Most U.S. doctors are signed up with Medicaid—and are thus poised, at least in theory, to treat the 70 million-plus people enrolled in the massive government health insurance program.

But there is a catch. In practice, many of those doctors don’t actually see Medicaid patients at all.

According to a nationwide study published Monday by Oregon Health & Science University researchers, more than 1 in 4 doctors enrolled in Medicaid delivered no actual care to Medicaid beneficiaries in a recent year under review. And many physicians who did see Medicaid patients did so only rarely.

Meanwhile, another subset of Medicaid-registered doctors saw Medicaid patients frequently. One in three treated more than 150 unique Medicaid patients in 2021, the most recent year for which researchers could access comprehensive data.

“There is a set of providers that are ‘ghost’ providers,” says Dr. Jane Zhu, an OHSU primary care physician and researcher who headed up the study, published in Health Affairs. “And then there are a set of providers that are core providers. They’re really providing an outsized role in care delivery in Medicaid.”

The finding, which is consistent with earlier, more locally based studies bolsters the case for policymakers to look beyond physician enrollment and insurance company directories as they seek to gauge and improve health care access, Zhu tells WW. It speaks to a dynamic that will be familiar to many in Oregon, where Zhu and colleagues have documented numerous cases in which Medicaid-based insurance plans list doctors in their networks who don’t go on to actually provide care to Medicaid patients.

The ghost provider phenomenon has real world consequences for say, someone on Medicaid with depression and anxiety who can’t find an in-network psychiatrist to treat them within a reasonable time frame.

Growing wait times—often the result of a doctor shortage—are a known problem for regulators. But “if you have no idea what percentage of your actual physicians are participating in Medicaid and your numbers are all false, you’re going to be focused on the wrong thing in order to incentivize participation,” Zhu says.

Various health care forces conspire to make more doctors seem available than actually are. Perhaps a health system wants to join a good-paying commercial insurance network. As part of the deal, it might have to contract with that insurance company’s Medicaid plan as well—even as the health system, perhaps wary of lower reinbursement rates, doesn’t actually intend to treat Medicaid patients.

Meanwhile, the insurance companies that manage Medicaid-funded plans are regulated by certain network adequacy rules, which incentivize them to fill their network with many providers on paper—without rigorously assessing whether those providers are actually providing care.

The problem is not hypothetical. Zhu led a 2022 study on “phantom networks” in Oregon—documenting a discrepancy between reported and actual mental health care access in the Oregon Medicaid program. Looking at managed Medicaid plans in the state, the researchers found that a whopping 58% of mental health providers listed in the insurance network directories were “phantom” providers who did not actually see any Medicaid patients during the year in question.

The latest study looks not at insurance networks per se, but a larger group: providers listed in state registries (many of whom would, down the line, participate—whether in real or ghost form—in insurance networks).

The study found notable differences between provider types. Psychiatrists in Medicaid directories were the most likely to be ghost providers (43%) while cardiologists and primary care doctors were less likely to be ghost providers (18% and 26%, respectively).

Meanwhile, whereas psychiatrists enrolled with Medicaid saw a median of three unique Medicaid patients in 2021, primary care docs enrolled in Medicaid saw a median 59 such patients.

One surprising aspect of the study was that doctor participation was not in all cases static. Over the three most recent years researchers looked at—2019 to 2021—a fifth of ghost physicians started to see Medicaid patients, while some once-peripheral physicians started to see such patients more.

“There’s going to be a set of physicians who are never going to take care of Medicaid patients no matter how hard you push,” Zhu says. “But there is, among this broader set of physicians, a set who probably want to see Medicaid patients, but may not have the capacity or the ability to do that in a sustained way.”

In this fact, she sees latent system capacity that policymakers might be able to leverage. Meanwhile, she says, lawmakers might focus less on engaging physicians who may never intend to provide care for Medicaid patients, and pay more attention to the core providers who provide the bulk of Medicaid care.

“Maybe we should be targeting more support to those providers to make sure that they’re sustained in the program,” she says. “Because if one of them leaves, it has a huge implication for the patient population that they’re seeing.”

Willamette Week’s reporting has concrete impacts that change laws, force action from civic leaders, and drive compromised politicians from public office.

Support WW