Insuring the Uninsured and the Distribution of Mortality

by Will Wilkinson on February 14, 2010

I’m firmly in the McArdle/Cowen/Cannon camp in the argument over whether insuring the currently uninsured will have any net effect on mortality within this class. But I think it’s important to emphasize more than others have done that

(a) Extending insurance to the currently uninsured will have no net effect on mortality,

and

(b) Extending insurance to the currently uninsured will save no lives,

are different propositions, and that the truth of (a) does not imply the truth of (b).

Extending insurance to the currently uninsured would save some people’s lives while killing others.

The people most likely to be saved are older, unhealthy people who are now uninsured because they cannot afford insurance. These people would get treatment they need but can’t afford. The people most likely to be killed are people who can now afford but do not want insurance because they are young, healthy and are able to self-insure (or are in a position to be insured by their relatively family). If compelled to buy a health plan, these people would tend to consume more medical care than they would otherwise, exposing them to the very real risks of unnecessary treatment.

I think it’s reasonable to guess that the people most likely to be saved are relatively poor (but not poor enough for Medicaid) and just below the age of Medicare eligibility. The people most likely to be killed are relatively wealthy and in their twenties.

So, while insuring the currently uninsured might come out as a wash in terms of lives saved versus lives lost (or even if there is a small net decrease in mortality), the way one would expect this to affect the distribution of mortality could lead to a decrease in QALYs, which is worth taking into account.

Of course, I don’t know that this is so. This is conjecture, but I think it makes sense. Generally, I’d like to see more about the distributive effects of the Democrats’ HCR plans.

[Update: Links added above for those unaware of the ongoing debate. See also links at end of Megan's linked post.]

  • alexmadison
    the distribution is never fair

    male yeast infection treatment
  • robinbg
    It is tragic that we as a great nation can't figure out a solution to the uninsured situation. Other countries have already done it and it seems the same"pride" that got us to the top will ironically bring us to our knees if we don't embrace change. Until then, we have to take advantage of existing programs. There are at least two of them which I feel everyone needs to know about. One is the "Patient Assistance
    Program" and two are free discount prescription cards that exist. I know of a site (www.medpap.ws)which goes over both .I hope this helps until we as a people truly realize we are only as strong as our weakest link.
  • Sarah
    The US is not the only country to look at in this debate!
    Let's see what happened to mortality in countries that expanded health insurance coverage.
    Chile has drastically lowered infant mortality since adopting its NHS in 1952.
    http://content.healthaffairs.org/cgi/reprint/26/2/458.pdf

    There's actually been quite a bit of study of the effect of medical care on mortality.
    http://www.cmj.hr/1999/40/2/10234053.pdf
    It's difficult to distinguish the effect of medical care (as opposed to other factors) in reduced mortality, but there seems to be some evidence for it.

    Some public health measures -- like immunization -- have demonstrably positive health outcomes.
    http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.0050211

    Also, I'm not sure if this has been cited, but this study
    http://www.earlylearning.ubc.ca/documents/development-health/Franks%20P,%201993.pdf
    shows the uninsured die more even after correcting for factors like income, education, race, and lifestyle factors like obesity, smoking, and alcohol consumption. When Megan McArdle says "the uninsured are different from the insured," I assumed she meant due to factors like these. Since the effect of health insurance remains independent of those factors, I'm not sure what additional confounding differences there might be.

    Using this study, with 45.7 million uninsured, and a hazard ratio of 1.25 for being uninsured (controlled for other factors), and 2.4 million deaths, and a population of 305 million, I do the simple back-of-the-envelope calculation and get 167,000 deaths among the uninsured that would have been prevented if the uninsured had the same likelihood of mortality as the insured. This is odd, because it's higher than the highest liberal estimates, but it's just based off current census numbers and the study I cited.
  • lemmycaution
    You seem to be making up the benefit to the young of not having healthcare. Medical interventions can kill you, but they are going to kill the weak more often than the strong.

    To me the natural consequence of strong Healthcare skepticism is a low-cost low-intervention single payer healthcare. Think England rather than Canada. The English healthcare system kind of scares me though.
  • lxm
    Here's the other camp

    http://voices.washingtonpost.com/ezra-klein/2010/02/all_you_ever_wanted_to_know_ab_1.html

    Sometimes people just really really really want to believe.

  • johndewey
    "The people most likely to be saved are older, unhealthy people who are now uninsured because they cannot afford insurance. These people would get treatment they need but can’t afford. "

    If they cannot afford it now, then the only way they will afford such care under Obamacare is for someone else to pay for it.

    We could probably make a case that "spending an extra $90 billion a year to provide new and safer automobiles" to millions of poor Americans will save/extend x number of lives. Or that "spending an extra $90 billion a year to provide three nutritious meals a day" to millions of poor children will save/extend x number of lives.

    The problem, though, is that the money to accomplish any of those things must be taken from some individuals are not willing to part with it voluntarily. Or, in the case of loading up the nation with trillions in public debt, it will be taking wealth from a minority of citizens who are not yet born and cannot even protest.

    When did extending/savings one's life stop being the responsibility of individuals and parents and start being the responsibility of the productive and the unborn?

  • anon
    Will,

    I'm curious as to your thoughts on the net effect that extending insurance will have on total happiness?

    The happiness question is more difficult but also more interesting/important than the question of how many lives will be "saved".
  • Re: "The people most likely to be killed are people who can now afford but do not want insurance because they are young, healthy and are able to self-insure (or are in a position to be insured by their relatively family). If compelled to buy a health plan [snip]"

    Within the list which follows your 'because' might you allow the addition of 'they know that if they come down with pancreatic cancer, their society will provide them with expensive treatment anyway'?

    It seems important to acknowledge many of the young uninsured are not 'self-insuring'--they're 'free-riding'.

    I'd further scratch your 'If compelled to buy a health plan' and replace it with 'If compelled to pay for the implicit health plan already provided to them by their indulgent social order...'
  • dgauthier
    Questions from an undereducated, regular Joe. Why is everybody seeming to equate the availability of heath insurance with the availability of health care?

    I live in Louisiana and I'm not well-travelled but here we have a small network of state-run charity hospitals, clinics and mental health clinics. I've had to avail myself to their services on a few occasions when I changed jobs. The wait for service is interminable and the facilities run-down, but I and my family got the care we needed. My late wife also had no trouble getting medication for her bipolar disorder.

    Is such a charity system unavailable elsewhere?

    Shouldn't some studies be focused on the mortality of those who have no available health care, as well as no insurance?
  • johndewey
    My brother and two other relatives have used the charity hospitals in Louisiana several times in the past few years. Yes, the waits were long, but they received adequate treatment.

    Most hospitals in this nation - about 60 percent - have applied for and received the status of "not-for-profit". By law, they are required to treat the indigent. But the extent to which they are providing such care is controversial.
  • Sarah
    They don't all provide care.

    I'm most familiar with the University of Chicago hospitals, which often (anecdotally; I don't know statistics) send indigent emergency-room patients to the county hospital instead of treating them. Of course, the added travel time can be risky for the patient.
  • Craig McGillivary
    I suspect that Hillary Clinton probably had a bigger effect on public health by banning smoking in the Whitehouse than this bill will.
  • Eric Auld
    Will--glad you're back blogging! Regarding the claim about unnecessary medical care, what you said seems intuitively implausible. Can you support it?

    Once again, glad to see you blogging. I was looking through a lot of great quotes I'd saved recently and remembering how good your blog is.
  • lark
    Wow.

    Here we have studies from Harvard and the Natl Academies finding tens of thousands of deaths from lack of health insurance and your response is an utterly inane finding that there will be thousands killed by medical treatment if they carry insurance. No studies backing you up, of course. No one would waste the money for studies on something so utterly and offensively dumb.

    Clue: You "think it's reasonable to guess" isn't good enough.

    What you are doing here is arguing against the clear benefit of universal health coverage by making up data. You are making up this data because you want to resist the conclusion from the most respected studies, that tens of thousands of Americans are killed every year from lack of health insurance.

    If this is the best you can do, your ideology is showing. And it's ugly, ugly, ugly. Why? Because you are using stupid ideological arguments to justify the current slaughter of Americans, due to lack of health insurance.

    How do I know about this slaughter? Two friends, dude. Two. Two who were far better people than you are.

    I bet you don't post my comment.




  • dfstearns
    Okay, it sucks you have friends who have died, all the more so if they were better people than Will is. But people dying from lack of medical care is not "slaughter" any more than gravity is tyranny. http://www.sacred-texts.com/bud/btg/btg85.htm
  • I've added links to the first sentence of the post to help you catch up on the ongoing debate. In particular please read through McArdle's recent posts in which she addresses the various studies. I'm very sorry to hear about your friends.
  • I don't think this post could feature a much bigger gap between claims and supporting evidence.

    Take as given that there are unnecessary treatments, that young wealthy people have the lowest needs for medical care, and that medical mistakes cause people to die. This does not establish that young people are, as a group, likely to experience an increase in mortality as a result of being compelled to buy health insurance. After all, the lower need for healthcare among the young also lowers their chance of suffering medical errors. Of course young people can still benefit from routine checkups, physicals, consultation about exercise, diet, etc, none of which have high rates of fatal medical errors.

    Absent some evidence (whether qualitative or simply suggestive) that there are likely to be many accidental deaths from increased care, you've outline something that's just barely a statistical possibility, and presented it as if we need to be concerned.
  • Mark
    Is there something wrong with Disqus, or are you deleting my comments manually?
  • There's something wrong with Disqus. I'm getting your comment in my email, but then not finding it here when I click through. I don't delete comments. I'll try to see if it's getting marked spam.
  • Mark
    I didn't *think* you were deleting them, but I saw them briefly on the page before they disappeared, so I knew they were getting through... anyway, thanks for looking into it.
  • Found it! It seems I've been getting lots of false positives. Sorry.
  • Mark
    "The people most likely to be killed are people who can now afford but do not want insurance because they are young, healthy and are able to self-insure (or are in a position to be insured by their relatively family). If compelled to buy a health plan, these people would tend to consume more medical care than they would otherwise, exposing them to the very real risks of unecessary treatment."

    This is an astonishing claim. Do 20-somethings who have generous health benefits from their jobs have higher mortality than those who don't? And if so, shouldn't we then, as a matter of public policy, encourage young people to forego options to buy health coverage whenever possible? For that matter, shouldn't the ideal health reform bill *ban* young people from buying coverage? Think through the implications of your claim, for a minute.
  • My point is that overconsumption of a dangerous service is dangerous. There ought to be nothing astonishing about suggesting that young healthy people with (over)generous health benefits in fact do worse than young healthy people with less-generous high-deductible plans.
  • It ought to be astonishing because, for one thing, even "free" healthcare services still cost in terms of trips to the doctor, discomfort, general unpleasantness, etc. Fillings are "free" for most nine year-olds (parents pay the cost); but we would certainly be astonished to hear your average nine year-old begging her parents, "Please, please - can I go to the dentist again?!"
  • Mark
    But it is astonishing! I don't doubt that people with good health coverage consume more unnecessary care than those without, but by the same token, we've got to assume they consume more necessary care as well. How do you know that this comes out as a net harm, instead of a net benefit? Again, I ask: Do young people with generous coverage have higher mortality then their peers with little or no coverage?
  • Mark
    "The people most likely to be killed are people who can now afford but do not want insurance because they are young, healthy and are able to self-insure (or are in a position to be insured by their relatively family). If compelled to buy a health plan, these people would tend to consume more medical care than they would otherwise, exposing them to the very real risks of unecessary treatment."

    This is an astonishing claim. Do 20-somethings who have generous health benefits from their jobs have higher mortality than those who don't? And if so, shouldn't we then, as a matter of public policy, encourage young people to forego options to buy health coverage whenever possible? For that matter, shouldn't the ideal health reform bill *ban* young people from buying coverage? Think through the implications of your claim, for a minute.
  • Wha?? Who dies from unnecessary treatment?
  • Many, many people do. Many people undergo unnecessary treatments. Medical mistakes cause many deaths.
blog comments powered by Disqus

Previous post:

Next post: