Health Care Data: Commence Nitpicking

Sep 10, 2009 3:42pm

President Obama this morning picked up on the latest Census Bureau report on uninsured Americans, noting that it says “the number of uninsured rose in 2008.” That's true, but not because of an increase in the uninsured rate – rather simply because the nation’s population is growing.

The bureau reports that 15.4 percent of Americans were uninsured in 2008, vs. 15.3 percent in 2007 – a difference that falls within its survey’s margin of sampling error. Its report says: “The percentage of people without health insurance in 2008 was not statistically different from 2007 at 15.4 percent. The number of uninsured increased to 46.3 million in 2008, from 45.7 million in 2007.”

I checked this today with Census spokesman Robert Bernstein. “The rate within the population did not change significantly," he told me. "The number did change, because total population increased. The number of insured people also rose, for the same reason.”

The bureau also says its measure apparently underestimates the true number of uninsureds, for various methodological reasons. But that shouldn't affect its time trend.

In historical data, the uninsured rate has ranged from a low of 12.9 percent in 1987 to a high of 15.8 percent, in 1998 and again in 2006. (This includes a change in estimation procedures that started in 1999. See p.59, Table C-1, here.) Joblessness likely is one factor; the recent low for uninsureds, 13.7 percent, came in 2000, when unemployment fell to just 4 percent.

Separately, while I’m picking nits, in his address last night, the president said "a strong majority of Americans still favor a public insurance option."

Not that I can see. In our own data in mid-August, 52 percent supported a public option; “strong” support was 33 percent, not near a majority. Support for a public option was a similar 55 percent in a recent CNN poll, 27 percent strong; and 53 percent in a CBS poll. Support was higher in the CBS poll, 60 percent, when the option was likened to Medicare. Its questions didn’t include a strength-of-support measure.

Worth noting, moreover, that the 52 percent in our mid-August data was down from 62 percent in June. And when we asked in June what people thought if a public option would put many private insurers out of business, as critics claim, support plummeted to 37 percent.

The problem with "strong" is semantic, but meaningful nonetheless. Broad support (if 52, 53 or 55 percent qualifies as broad) is not the same as strong support; width is not depth. And even if using "strong" for "broad," having support go from 62 to 37 percent with pushback, from 62 to 52 percent from June to August, and over 55 percent only when likening the proposal to the popular Medicare program, does not seem to fit the bill.

You are using an outdated version of Internet Explorer. Please click here to upgrade your browser in order to comment.
blog comments powered by Disqus