Appointment audit studies, like the one I wrote about yesterday, have limitations. They assess the degree to which doctors' offices will accept new patients, by insurance status. Some might call that "access," but it's not the full story. Audit study results do not directly reveal the extent to which variations in new appointment availability translate into variations in receipt of necessary or appropriate care. "Access" is not so simple a concept.
Recall that based on the findings from the JAMA Internal Medicine (JAMA IM) study I summarized, to achieve the same level of success at making a primary care appointment, a new Medicaid patient, calling offices at random, would have to call about two offices for every one that a new privately insured patient called. Is this a big deal?
A companion study led by Genevieve Kenney and including some of the same researchers as the JAMA IM paper, addresses this question. They used 2012 National Health Interview Survey (NHIS) data to examine access issues for low income adults with Medicaid coverage, private, employer-sponsored insurance (ESI), or no insurance. The NHIS is a nationally representative survey of 35,000 households, focusing on the non-institutionalized, civilian, U.S. population.
For low-income (<250% of the federal poverty level) adults with new coverage, their findings corroborate those of the JAMA IM paper. New Medicaid enrollees were more likely to have trouble finding a doctor than those with new ESI. However, even there, just 11.3 percent of new Medicaid enrollees said that they had difficulties finding a general doctor or provider and only 2.8 percent said that they could not find a general doctor or provider with availability.
Where they add important nuance to the JAMA IM work is in their examination of low-income adults with continuous, not new, coverage. Their sample included about 1,800 and 3,500 low-income, full-year insured adults who had Medicaid or ESI coverage, respectively, at the time of the survey. They found little access differences across these two continuously insured groups.
For example, 86% of Medicaid enrollees said they had a usual source of care, about the same percentage as ESI enrollees. Likewise, close to two-thirds of both Medicaid and ESI enrollees had had a visit with a general doctor in the prior year. (Only 38% of the uninsured had a usual source of care and 32% had had a visit with a general doctor.) And, 4% of the Medicaid covered and 2% of those with established ESI had trouble finding a doctor, a difference that was not statistically significant after adjusting for demographic and health factors. (About 7% of the uninsured had such difficulty.) However, Medicaid enrollees were more likely to experience delays in care because of provider availability than those with ESI or with no insurance—but even there, just 8.4 percent of established Medicaid enrollees said that that they had delayed care for this reason.
They found differences in the area of affordability concerns. Medicaid enrollees have fewer of them.
[L]ow-income adults with Medicaid coverage were 3.8 and 5.4 percentage points less likely than low-income adults with employer-sponsored coverage to say they had experienced an unmet need for medical care or for prescription drugs because of affordability concerns, respectively. […] Controlling for other factors, we find that Medicaid enrollees were 2.9 percentage points less likely to say they delayed getting needed medical care because of affordability concerns and were 6.2 percentage points less likely to have delayed filling a prescription or have made some other change in response to cost concerns. [All differences are statistically significant.]
(They also found that the uninsured also have greater affordability issues than Medicaid patients.)
So, continuously insured Medicaid patients face delays, though no more trouble finding doctors, as ESI patients. Medicaid patients also experience less financial strain. How does all this shake out in terms of outcomes? The investigators considered receipt of preventive care. With one exception, there were no statistically significant differences in the area of preventive care: receipt of blood pressure checks, cholesterol checks, blood sugar checks, pap smears and mammograms for women, colon cancer screening, diet counseling, and discussions about smoking all occurred at the same rate for Medicaid enrollees as for those with ESI. However, those with ESI were 5 percentage points more likely to have received a flu shot. (The uninsured had much lower rates of receipt of preventive care than Medicaid patients.)
These results suggest that in real-world conditions, Medicaid enrollees with continuous coverage (as opposed to new Medicaid enrollees) are able to achieve comparable access to care as those with ESI, albeit with some delay. Perhaps this is the case because, with experience, Medicaid enrollees find which providers are more likely to accept them, but those providers are busier. The findings also show that Medicaid offers greater financial protection than ESI and, in general, offers no lower quality in terms of preventive care.
The two analyses I've reviewed this week demonstrate the nuances in the study of "access." When you audit offices, Medicaid appears to offer potentially problematic access. But when you examine the experience of enrollees themselves—particularly ones with continuous coverage, as is more likely to be the case as the uninsurance rate drops—few access issues are apparent. This difference in "access" across methodology has been observed before in other studies. But it's not one many commentators and pundits seem to be aware of.
You can make Medicaid look dreadful by cherry picking how you define "access." Or you can take a broader, more nuanced view of “access,” considering also patients' experience. When you do, you find that Medicaid works pretty well.