Is Kansas really an Obamacare outlier? Here’s why it might not be

For all the controversy that continues to surround the Affordable Care Act, one thing seems pretty clear: After the reform law was passed, uninsured rates declined across the country.

Clear, that is, except here in Kansas, where the headlines have told a different story. “Poll: Kansas Uninsured Rate Goes Up By 40 Percent,” reported KCUR, the Kansas City public radio station, in August 2014. “Poll: Kansas only state whose uninsured rate increased in 2014,” a Feb. 25 Topeka Capital-Journal headline read. The most recent news was less bad, but still not very good: “Change in Kansas’s uninsured rate lags other states,” the nonprofit Kansas Health Institute news service reported on Aug. 11.

None of these headlines were incorrect, and the accompanying articles often made important points—for example, that Kansas officials’ refusal to expand Medicaid, as the law envisioned, has limited the gains in the insured rate. But this grim picture of Kansas as a unique negative outlier is just one way to look at the story. In fact, the numbers deserve another look—and while it’s important to note that we don’t have the whole story yet, it’s possible that the state’s uninsured rate has been improving more significantly than press coverage has indicated.

The discussion about Kansas’ uninsured rate in the Obamacare era dates to the spring of 2014, when Rep. Tim Huelskamp, a Tea Party Republican from the state’s 1st District, asserted that more Kansans were uninsured after the ACA took effect than before. The Washington Post, several local outlets, and CJR all soon debunked Huelskamp’s claim. But that summer, a report from the Gallup Healthways survey, the source for much of the media coverage about uninsured rates in the state, seemed to confirm it. In its midyear survey, Gallup found that the state’s uninsured rate had risen a shocking five percentage points since 2013—from 12.5 percent to 17.6 percent. This unexpected result didn’t quite create a firestorm of media attention, but it still drew more press coverage than any other recent piece of news about Kansas’ uninsured.

Sign up for weekly emails from the United States Project

Six months later, Gallup released its full-year results for 2014, showing a 14.4 percent uninsured rate in Kansas. This result drew less coverage—and what coverage there was emphasized the increase over 2013, rather than the decline from the midyear figure.

Then, on Aug. 10, Gallup released the 2015 midyear results, showing that Kansas’ uninsured rate had hit a new low in the survey of 11.3 percent—a startling decline from the mid-2014 figure that drew a wave of coverage. This seems like good news, but I could find hardly any mention of the finding in the state’s media, except for a report from the KHI news service, which focused not on the sharp turnaround from 2014 but the modest decline from 2013.

Comparing the new results to 2013 makes sense, because that’s the year before the insurance exchanges set up by the law opened for business. But it also raises questions for Kansas, where that year was something of an anomaly in Gallup’s findings: lower than results for the four preceding years, which ranged from 14.1 to 16.2 percent, and the following year, 2014, when it was 14.4 percent. As Dan Witters of Gallup Healthways told CJR last year, this could be one explanation for the magnitude of the apparent spike in the midyear 2014 result—the 2013 number might have too low to begin with. If that’s the case, and the true baseline was higher, then the latest finding, of an 11.3 percent uninsured rate, would represent a more substantial decline.

Of course, that finding might be off to some degree, too. Still, as Witters points out, we have more to go on than just the last six months. Though Gallup doesn’t release results just for the latter six months of the year, if the 2014 findings went from 17.6 percent at mid-year to 14.4 percent by the end of the year, then the figure for the last six months of the year must have been “in the low 11s,” Witters says. Thus we now have a full year’s sample showing a result not much above 11 percent.

And Gallup isn’t the only survey showing a recent decline. The Centers for Disease Control and Prevention’s National Health Interview Survey release from June shows the Kansas uninsured rate all the way down at 9 percent for 2014, after having been at 12.7 percent in 2013, 12.9 percent in 2012, and 15.2 percent in 2011. I haven’t seen any media coverage of this poll result.

If accurate, the latest findings on Kansas from Gallup and the CDC do seem to track with what is happening in other states. As Gallup has found, in states like Kansas that have not expanded Medicaid or set up state exchanges, insurance rolls did not grow as quickly or as dramatically as in states that did take those steps—but they have grown nonetheless. “The states that did both got out of the gates a lot faster than the states that didn’t,” Witters says. “That trend has now leveled off; the rate of decline is now the same for both groups. Kansas is kind of a good illustration of that.” He added: “Whatever we were picking up in the beginning of 2014, that’s now in the rearview mirror.”

To be clear, I’ve looked at some of the data here from an angle that highlights a potential decline in the state’s uninsured rate—a counterpoint to press coverage that has often come from the opposite perspective. Meanwhile, at least one expert observer is counseling caution. “With Gallup, the amount of change in the numbers is within the margin of error,” KHI senior analyst Scott Brunner wrote in an email. (The margin of error for the 2015 midyear result, Witters says, was 2.5 percentage points.) Brunner said he couldn’t speak to the CDC survey without knowing the sample size for Kansas. (I posed this question and others about the survey to the CDC, but was told that the survey expert was out of town.)

But, Brunner promised, more data is on the way. “What we are waiting for is the American Community Survey and Current Population Survey results in September,” he wrote. “Those are the sources we have relied on for our estimates. The sample size in Kansas is large enough to make clear comparisons over time.”

When that data comes out, it won’t tell us everything we want to know about the current moment: the ACS operates on a lag, so the soon-to-be-released figure will be for 2014. Still, it may tell us a lot about what happened in Kansas after the Affordable Care Act went into full effect. Whether the number is up, down, or flat, here’s hoping reporters in the state are paying attention.

Has America ever needed a media watchdog more than now? Help us by joining CJR today.

Deron Lee is CJR’s correspondent for Iowa, Missouri, Kansas, and Nebraska. A writer and copy editor who has spent nine years with the National Journal Group, he has also contributed to The Hotline and the Lawrence Journal-World. He lives in the Kansas City area. Follow him on Twitter at @deron_lee.