Does losing health insurance make you more likely to die? Several years ago, while I was writing my book on health care, I spent a lot of time on that question. I read the two highly publicized reports by the Institute of Medicine (IOM): Care Without Coverage: Too Little, Too Late and Consequences of Unnsurance. I read through the underlying research. And I spoke to as many experts as I could find.

The IOM had concluded that lack of health insurance led to 18,000 deaths a year--a figure that, some experts suggested, was a bit arbitrary. Isolating the effects of uninsurance was just too difficult to arrive at such a specific number. But, plainly, lack of health insurance left people in worse health. Many of them, most likely in the thousands, died every year. The cause-and-effect was certainly consistent with my own observations as a reporter, some of which I'd written about over the years. And so while I didn't make a habit of citing the 18,000 figure, I would (and still do) frequently cite the IOM report, explaining that "thousands" of people die every year because they can't get afford to get recommended medical care.

As you may have heard by now, Megan McArdle has written an article for the Atlantic Monthly questioning that assumption. She's not foreclosing the possibility that uninsurance might make people more likely to die, but she thinks the data hardly support that conclusion. It's not the first time she, or other critics of health care reform, have raised questions about the relationship between health insurance and health outcomes. Since that article's publication, I've been hoping to do what I did a few years ago--sit down with her article, the underlying data, and the compare those with the latest studies on the subject. But I've been too preoccupied with other assignments and tasks.

Ezra Klein, fortunately, has done the due diligence, publishing not one but two lengthy responses to McArdle. He's also posted a guest blog item from the Urban Institute's Stan Dorn, who was the lead researcher on that original IOM study. Austin Frakt has weighed in on this; so has J. Michael McWilliams (guest-posting on Frakt's blog). All of them think McArdle is simply wrong, on the merits. My reading is that they have the better of this argument, by far. But you should read the items, including the rejoinders McArdle posted on her blog, and decide for yourself.

Peter Harbage is a Washington DC-based health policy analyst who worked with both Senator John Edwards and Governor Arnold Schwarzenegger on their individual mandate proposals. He has also worked at both the Center for American Progress and the New America Foundation.

With the release of the Congressional Budget Office report on the Senate Finance health bill, there has been significant concern about how health reform may not cover all of the uninsured.  From leaders in the progressive movement (read Jon's list of top health reform issues) to Karen Ignagni's October 11 cover memo on the infamous PwC "report", people have been talking about the size of the penalty for families failing to purchase insurance.  It's a good question, but it misses the point.  Getting everyone covered depends on a three step process:  coverage must be made affordable, coverage must be made easily available, and once that is achieved, then you can have an effective mandate and penalty.

It is this second step--making coverage easy to get--that has gotten too little attention. Six out of ten children eligible for Medicaid and CHIP (basically free coverage) are estimated to be eligible and not enrolled. The forms involved can be overly complex and the system can be difficult to navigate. In fact, today's health insurance system is seemingly designed to keep you out of insurance. Job change, marriage, change in age, all mean that you could lose your current insurance.