4 ms·

Eh, the article does mention that point: "It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine." So health insu

Eh, the article does mention that point: "It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine." So health insurance may be useful, but health insurance for hospital treatment isn't. I don't think that conclusion is much weaker.

gumby 5y ago
> Eh, the article does mention that point: "It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine." What you quote is a mischaracterization by Hanson. From the study (from the GP comment): > the study was designed to be powered to detect a change in the hospitalization rate, not necessarily changes in health outcomes. The interesting questions are: which causes a better gross outcome (what Hanson discusses) and, if outcomes are the same which costs less. If insurance means there is more primary care and that causes less hospitalization then you clearly save money and have a better QoL. Insurance also handles a different class of cases, e.g. you are hit by a car. These are outlier cases (for which insurance makes the most sense and implies no moral hazard). This is ignored by Hanson who considers only baseline chronic conditions like cholesterol levels. The study could imply that insurance coverage could be modified (covering less routine care), though more work would be needed. I’m dubious, but that could be the right thing.
dragonwriter 5y ago
> Eh, the article does mention that point: "It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine It mentions only one of the three points. > So health insurance may be useful, but health insurance for hospital treatment isn't. No, that’s not justified by the study, because as the study says, it only was designed with the power to detect changes in hospitalization rates over the term of the study. not health outcomes, which would require a much larger study. The study is useful if you are trying to use pre-policy hospitalization rates to predict the usage under (and therefore cost of) a public hospitalization-only insurance policy over the first several years, but it is very much not useful in evaluating whether that or any other insurance policy is useful.

gumby 5y ago

> Eh, the article does mention that point: "It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine." What you quote is a mischaracterization by Hanson. From the study (from the GP comment): > the study was designed to be powered to detect a change in the hospitalization rate, not necessarily changes in health outcomes. The interesting questions are: which causes a better gross outcome (what Hanson discusses) and, if outcomes are the same which costs less. If insurance means there is more primary care and that causes less hospitalization then you clearly save money and have a better QoL. Insurance also handles a different class of cases, e.g. you are hit by a car. These are outlier cases (for which insurance makes the most sense and implies no moral hazard). This is ignored by Hanson who considers only baseline chronic conditions like cholesterol levels. The study could imply that insurance coverage could be modified (covering less routine care), though more work would be needed. I’m dubious, but that could be the right thing.

dragonwriter 5y ago

> Eh, the article does mention that point: "It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine It mentions only one of the three points. > So health insurance may be useful, but health insurance for hospital treatment isn't. No, that’s not justified by the study, because as the study says, it only was designed with the power to detect changes in hospitalization rates over the term of the study. not health outcomes, which would require a much larger study. The study is useful if you are trying to use pre-policy hospitalization rates to predict the usage under (and therefore cost of) a public hospitalization-only insurance policy over the first several years, but it is very much not useful in evaluating whether that or any other insurance policy is useful.