Was the GAO Medicaid Study a Cover-up?

Judging by titles and headlines, the GAO is just as confused about the contents of its own reports as the news outlets reporting on them.

In November, 2012, GAO released a report entitled States Made Multiple Program Changes, and Beneficiaries Generally Reported Access Comparable to Private Insurance. A Washington Post headline read “Study: Medicaid, private insurance give same access to health care.” A Modernhealthcare.com headline read, “Medicaid providers tough to find for many states, report says.”

So what did the report really say? Read on.

The report begins by discussing the results of a web-based survey of state officials done in the first half of 2012:

Of the 55 states [and territories] responding to our survey, 38 states reported experiencing challenges to ensuring enough participating providers for Medicaid beneficiaries. Ensuring sufficient dental providers was particularly challenging — but states also reported that ensuring sufficient provider participation in specialty care was problematic. Specifically, states most frequently reported having difficulty ensuring sufficient Medicaid providers for psychiatry, obstetrics and gynecology, surgical specialties, and pediatric services.

Reasons cited by officials for Medicaid network inadequacy included overall shortages of physicians, low Medicaid payment rates, missed appointments, and the administrative burdens of enrolling in Medicaid and getting paid by it.

After the discussion of provider participation as reported by state officials, the GAO examines whether Medicaid beneficiaries reported problems accessing Medicaid care. That section of the report relies on data from the National Health Interview Survey (NHIS) and the Medical Expenditure Panel Survey (MEPS).

Making sense of the section of the report analyzing the patient survey results is challenging. The report does not detail how it combined the MEPS and NHIS data to measure access, and the full sample, full-year, coverage comparison is the only one that shows a statistically insignificant difference between the percent of people on Medicaid who reported access problems, 3.7 percent, and the percentage of people with private coverage who reported access problems, 3.0 percent. Of those in fair or poor health with full-year coverage, 9.9 percent of those on Medicaid and 8.4 percent of those with private coverage reported access problems, also a statistically insignificant difference.

Differences in reported problems for Medicaid and private coverage widened for people with coverage for 11 months or less. But people within partial year coverage groups reportedly overlapped, with some individuals having had both Medicaid and private insurance for part of the year. Adding to the confusion is the fact that the Medicaid children category included those in the Children’s Health Insurance Program (CHIP). This assumption blurs distinctions. In some states, CHIP pays better than Medicaid and acts more like private coverage. Finally, although Census data suggest that 9.4 percent of US children were uninsured and 13.3 percent of US residents were over 65 in 2011, the report states that “sample sizes for uninsured children and for adults aged 65 and older were below 100,” leaving one to wonder just how big, and how representative, the sample actually used was.

Given the confusion over the kids, most of whom don’t need much medical care, and the fact that people over 65 get generally get medical coverage from Medicare, the result for the working-age group makes the most sense. When only working-age adult samples are compared, 7.8 percent of working-age adults with full-year Medicaid coverage, people aged 18 through 64, reported difficulty obtaining care. Only 3.3 percent of those with private coverage reported a problem. The report helpfully notes that this is a “statistically significant difference” and, on page 26, concludes that “Working-Age Adult Medicaid Beneficiaries Reported Greater Difficulties Obtaining Care than Those with Private Insurance.”

This is a fairly big deal. It indicates that Medicaid access is not as good as private access for reasons that include lack of adequate transportation, long waiting times at appointments, inconvenient appointment times, and long waits for appointments.

It’s too bad that those who wrote the report did not see fit to put this result anywhere in the report abstract.

Print