that if we're going to be honest we should start addressing them as such. But that in itself mught be mood affiliation.

Now there's a question. Why does humanity tend do health care as an insurance scheme, where some people incur hundreds of thousands of dollars in cancer treatment while most incur nothing and who's who isn't known beforehand, rather than a fixed across-the-board cash handout, as Tyler suggests? This could be a revolution in our understanding of insurance markets.

See The Incidental Economist's recent review of Catastrophic Care by David Goldhill.

The conclusion one should draw seems pretty clear. A la Robin Hanson, we know from the RAND HIE that more utilization does not produce better health outcomes.

This should be one's reasonably strong prior. The current evidence from the OHS certainly provides no reason to reduce the strength of this prior.

If I recall correctly, the Moving to Opportunity housing voucher experiments, which had nothing to do with health care were also found to improve the mental health of poor mothers. Having a nicer apartment in a better neighborhood made the women happier. (It did not, sadly, have the effects on employment and welfare use that people hoped for.)

Tyler's suggestion that we should compare effects of receiving Medicare to effects of receiving cash is a good one.

There are a number of studies on inexpensive yet effective ways outside of traditional medical treatment to reduce moderate depression.

These articles refer to studies that consumption of Omega -3 and other fish oils reduce depression.

This Harvard link refers to several studies showing that moderate exercise reduces depression.

Updated here:

Kevin Drum's post (, while still containing the requisite laundry-list of excuses, does have the benefit of opening with an honest assessment: "This is a disappointing result that raises obvious questions about Medicaid expansion." (The next word is, of course, "but".)

Drum also does make one philosophically interesting point further in: "Overall, I'm a little unclear about what the conservatives who are crowing over this study really think. They obviously believe that access to healthcare is a good thing for themselves. (At least, I haven't heard any of them swearing off doctor visits.) But you can't have it both ways. If it's a good thing for us middle-class types, it's a good thing for poor people too. Conversely, if it's useless for poor people, then it's useless for the rest of us too. So which is it?"

First, I would say that (while there are a lot of other contributing priors, and the applicability of this study to the rich, who have a different set of behaviors, isn't clear, etc. etc.) this study does indeed lower my confidence in the value of preventative care for myself somewhat.

Second, I would say that the point of welfare systems isn't to insure that the poor get all the stuff that rich people value. Some of the medical care that some rich people get I think is a total crock (e.g. psychotherapy), but it's fine with me if they buy it on their own dime. Some of the medical care that some rich people get (e.g. knee replacement surgery) I think is of questionable value compared to cheaper alternatives, but again who am I to complain if I don't have to pay? Some of the medical care I buy myself is probably providing me peace of mind (e.g. MRIs where x-rays would do), bragging rights (e.g. cool 3D motion ultrasounds of babies), and other things of little objectively measureable value. But as soon as you ask me to pay for someone else, I am very interested in knowing that what I am buying is providing objective (not psychological) value for the absolute minimum cost.

"Overall, I’m a little unclear about what the conservatives who are crowing over this study really think. They obviously believe that access to healthcare is a good thing for themselves. (At least, I haven’t heard any of them swearing off doctor visits.) "

As a conservative, at least as Drum would define that term, I kinda have "sworn off doctor's visits." I haven't been in over 3 years, even though my (company-provided) insurance would pay for it, because I haven't been meaningfully ill and I don't see any benefit in going to the doctor just for the hell of it. I have health insurance in case I get cancer, or find myself on the wrong end of a bus bumper. So if we accept Drum's "good for the gander, good for the goose" argument, *that's* the kind of coverage we should be providing to poor people. But I suspect Drum would not like that suggestion, and I KNOW Drum's readers would basically call me Hitler for even bringing it up.

You've said that very well.

What is the moral argument to compel another person to pay for objectively ineffective treatment? Drum tries to generalize his retort at a level where the effectiveness of the treatment is not a factor. But that is the whole point.

The point about low power is not the aggregate sample size. It is the fact that within the aggregate sample, only a small fraction of that population is at risk of heart disease, diabetes, etc. So, the power of the tests to determine whether the program benefited that sub-set of the sample is probably fairly low. This is a familiar problem in micro-research when looking for program effects on sub-sets of the population. Let me give an example. Let's say that I'm using an Australian household data set with a sample of 10,000 and looking to see whether a policy affected the relative employment rates of men and women. The sample size should be pretty powerful in those circumstances. But it will be a lot less powerful if I am interested in the impact of the policy on indigenous youth, as the survey may include only 20 relevant people. The Oregon study did find that the point estimates mostly went in the right direction for these more specific health conditions, so I'd say that the conclusion should be "more research needed", rather than Medicaid expansion is useless.

If you cannot find effects when studying two groups of 10,000 each, your NNT must be huge.

(I'm a little too flip here, because you are looking at groups of people that aren't going to die in large numbers anyway. But another way to look at that is that you are looking at groups of people that aren't likely to die anyway, health care or no.)

Please, please -

Stop referring to these programs as "Insurance."

Insurance is the transfer of **Risk**.

These programs are transfers of **costs**.

I'm curious about the big picture financials:

* Are hospitals in Oregon doing better financially because they are absorbing fewer costs of uninsured people?
* Are regular insurance costs in Oregon going down because hospitals are not cross-subsidizing the uninsured?
* With more insured people, could non-hospital forms of medical care grow and thrive now that they no longer need to treat large numbers of insured people at hospitals to cross-subsidize the care of the uninsured?

These are good questions.

I believe the Oregon study found no significant change in ED usage, but I can't find a citation to that now, so treat my citation warily. (Also, ED spending is only about 1% of America's health care costs, so any savings there by necessity will be small.)

I am very confused by Tyler's comments on this one. Usually I understand his points and references, but I'm not even sure who his "someone" and "those commentators" are referring to.

My 2 cents:

1) This study seems to provide clear evidence for the effectiveness of the "catastrophic coverage" portion of spending.
2) This study raises questions about the value of preventative care in general - at least on a two year time horizon.
2a) The point estimates cut the other way here. I was surprised some of them were statistically insignificant given the magnitude of change.

My prior here was that Medicaid expansion was "the right thing to do" and worth doing. Nothing here really makes my confidence in that prior higher or lower. I would not be particularly interested in comparing to cash transfers, but I would be interested in comparing to a catastrophic coverage regime. Real value delivered so far seems to be the 80% reduction in major medical expenses.

From Wonkblog's coverage:

A big criticism of Medicaid is that it pays doctors so little that it’s essentially worthless because no doctor will see you. But the Oregon residents who won the Medicaid lottery got much more health care — including preventive health care — than the residents who lost it. They also saw catastrophic health costs basically vanish.
There are a number of possible spins you can put on that finding. One is that the study was simply too small, with too few sick people, to show the kind of quick health changes the researchers were looking for.

So, 10,000 patients is too small, but the 15-20 doctors they saw that are still in business - in an environment in which other people had other choices, including cash payment - is an appropriate sample?

And I agree that this shows the importance of catastrophic care (what insurance used to mean and still does in every other context) while calling into question the supposed benefits of preventive medicine.

This post is pure mood affiliation. There is no analogy to RR - that study's results were proven false after being loudly trumpeted by its authors as a definitive answer to a public policy question. No one is disputing this study in and of itself - only what it's implications are for public policy. Point taken about the comparison to cash grants, but Medicaid is insurance - a financial product - and the study confirmed its benefits as a financial product. It seems their are legitimate disagreements about the implications of the study's findings, but Medicaid's proponents are far from promoting false results as justification for a dubious policy prescription.

If anyone should admit egg on their face, it is health care providers. While they are renumerated among the top 1%, and most likely believe their compensation is well deserved because of their great contribution to society, and might largely believe their services are too valuable to waste on Medicaid patients, yet another study proves their marginal product is zero.

If people are poor enough to be on Medicaid, their whole life is kind of an economic catastrophe. They might get free meds for diabetes, but that does not help them get a decent job. (if they did get a decent job, they might well lose Medicaid).

If they lose Medicaid and get hospitalized, they very rarely have to pay the bills.

Therefore the real goal of Medicaid is to pay doctors and hospitals and nursing homes for the charity care they used to provide for free.

But America being America, we want to channel all this money on a per patient basis -- rather than having free community clinics and free public hospitals.

And being America, we want to measure the personal health consequences of our social insurance programs.

I favor what I will loosely call the British approach -- i.e. here is a National Health Service that your tax dollars have paid for. If it makes you healthier, fine, if not, who cares? You wait not to pay, or you pay not to wait. The government is not responsible for making you healthier -- the government is responsible for maintaining a place where you can get health care.

When I managed a public library, I did not know or care if my patrons became smarter. My job was to provide books for free.

Bob Hertz, Director
The Health Care Crusade

Comments for this post are closed