Health Insurance and Death
Cato’s Michael Cannon has a post up which, unlike Charles Lane’s writings on the matter, actually attempts to engage on the merits with the argument that lack of health insurance leads to people dying. But even though he cites a number of studies on the subject, only one, the Richard Kronick study, actually address the point the Institute of Medicine study was referring to. Kronick concludes that the approximately 20,000 deaths per year is an overestimate due to a failure to control for relevant demographic variables. Conversely, Andrew P. Wilper, Steffie Woolhandler, Karen E. Lasser, Danny McCormick, David H. Bor, and David U. Himmelstein found in their “Health Insurance and Mortality in US Adults” that the IOM study is a serious underestimation:
The study, which analyzed data from national surveys carried out by the Centers for Disease Control and Prevention (CDC), assessed death rates after taking education, income and many other factors including smoking, drinking and obesity into account. It estimated that lack of health insurance causes 44,789 excess deaths annually.
Previous estimates from the IOM and others had put that figure near 18,000. The methods used in the current study were similar to those employed by the IOM in 2002, which in turn were based on a pioneering 1993 study of health insurance and mortality.
A separate study indicated that individuals lacking health insurance were more likely to die even in the context of emergency room visits:
The researchers took into account the severity of the injuries and the patients’ race, gender and age. After those adjustments, they still found the uninsured were 80 percent more likely to die than those with insurance — even low-income patients insured by the government’s Medicaid program.
“I’m really surprised,” said Dr. Eric Lavonas of the American College of Emergency Physicians and a doctor at Denver Health Medical Center. “It’s well known that people without health insurance don’t get the same quality of health care in this country, but I would have thought that this group of patients would be the least vulnerable.”
If “no health insurance means bad health care” or “bad health care can lead to death” were some kind of wildly counterintuitive propositions at odds with sound theory, then I think we might have good reason to doubt these results. But in fact they’re perfectly intuitive results that accord perfectly with, among other things, the elementary economic theory that says people wouldn’t pay for health insurance unless it provided something of value. The bulk of empirical studies also seem to confirm this. Lack of insurance is associated with low-quality health care which is associated with enhanced risk of death.
63 Responses to “Health Insurance and Death”
Register
RSSLogin
Yglesias Tweets
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
December 14th, 2009 at 4:52 pm
There is uncertainty concerning the issue, therefore, the issue is invalid and never to be mentioned again. Back to praising Holy Joe and attacking his disgusting left wing attackers!
December 14th, 2009 at 4:57 pm
The failure of the Nelson amendment siginificantly reduced the odds of health care reform passing.
Yet, most Democratic Senators voted against it, and it was defeated.
Can we now consider them all the moral equivalent of mass murderers?
I recall MY acknowledging that without Nelson, health care reform did not have the votes, but he said voting against it was the right thing to do.
Is he also complicit in thousands of deaths?
December 14th, 2009 at 4:59 pm
Furthermore, the Cato-ites are arguing against the standard American Health Calvinism rationalization, whereby access to health insurance and thus healthcare is a motivator, and having timely, appropriate, affordable healthcare is an indicator of one’s worthiness:
It also suggests that Cato’s forthright defenders of economic liberty don’t have the courage of their convictions to say “well, sucks to be you” to those who are third-class healthcare citizens. The appropriate Cato position ought to be that a little premature death and debilitation is a price worth paying for the unique freedom available to the consumers of healthcare in the US.
December 14th, 2009 at 5:07 pm
Can we now consider them all the moral equivalent of mass murderers?
Because they won’t collectively pay the ransom on a hundred thousand lives? You are sick.
December 14th, 2009 at 5:07 pm
“Back to praising Holy Joe and attacking his disgusting left wing attackers!”
Matt’s just doing his job. Podesta clients Cigna and HealthSouth love the Lieberman/Romneycare bill.
With a flood of government mandated new privitized customers, the Lieberman/Romneycare bill means big profits for them.
December 14th, 2009 at 5:08 pm
Matthew Yglesias,
The bulk of empirical studies also seem to confirm this. Lack of insurance is associated with low-quality health care which is associated with enhanced risk of death.
Nonsense. Have you made a careful review of the literature to evaluate what “the bulk of empirical studies” suggest? No, of course you haven’t. You simply make this claim because you desperately want it to be true.
And your assertion that only one of the studies cited by Cannon “actually address[es] the point the Institute of Medicine study was referring to” is also false. In fact, the Levy and Meltzer meta-analysis Cannon mentions examines far more studies than either the IOM report or Kronick’s paper, and L&M’s conclusion is that the question is essentially unresolved.
And then there’s the famous Rand Health Insurance Experiment, which was by far the largest and most comprehensive study of the effects of health insurance on health ever conducted. And the central finding of that study after many years of tracking thousands of people was that health insurance has little or no impact on health.
December 14th, 2009 at 5:09 pm
“A separate study indicated that individuals lacking health insurance were more likely to die even in the context of emergency room visits:”
That’s hardly surprising. My brother went into a hospital with appendicitis and no insurance. He needed emergency surgery, but without insurance, the hospital sent him home saying he must not have appendicitis because he was able to walk. That night, his appendix burst, nearly killing him. Instead of a routine appendectomy, he spent ten days in the ICU being treated for sepsis and then got the appendectomy. And the extra cost was $120K. He managed to knock $90K off the cost by threatening a malpractice suit. But he still had to pay $30K extra to have the hospital nearly kill him. As it turned out, he died 10 months later anyway because his appendicitis was actually carcinoid syndrome that had already metastasized to his liver. They could have caught that a year earlier when he started seeing a doctor about the symptoms (which are actually rare). But again, he had no insurance to pay for the tests, so he didn’t get them. Had he had insurance, he’d probably be alive today.
December 14th, 2009 at 5:16 pm
The problem with your example fostert is that it tracks both with the empirical evidence and with logic. It must therefore be dismissed by logic impaired fuckwits who think that there can be no such thing as torture or side-effects from a health “care” system that avoids treating millions.
December 14th, 2009 at 5:17 pm
Just sayin’.
December 14th, 2009 at 5:17 pm
From the linked study:
Now, Matthew is assuring me that uninsurance causes deaths. The study only says that uninsurance is associated with” mortality.
As usual, the Left doesn’t understand the difference between correlation and causation.
It is clear that the Left hates science. After all, how many times does it have to be pointed out that correlation is not the same as causation.
December 14th, 2009 at 5:18 pm
Damned tags.
December 14th, 2009 at 5:22 pm
Is this sort of thinking really something Matt believes? He’s described a significant percentage of modern medicine as witchcraft, and yet people pay for that, don’t they?
December 14th, 2009 at 5:26 pm
Matt’s mistaken when he says that only one of the articles that Cannon cites are responsive to the IOM study. All of them are, and to a certain extent the Medicare studies (cited by Cannon and Kronick) are superior to the other studies, because they offer a natural experiment.
And the fact that studies disagree isn’t a reason for us to adopt Matt’s priors. Or rather, with some of them, since Matt isn’t consistent in his thinking.
December 14th, 2009 at 5:28 pm
Al,
Yes indeed. The text Matthew quotes making the claim of causation is not from the study but from a press release about the study put out by Physicians for a National Health Program, a political lobbying organization that promotes a single-payer health care system.
As you note, the study abstract makes no claim of causation, merely a correlation.
December 14th, 2009 at 5:30 pm
Exactly, Al. Clearly the causation arrow points in the other direction and death causes uninsurance. OMG the libs are sooo stupid why cant they understand our superior conservative logic? [end snark] Moron.
December 14th, 2009 at 5:31 pm
Let’s do a little gedanken experiment. Everone, even Mixner, has heard or know of people who died because lack of insurance delayed diagnosis and treatment of a life threatening illness.
Therefore, if Mixner is right, and health insurance has no effect on overall death rates, the number of lives saved by having insurance would have to be balanced by an equal number of individuals who died because they had health insurance and received unnecessary fatal treatment at the hands of physicians.
Where are those stories? How many thousands of Americans lose their lives each year because the deadly effects of health insurance? I’m waiting to see the data.
Another thing – does Mixner and the rest of his ilk have the courage of their convictions? If they did, they should give up their own health insurance, as it doesn’t improve their health.
December 14th, 2009 at 5:34 pm
Mixner is lying about the RHIE, but that is mainly because he wants to sidetrack the thread.
December 14th, 2009 at 5:34 pm
Sure, but that’s hardly the intuitive conclusion, is it? In other words, of the two propositions, it’s the one about a significant amount of modern medicine not actually having any merit or benefit (but is nontheless purchased) that is the counterintuitive one, and should be subject to a greater demand for supporting evidence for acceptance – rather than the idea that people buy health insurance because it makes available health care they would not otherwise get, which is quite intuitive and shouldn’t be particularly contentious.
So I really don’t understand your objection. Surely it’s not inconsistent for Yglesias to believe that people often act rationally in the marketplace, and also often don’t.
December 14th, 2009 at 5:37 pm
dan, poor health may cause uninsurance. Poverty may cause uninsurance. Poor health and poverty are both associated with higher mortality. Hence, uninsurance may be associated with higher mortality. That doesn’t mean uninsurance causes higher mortality. Moron.
December 14th, 2009 at 5:39 pm
Troll Supervisor is lying about Mixner, but that’s because he wants to disrupt the thread.
December 14th, 2009 at 5:39 pm
Platypus, you’re mixing some ideas. It’s perfectly possible that the uninsured get health care at levels similar to the levels they would get if they were otherwise insured, and that health care does in fact kill lots of people, reducing some of the benefit of having health insurance. Protecting one’s health isn’t the point of health insurance, or at least not the only point; health insurance, for those with assets, protects those assets.
December 14th, 2009 at 5:41 pm
fred, that’s right. Also, some studies suggest that the connection between poor health and and low income are stronger in the UK and Canada than they are in the US. It seems to me that Ezra and Matt should answer why they want us to move toward a system in which income is more important to health outcomes rather than less. (And, no, good intentions don’t count.)
December 14th, 2009 at 5:44 pm
Re: appendicitis because he was able to walk. That night, his appendix burst, nearly killing him. Instead of a routine appendectomy, he spent ten days in the ICU being treated for sepsis and then got the appendectomy. And the extra cost was $120K.
This highlights exceptional financial stupidity on the part of the hospital. Had they done an appendectomy sooner they would have taken a much smaller loss. If accountants are making these decisions they must be the ones who graduated at the bottom of their class.
By the way does anyone else think that Al and Mixner would also dispute that smoking results in a lower life expectancy or that seat belt use reduces automiotive fatalities if the GOP favored smoking and hated seat belts?
December 14th, 2009 at 5:47 pm
Lack of health insurance may be real. But unlike the left wing nut jobs I do not believe that the lack of insurance is man-made. Neither is death due to lack of health insurance.
It’s all a scheme by Al Gore to make a billion dollars off his investment, and get one more Nobel Prize for making yet another Power Point presentation.
December 14th, 2009 at 5:47 pm
Shorter Platypus: Ignore the evidence and try to answer the question with a thought experiment.
And even your thought experiment doesn’t make any sense. How come Britain has a higher cancer mortality rate than the US even though everyone in Britain is “insured” for cancer treatment?
December 14th, 2009 at 5:48 pm
I assume the point was not that “death causes uninsurance,” which would be absurd, but that factors that lead to people lacking insurance are likely to be independently associated with worse health and mortality outcomes.
December 14th, 2009 at 5:51 pm
“It’s well known that people without health insurance don’t get the same quality of health care in this country, but I would have thought that this group of patients would be the least vulnerable.”
Perhaps they’re less vulnerable to getting poor treatment in the ER when show up that day, but they’re still going to be more likely to die because of their history of not getting the health care they need for the years before that.
December 14th, 2009 at 5:54 pm
Thomas-
I don’t doubt that some uninsured get health care at levels comparable to theose they would get if otherwise insured. However, as a rule, the uninsured use less health care than the insured and delay seeking diagnosis and treatment for disease. It is this fraction, which represents millions of Amerians, that is at risk for poor outcomes, including death.
I agree that there is a real risk of iatrogenic death that would be prevented by avoiding health care. However, modern medicine has improved to the point where the odds favour seeking medical attention rather than avoiding it when faced with a serious disease.
Finally, for those fortunate enough to have assets, health insurance may protect their assets (but watch out for co pays, maximum benefits, pre-existing conditions, etc). However, for the poor and near poor, there are no significant assets to protect and protecting one’s health is the primary point of insurance.
December 14th, 2009 at 5:54 pm
Platypus,
… individuals who died because they had health insurance and received unnecessary fatal treatment at the hands of physicians. Where are those stories?
Right here, for example. Adverse drug events (incorrect dosage, harmful drug interactions, allergies, etc.) are estimated to cause almost 800,000 deaths and injuries in the U.S. every year.
December 14th, 2009 at 5:54 pm
You know, lots of people take a chunk out of Mixner – he’s like the village bicycle in that way – but when the “Not as Stupid” guys shows up and really pisses him off, and he lashes out blindly, he’s worried about me.
I love that.
December 14th, 2009 at 5:55 pm
Unemployment is also strongly associated with higher rates of mortality. Is it now perfectly fair to suggest that anyone voting for or supporting any bill likely to be a job-killer (such as, say, the one in question, or even more clearly cap-and-trade or minimum wage hikes) is a mass-murdering sociopath?
Also, how many will die because of the medicare cuts in the current bill?
December 14th, 2009 at 6:02 pm
platypus, It isn’t enough to say that the uninsured get less care than the insured. You have to show that the demographically-similar uninsured get less care than their counterparts who are insured. So, that poor uneducated slob who doesn’t trust doctors? Yeah, you have to show that health insurance makes a difference to him. And the fact that Medicare didnt’ make a difference historically, and the data doesn’t show any discontinuities in mortality today at age 65 suggests that may be difficult.
Yes, poor people have less need for health insurance. (That is what you mean, right?) And on maximum benefits: think of who that “reform” helps. Health care providers and rich people.
December 14th, 2009 at 6:02 pm
Adverse selection. The people in the US who get cancer and are uninsured are never treated for it, thus they’re never counted among the people who died after being treated for cancer, which is how those cancer longevity rates are calculated.
December 14th, 2009 at 6:03 pm
Jeff, that’s not fair. Medicare spending can be cut, because most of it is wasted. (Don’t think too hard.) And job killing is ok if you have good intentions and grand theories.
December 14th, 2009 at 6:05 pm
Clearly the causation arrow points in the other direction and death causes uninsurance.
Really, the only two options are “uninsurance causes death” and “death causes uninsurance”?
December 14th, 2009 at 6:07 pm
Did you come here from Red State or Bizzarro World? It’s so weird to read conservatives on issues like healthcare. It’s like reading dispatches from an alternate universe where health care is for the healthy and unneeded by the sick, where the richer you are, the more you need coverage for catastrophic health care events, where voting against something because it includes a provision you yourself were advocating not three months earlier is a “principled stand”, and so on. Is Superman a villain in your universe?
December 14th, 2009 at 6:08 pm
Thomas-
Regarding demographically similar uninsured vs insured. Just one example – There have been several studies of breast cancer survival in Canada vs. the US that have shown that there is a “survival advantage” (i.e., you live longer) for Canadians vs. the US. This advantage is strongest when comparing lowest social economic status women in Canada vs the States. (See Int J Epidemiol. 2009 Dec;38(6):1543-51 for a recent meta-analysis).
December 14th, 2009 at 6:15 pm
The people in the US who get cancer and are uninsured are never treated for it, …
This premise is false.
… thus they’re never counted among the people who died after being treated for cancer, which is how those cancer longevity rates are calculated.
I said cancer mortality, not cancer longevity. But since your premise is false anyway, it doesn’t really matter.
December 14th, 2009 at 6:18 pm
Oh, I’m sorry, I thought you were making a non-stupid claim. My bad. To reply to your specific point, then, no – far more people die of cancer in the US than die of it in the UK or Canada.
December 14th, 2009 at 6:27 pm
How come Britain has a higher cancer mortality rate than the US even though everyone in Britain is “insured” for cancer treatment?
Because their statistics count 100% of the population while ours count only the more limited percentage of the American population which gets treatment. If you never get treated for cancer by a doctor and then die, you don’t get included in the mortality rate stats, and so we achieve a seemingly “higher” result by throwing out our worst cases.
December 14th, 2009 at 6:27 pm
Oh, I’m sorry, I thought you were making a non-stupid claim.
Oh, I’m sorry, I thought you weren’t a total moron.
To reply to your specific point, then, no – far more people die of cancer in the US than die of it in the UK or Canada.
Huh? So what? How is this a “reply” to my point? Go back and read the post again, more carefully this time.
December 14th, 2009 at 6:32 pm
Because their statistics count 100% of the population while ours count only the more limited percentage of the American population which gets treatment.
Wrong, wrong, wrong. Cancer mortality rates are measured in deaths per unit of population. Not per unit of “population which gets treatment.”
December 14th, 2009 at 6:35 pm
Chet, wealthy people generally have more insurance on their cars and more insurance on their lives, and not because they are more likely to have car accidents or are more likely to die. If you’re wealthy and need medical care, you’re going to pay one way or another, which isn’t true if you’re poor. Health insurance ensures that the care will be paid, not that it will be received. We shouldn’t be surprised that lots of people don’t choose to pay thousands of dollars for health insurance, and we shouldn’t be confident that taking thousands of dollars from them for health insurance will make them better off. (It will make health care providers better off–they’ll know that they’ll be paid.)
December 14th, 2009 at 6:37 pm
Enrolling a uninsured person in a government run health insurance program and handing that person a health insurance card does not necessarily result in better health care.
The essence of the liberal argument is that if government redistributes the health care wealth, we’ll all be a little bit healthier.
That is by no means intuitive.
December 14th, 2009 at 6:39 pm
US Cancer Care Is Number One
The results for the U.S. include both insured and uninsured patients.
December 14th, 2009 at 6:52 pm
Health care actually doesn’t ensure that at all, which is much of the problem.
Again – only the people who get treated for it can be measured for it, hence the disparity. It’s an adverse selection effect, not evidence that we can cure more people of cancer by treating fewer of them and bankrupting others.
It’s eminently intuitive, similar to how it’s intuitive that if you have an island where 10 of the people have 90% of the food, taking a lot of that food and handing some of it out to the other 90 people will do much to alleviate starvation.
At least, it’s intuitive if you believe that health care is something that, when the sick get it, it probably makes them better. I understand that in conservative Bizzarro World, that may not be true.
December 14th, 2009 at 7:08 pm
Mixner-
I’ve leave attacking the underlying premise of your latest post – that the US always has the best outcomes – and just point out that your reply didn’t address the fundamental question. Can the US improve its own survival rates and other health care outcome measures by extending reliable and effective health insurance to millions of currently uninsured?
I would also put more weight to papers you cite if they were actually peered reviewed rather than in house publications.
December 14th, 2009 at 7:13 pm
Again – only the people who get treated for it can be measured for it
Again – you’re just spouting nonsense. Cancer mortality means death from cancer, period. It includes all cases in which cancer was determined to be the cause of death, whether the victim received medical treatment for the cancer prior to his death or not. You just don’t know what you’re talking about.
December 14th, 2009 at 7:19 pm
When I was diagnosed with cancer, I thought it was a death sentence, or that I’d need to sell my home or cash in my retirement to pay for treatment.
Then a friend asked me: have you tried standing in the hospital parking lot Standing in the hospital parking lot means you get a high dose of Healtheon, the colorless, odorless vapor that hospitals emit while making people better. So I did, and within days, I was completely cancer-free!
December 14th, 2009 at 7:21 pm
Mixner-
One more thought – before you get too excited about the National Center for Policy Analysis in house study you cited, you might want to read some surveys in peer reviewed journals. Take this one, for example – a recent Lancet Oncology survey (Lancet Oncol. 2008, 9:730-56) that found that the 5 year survivals for colo-rectal cancer and breast cancer were best in Hawaii (the state with the lowest percentage of insured in the US) and that Cuba (yes, Cuba) had even better 5 year survival figures for these cancers than the States.
December 14th, 2009 at 7:26 pm
I’ve leave attacking the underlying premise of your latest post – that the US always has the best outcomes –
That is not the “premise” of my post, and nowhere have I claimed or suggested that the US “always has the best outcomes.” You seem incapable of understanding even plain English sentences.
and just point out that your reply didn’t address the fundamental question. Can the US improve its own survival rates and other health care outcome measures by extending reliable and effective health insurance to millions of currently uninsured?
Since you didn’t ask that question, the lack of a reply to it isn’t terribly surprising. I certainly don’t think it’s “the fundamental question.” Perhaps the US could “improve its own survival rates and other health care outcome measures by extending reliable and effective health insurance to millions of currently uninsured,” but only at a price it’s currently unwilling to pay. Perhaps alternative policies (anti-smoking policies, for example) could achieve equivalent health and mortality improvements at lower cost. You need to stop obsessing over “health insurance” and think about the relationship between public policy and health more broadly.
I would also put more weight to papers you cite if they were actually peered reviewed rather than in house publications
The cancer data is from Lancet Oncology, perhaps the most prestigious oncology journal in the world. And it’s peer-reviewed.
December 14th, 2009 at 7:26 pm
[...] the original post: Matthew Yglesias » Health Insurance and Death Tagged as: cancer, Finance, food, health, history, mortality, Security, sports No Comments [...]
December 14th, 2009 at 7:28 pm
that found that the 5 year survivals for colo-rectal cancer and breast cancer were best in Hawaii (the state with the lowest percentage of insured in the US) and that Cuba (yes, Cuba) had even better 5 year survival figures for these cancers than the States
Are you familiar with the term “cherry-picking?”
December 14th, 2009 at 7:40 pm
Speaking of cherry-picking:
December 14th, 2009 at 7:48 pm
Yes, you’ve cherry-picked a comment from a blog by an anonymous author who may have no clue what he’s talking about. That sure is persuasive.
December 14th, 2009 at 8:01 pm
Jd-
Yes, I’ve familiar with cherry picking. I suggest that you read the article and judge for yourself how bad my cherry picking is. The survey only covered three cancers: breast, colo-rectal and prostate. Hawaii was #1 of the US States surveyed for the breast and colo-rectal, but 10th for prostate. Overall, that was the best performance by far for any individual state within the US. Cuba was 1st of 32 countries surveyed for breast and colo-rectal survival, and 8th out of 32 for prostate. The best performance overall for any country in the survey.
December 14th, 2009 at 8:02 pm
Shorter Mixner: I have no clue what I’m talking about.
December 14th, 2009 at 9:22 pm
For the record, guys, I’m pretty sure death really does cause uninsurance. 100% of the people I’ve known who have died stopped having health insurance after they died.
December 14th, 2009 at 9:46 pm
[...] this link: Matthew Yglesias » Health Insurance and Death a-number-, actually-attempts, charles, charles-lane, engage-on-the, even-though, health, [...]
December 14th, 2009 at 10:29 pm
Again I say, MY was perfectly fine with these hundreds of thousands of deaths when the obstacle to preventing them was accepting no funding of abortion coverage.
December 15th, 2009 at 12:04 am
There is only one other person, besides me, who knows how funny it is that Mixner thinks I’m the “Not as stupid” guy.
He he he. I wonder if it’s a regular?
December 15th, 2009 at 1:36 pm
[...] Yglesias responds to Cannon [...]
December 15th, 2009 at 5:35 pm
Cancer in america when one is uninsured is often treated by extensive networks of privately-funded hospitals devoted to cancer research and treatment, like the MD Anderson Cancer Center in Texas.
It’s the insured people who hit lifetime limits and then don’t pursue the private-charity routes noted above who die.