Ready for more bogus stats about the uninsured? Well, Families USA has got some for you and the “news” media are reporting them just as if they came from a legitimate study:
One out of three Americans under 65 were without health insurance at some point during 2007 and 2008, according to a report released Wednesday.
These “findings” seem curiously at odds with the calculations of the Census Bureau:
The number of Americans without health insurance reported by Families USA is much higher than those reported by the U.S. Census Bureau. According to the census numbers, in 2007 there were 45.7 million uninsured Americans.
So, what does Families USA say about that discrepancy?
Families USA says those numbers tell only part of the story. The Census bureau counts only people who were uninsured for the full calendar year.
Those who have bothered to acquaint themselves with the facts will recognize this assertion as pure BS. As Sally Pipes puts it:
The [Census] Bureau counts anyone who went without insurance for any part of the previous year as “uninsured.” So, if you weren’t covered for just one day in 2007, you’re one of the 47 million.
But what do mere facts matter? The Families USA “study” isn’t really about facts. They have stretched the window from one to two years, and generally fudged the numbers.
But why? Hmm … could this be about the President’s upcoming health care summit?
Comments 2
There is a fairly straightforward reconciliation between this manipulation of the data by Families USA and the Census bureau. It doesn’t take an actuary (I happen to be one, however) to dissect the data. Families USA wanted to find a way to take the facts and make the numbers look worse somehow. So, yes, they looked at a two year period, and are able to say (and this is probably correct) that 86.5 million different persons under age 65 were without formal health insurance at some point within that 2 year period. The point, however, is that when they compare that figure with 262.3 million persons under age 65, they are mixing two years of a subset with an implicit single year of “exposure.” Using mid-points for the brackets of length of uninsured status, and calculating a weighted average, we arrive at an average period of uninsured status of 13.1 months. That is not important however, what is important is that the total number of “uninsured person months” during the two year period is 1,132.75X10^6, while the number of total person months exposed during the two year period is 6,295.2X10^6, for a ratio of uninsured months to total exposed months of roughly 18%. (The total exposure is the 262.3X10^6 population times 24 months each.) Guess what 18% of 262.3X10^6 is? 47.2 million (some rounding error). That looks suspiciously like the census bureau figure for the average, at any given time, of the number of uninsured persons.
Posted 31 Mar 2009 at 2:24 pm ¶Want more Families USA BS? Read this post by Linda Gorman about their claims about the cost-shift from the uninsured:
http://www.john-goodman-blog.com/families-usa-makes-strong-bid-for-worst-study-of-the-year-award/
Posted 18 Jun 2009 at 4:26 pm ¶Post a Comment