A Project of The Annenberg Public Policy Center

‘Millions’ Lost Insurance


Party Lines Red InsertThe Line: Millions of people have lost their health insurance and their doctors because of the Affordable Care Act.

The Party: Republican

President Obama gave ad-makers plenty of fodder last year when his promise — “If you like your health care plan, you can keep your health care plan” — clearly was proven false. We had said years earlier that Obama couldn’t make that promise to everyone, but the claim made headlines when Americans received cancellation notices for individual market plans that no longer met the law’s requirements.

Critics of the law now say millions lost their health insurance. But that’s misleading. Those individual market plans were discontinued, but policyholders weren’t denied coverage. And the question is, how many millions of insured Americans had plans canceled, and how does that compare with the millions of uninsured Americans who gained coverage under the law.

There is evidence that far more have gained coverage than had their policies canceled.

The conservative Americans for Prosperity has made the canceled policies a theme in its advertising. In one series of ads, a soft-spoken woman says: “Millions of people have lost their health insurance. Millions of people can’t see their own doctors.” That ad, which aired in February and March, targets Democratic senators in three states: Sens. Mark Udall in Colorado, Mary Landrieu in Louisiana and Mark Pryor in Arkansas. The ad aired against Sen. Kay Hagan in North Carolina in November, and it’s also been used to target a few House members.

Another AFP ad targeting Landrieu — and airing in January — said that “millions of Americans have lost their health care.”

It’s true that insurance companies discontinued health plans that had covered millions of people who had bought them directly rather than through an employer. That’s because those plans didn’t meet the coverage standards of the new law.

But those policyholders didn’t lose the ability to have insurance. In most cases, insurers offered them an alternative plan, though there were some instances of companies exiting the individual market altogether.

Whether offered an alternative or not, individuals could shop for insurance on the federal and state marketplaces, or through a broker or insurance carrier directly. Many were likely eligible for federal subsidies to help pay for insurance, resulting in better coverage and lower rates for some. But the specific plan they had was indeed discontinued. (More than half of those with canceled policies were likely to be eligible for federal assistance, according to Urban Institute research, and about 80 percent of all those buying plans on the exchanges are expected to qualify for subsidies, according to the Congressional Budget Office.)

How many individual market cancellations were there?

The most commonly used figure is 4.7 million, based on reporting by the Associated Press last December. But there’s reason to doubt the accuracy of that figure. An analysis of a more recent poll by researchers at the Urban Institute puts the figure at somewhere around 2.6 million.

An AP story that ran Dec. 26 said that “at least 4.7 million Americans received the cancellation notices,” and gave state-by-state figures for the “number of policies scheduled to be canceled.”

But the news agency didn’t say exactly how it arrived at the other figures that went into the 4.7 million total, making the reporting impossible for outsiders to verify. In three states, the figures appear to be inflated. Washington state’s insurance commissioner, for example, has publicly stated that the AP’s figure of 290,000 discontinued policies in that state is “inaccurate.” In a news release on his official website, Insurance Commissioner Michael Kreidler said that there were only 278,000 total in the individual market at the end of September. Recent reports by our fact-checking colleagues at Politifact.com and the Washington Post show the numbers were too high in Florida and Kentucky.

And now, new research also gives reason to think the AP estimate may be inflated.

In a March 3 posting on the website of the journal Health Affairs, two researchers from the Urban Institute analyzed findings from a nationwide poll and said, “Our findings imply that roughly 2.6 million people would have reported that their plan would no longer be offered due to noncompliance with the ACA.” And in this case, the methodology is made explicit.

In December 2013, the Urban Institute’s quarterly Health Reform Monitoring Survey of adults ages 18-64 included this question: “Did you receive a notice in the past few months from a health insurance company saying that your policy is cancelled or will no longer be offered at the end of 2013?” And of the 522 people polled who were covered by non-group policies, 18.6 percent said yes, their old plan would no longer be offered because it didn’t meet the new coverage standards that went into effect Jan. 1.

And if 14 million people were covered by non-group policies nationwide (as indicated by the National Health Information Survey of the U.S. Centers for Disease Control and Prevention), that percentage translates to 2.6 million non-group policies discontinued, the authors stated.

To be sure, there is always a statistical margin of error in any random-sample poll. Lead author Lisa Clemans-Cope told us in an email that statistically, there is a 95 percent certainty that the true percentage whose non-group policies were discontinued falls somewhere between 16.2 percent and 23.3 percent. That would put the number at anywhere between about 2.3 million and 3.3 million.

That range could be higher or lower depending on what number is used for the total who had non-group coverage in the first place. The Urban Institute authors cite a study published last year that found estimates of the total number of people covered by non-group policies ranged from 9.55 million to 25.3 million. So if 18.6 percent of non-group policyholders got notices that their policies were being dropped because of the new law, as the poll indicates, then the actual number whose plans were dropped could be as low as about 1.8 million or as high as 4.7 million (coincidentally, the same as the AP’s figure), depending on how many had such policies in the first place.

The authors, as noted, picked an estimate that fell in the middle of this range to arrive at their figure of 2.6 million discontinued policies. Until and unless better evidence comes along, that’s the most solidly based figure available.

How many “millions” so far have gained coverage?

The early numbers on enrollment in the exchanges and Medicaid don’t tell us how many of the enrollees were previously uninsured — despite some claims from Democrats to the contrary. The Obama administration disclosed on April 10 that 7.5 million had signed up for plans on the exchanges, but we don’t know how many previously had insurance. The Medicaid rolls increased by more than 3 million through the end of February, the administration also said, a figure that would reflect both those newly eligible under the law and previously eligible but now signing up.

But a survey funded by the Robert Wood Johnson Foundation and conducted by the Urban Institute indicates that many of those signing up for the exchanges and Medicaid may have been uninsured. It found that 5.4 million of the previously uninsured had gained coverage between September and the beginning of March. The exchanges launched Oct. 1.

An April 8 report by the nonprofit RAND Corp. put the figure of newly insured higher. Based on a nationwide poll, Rand estimated that there had been a net gain of 9.3 million insured “adults” as of mid-March, when the poll was being conducted. That includes marketplace and Medicaid enrollment, as well as an increase in employer-based enrollment.

Neither of those figures includes an estimated 3 million young adults who gained coverage in 2010 and 2011, likely because of the law’s provision allowing them to stay on their parents’ policies.

RAND also estimated that 700,000 who previously had individual market plans were now uninsured. The survey didn’t ascertain whether those newly uninsured were due to cancellations or voluntarily dropped coverage.

It will be some time before more concrete coverage numbers are available. The RAND numbers are extrapolated from a survey, and one with sizable margins of error. The estimate of 9.3 million newly insured has a margin of error of 3.5 million people, meaning researchers have a high degree of confidence that the true number would be between 5.8 million and 12.8 million. And the estimate of 700,000 uninsured who previously had individual market plans carries a margin of error of 900,000, putting the likely real number somewhere between zero and 1.6 million people.

Millions more are expected to gain insurance because of the law nationwide in the coming years. The nonpartisan Congressional Budget Office estimates that there will be 25 million fewer uninsured due to the ACA as early as 2016.

Losing Doctors?

The AFP ad also makes the claim that “millions of people can’t see their own doctors,” but there’s no evidence that all those who had individual market policies discontinued ended up not being able to keep their own doctors. Anecdotally, we know of some folks who were able to keep the same doctor on a new insurance policy. But those are only a few individual stories. One of our guiding principles here is the saying, “The plural of anecdote is not data.”

It is true that using a smaller network of providers is one way insurers can reduce premium costs, and there is evidence that insurers are indeed doing that for exchange plans. As Deborah Chollet, a senior fellow at Mathematica Policy Research, a nonpartisan research firm, told us in December: “The narrow-network plans offered by some issuers are intended to (a) maximize negotiating leverage with providers by narrowing their PPOs; and (b) thereby reduce premiums to attract consumers.”

Limited networks have existed for some time, as anyone with an HMO, PPO and the like can attest. There are no available statistics showing whether the plans on the new exchanges have more or less narrow networks than existed in the individual market previously. But, again, insurers certainly are limiting their networks to price their plans competitively.

Karen Pollitz, a senior fellow at the Kaiser Family Foundation, told us: “It’s definitely the case (based on conversations with insurers and with providers) that insurers have decided to limit networks in some instances in order to price their health plans more competitively.” She continued: “It’s also definitely the case that some providers have declined to participate in some of the new health insurance networks, holding out for higher fees from some insurers in return for a promise to participate exclusively in their networks. This is market competition at work — not entirely transparent, unfortunately, so it’s not yet clear what the impact will be on patients.”

— Lori Robertson and Brooks Jackson