A while back the Bloviator and I had some discussions parsing out the 2002 uninsurance numbers. There was some controversy (that the two of us settled to our satisfaction, at least) about those numbers from the census bureau to do with how many of the 43 million it counted as uninsured were uninsured for the whole year. Now Health Affairs has published a Commonwealth Fund-sponsored article by Pamela Short and Deborah Grefe at Penn State that examines in great detail uninsurance between 1996 and 1999. While this data is of necessity a little old, you must remember that we were in an employment boom then–so things were as good as they were ever going to get for employment-based insurance in the modern economy–and also that things are worse now.  Still, onto the highlights.

The authors looked at large slice of the non-Medicare under-65 population which had approximately 225 million adults.  (It excluded immigrants, newborns and some others). Out of that 225 million number 84 million (37%) were uninsured at some time in the 4 years. Of those 84 million roughly 15 million (or 6% of the total) were more or less uninsured the whole time. As for the rest, the authors use a fairly complicated 6 way breakdown which I will grossly over-simplify into the fact that 50 million were uninsured for 5 months or more during that four year period. Roughly 32 million of that sub-group (64%) were uninsured for at least a year or more. If you are counting along at home that leaves another 20 million who had one or more short breaks in their coverage of less than 5 months.

So in my assessment, what’s new about this research? 

1) Well it’s usually assumed that at any one time 20 odd million are uninsured for a whole year (a little less than half the 43 odd million uninsured at any one time).  But if you take this rolling view rather than the snapshot, you have 15 million hard-core uninsured essentially for ever and another 32 million who’ve had a year or more uninsured in a four year period.  So rather than the 43 million oft-quoted snapshot number, some 57 million have been uninsured for more than a year in a four year period. These are the hard core uninsured and they measure nearly 25% of adults. And incidentally that is more people than voted for any one candidate in the 2000 election.

2) Counting this crudely, and making some assumptions, there seem to be three groups; one that is nowhere near getting insurance, One that is swinging between government programs like Medicaid, some employer based insurance and no insurance, and a smaller group that is cobbling together a patchwork of employer insurance, individually-bought insurance and uninsurance. The first two groups are the lower income ones.

The authors conclusions are that separate policy solutions are needed for each group–unless we have universal coverage.  That’s true in so far as it goes, but the authors know and (restrained by the terms of this data study) don’t state that the peverse dynamics of the individual insurance "market", the cost of COBRA coverage, and the difficulty of maintaining Medicaid coverage, all combine to make viable policy solutions targeted to sub-groups of the uninsured almost impossible to create.  The only actual options for universal insurance are :
a)some kind of employer-mandate, or
b)some kind of individual-mandate, both backed-up by government schemes either in terms of premium support for the poor or guaranteed insurance (e.g. Medicaid expansion). Or
c) of course single payer, Medicare for all.

There is clearly no political will for any of these reforms now. But perhaps if word gets out that not only is one in seven people uninsured now, but over one in three of us might be in this jam sometime in the next four years, that political will might become more apparent. One thing we do know: voluntary universal insurance is a fiction.