There's been a great deal of blog-buzz recently about the new Johns Hopkins study. It purports to show that a sizeable chuck of "uninsureds' earn too much to qualify for public assistance, but nevertheless cannot afford insurance premiums.
Let me state the obvious at the outset: the study itself is fatally flawed, and its presentation so full of holes that I want a Reuben. About the only redeeming feature I could find was that it eschewed the common mistake of conflating health insurance with health care.
Okay, now to brass tacks:
According to the “study,” well over half of those folks who identify themselves as uninsured claim that they make too much money to qualify for the various public assistance programs, but not enough to afford insurance premiums. As my co-blogger, Bob Vineyard, reminded me, this seems to fly directly in the face of many previous such studies, which indicated that about 40% of the uninsured make in excess of $50,000 a year. Granted, this isn’t exactly Bill Gates territory, but it certainly enough to be able to afford some basic, catastrophic coverage.
Second, the type of plans for which these folks are supposedly shopping is never revealed. Are they looking for (expensive) first-dollars type plans, with office co-pays and drug cards? Or are they looking at, for example, high deductible plans which offer an affordable safety net? The study never says (although I’d hazard a guess). The point is, absent some critical illness, most folks will qualify for at least some basic level of coverage.
Third, it’s never made clear (if they bothered to ascertain this at all) how many of these fine folks have cable TV, internet access, or cell phones. It would be nice to know this: assuming that one has a finite income, the choices one makes as to how it’s to be spent is certainly relevant to this discussion. Absent this type of information, the study itself is meaningless.
But Professor, you may object, how can you say this when everyone knows how expensive health insurance is?
The “study” itself gives us the answer: “(W)e do not know what the benefit packages or cost sharing are for the policies for which we have premium data." That is, they have a set of values, and a set of variables, and there is absolutely no effort made to correlate the two. In other words, they just made the whole thing up.
Much ado about nothing, indeed.