Earlier this month, Oregon Health and Science University released new nationwide data that found low physician participation in Medicaid.
Researchers deemed these physicians “ghost” providers: physicians who are enrolled in Medicaid, but don’t care for even a single patient covered by the federal health insurance program.
Those findings also revealed that another third of physicians who are enrolled in Medicaid may be overburdened, with higher-than-average yearly patient volumes.
Dr. Jane Zhu, associate professor of medicine at OHSU, joins us with more details.
Note: The following transcript was transcribed digitally and validated for accuracy, readability and formatting by an OPB volunteer.
Dave Miller: From the Gert Boyle studio at OPB, this is Think Out Loud. I’m Dave Miller. Nationwide, nearly a third of physicians who are enrolled in Medicaid don’t actually care for a single patient covered by Medicaid insurance. That’s the top line from a new study led by a researcher at Oregon Health and Science University. Jane Zhu is an associate professor of medicine at OHSU. She also found that another third or so of physicians who have been approved to accept Medicaid patients might see so many of them that they are overburdened. Jane Zhu joins us now. It’s great to have you back on the show.
Jane Zhu: Thanks for having me, Dave.
Miller: So as I mentioned, one of the key findings from your study is this low physician participation rate among doctors who’ve been approved to offer care to people on Medicaid. What does a doctor have to do to get on that list?
Zhu: Not much actually, Dave. Sometimes hospitals, for example, on condition of employment, will require their providers – all their physicians and nurse practitioners, physician assistants – to register in the Medicaid program in order to be employed. Provider Medicaid eligibility simply refers to whether a clinician organization is approved and enrolled to bill Medicaid for covered services. You fill out some paperwork, enrollment confirms that the provider meets program requirements. They can receive reimbursement. That’s often done on a system level.
Miller: That’s a helpful point. So as opposed to an individual level, a doctor saying I want to do this, I’m going to go through a lot of hoops, and then providing that care, you’re saying that’s actually not the way the system works. Often, it’s your boss’s boss’s boss who’s going to do this, and everybody just gets that paperwork filled out.
Zhu: Yeah, exactly. It depends on the practice organization, obviously if you have a small solo practice or a small group practice, maybe that responsibility lies on the individual providers, but now that healthcare is increasingly consolidated and part of large health systems and organizations, often that function is elevated to the system level.
Miller: So what does it mean, that about 30% of these doctors – and am I right, that we are talking here, in your study, solely about physicians, not nurse practitioners or physician assistants?
Zhu: That’s right, we focused only on physicians, in part because in the Medicaid claims data, billing from physician assistants and nurse practitioners can be a little bit less reliable. There’s incident-to billing, there’s lots of other reasons why other providers are not as well identified.
Miller: So what does it mean, that 30% of these doctors did not see a single Medicaid patient over the course of a year?
Zhu: Let me just take a step back and paint the picture for why this is actually an important question. It’s important but understudied. We all know that Medicaid covers 80 million people in the U.S. now, but one of the most well-documented issues in Medicaid remains access to care. We often point to physician participation in the Medicaid program as being one of the main driving factors for low access to care. It’s lower than in commercial insurance and Medicare, so understanding whether and how physicians participate in Medicaid is a really key question.
So what we did in our analysis is, we first assessed the proportion of active physicians who are enrolled in Medicaid, and we found that actually enrollment on paper is pretty high, ranging from about 70% to 90% depending on the specialty that we looked at. And then we looked at the second question, which is, to what extent are these Medicaid-enrolled physicians actually filing claims or seeing Medicaid patients? What we find is that enrollment does not actually mean meaningful engagement, and our data really illustrates that. The key question that we’re answering, and the reason we care about this, is because in the end this has an effect on patient access.
Miller: And so, in other words, you might look at the participation rate of internists or cardiologists and say, hey, there’s actually a pretty good number of those doctors who take Medicaid in this area. Therefore, there’s pretty good access. But you’re saying, if a third of them don’t actually see patients, then that the participation rate is actually a truly misleading number.
Zhu: Yes, that’s right. We often, from a policy perspective, from an administration perspective, look at these lists of providers – both enrolled providers in the Medicaid program, but also when we comb down to the managed care plan level, the insurance plan level, to managed care directories, providers who are contracted with managed care plans to take care of patients – we know from this study and from many others that those lists grossly overestimate who’s actually delivering care in the Medicaid program.
And that means, for patients, really real burdens. It’s not just an administrative burden, it’s also a psychological burden. Because if patients think that there are X, Y, and Z physicians that are able to accept Medicaid and are taking Medicaid patients, they’re making calls, they’re going to try to get appointments to doctors that aren’t actually accepting Medicaid insurance or who have no availability, or who have long waitlists that don’t translate into timely access to care. These are not just annoying frictions, they’re real hassle costs, and they often lead to delays in medical care and unmet needs.
What we found, for example, in psychiatry, are particularly high rates of ghost physicians in psychiatry, and so we’ve seen this as well. If you think about what people with mental health burdens are facing when they’re looking for access to care for physicians, add that to the clinical conditions that they’re already facing, you’re getting unmet needs from the patient perspective.
Miller: Right. I’m glad you mentioned psychiatry as one of the specialties that you looked at, because, in addition to family practice, you looked at cardiology, dermatology, ophthalmology, and psychiatry.
Zhu: And dermatology, and it wasn’t just family medicine, it was primary care more broadly. I’m a primary care physician and you know, it has a special place in my heart.
Miller: What kinds of variances did you see among those different specialties in terms of ghost provider rates – people who are actually authorized to accept Medicaid but didn’t see Medicaid patients over the course of a year?
Zhu: Psychiatry really stood out. About 43% of enrolled psychiatrists were ghost physicians, and taking that a step further, even among those who did see Medicaid patients, the median psychiatrist that we looked at saw just three Medicaid patients and billed seven encounters in a year. And that’s compared to primary care, where median volumes were far higher, about 59 patients, and in cardiology, where the median was over 80.
So we’re seeing a lot of specialty-specific variation – not to say that there is not a reason for that. Psychiatrists often see patients longitudinally. They see fewer patients over time. They might have less clinical capacity compared to high-volume procedural specialties. So this is not meant to be a comparison of what is the right patient panel, it’s meant to just give us an ability to compare across specialties in a standardized way.
Miller: What are the reasons that you’ve either been able to identify and research or are assuming, based on your own clinical knowledge and maybe conversations with other providers, the reasons for doctors not seeing Medicaid patients?
Zhu: I mean, there’s a lot of potential reasons. Some of them are structural. I mentioned earlier that there are contracting or employment requirements. So if you’re working for a hospital system or you’re contracting with a managed care plan, those arrangements might require physicians to participate across all the plan’s lines of businesses. So if you want to take care of Medicare Advantage patients, you also have to include your physicians in their Medicaid plans. And there might be a lot of reasons where you end up with enrollment without any real intention or capacity to actively treat Medicaid patients, and that is a structural reason.
There’s lots of other reasons, like capacity constraints. We don’t have a window of visibility, obviously, into capacity – things like closed panels, there might be staffing shortages or practice disruptions. Physicians themselves, on the provider side, might have limited outpatient roles or be in transition. So they might not be taking new patients, whether they’re Medicaid or otherwise.
But there are also potentially Medicaid-specific reasons. It’s been well documented, for example, that Medicaid billing rules and prior authorization requirements, some of the documentation burdens can be really complex. So it’s possible, for example, that some physicians might enroll to preserve the option to see Medicaid patients, but they might limit participation if, as a rational actor in the healthcare system, they think that administrative costs outweigh a relatively low reimbursement, for example.
Miller: Just to be clear, is there anything that goes against the law if a doctor who’s authorized to see Medicaid patients doesn’t see any?
Zhu: No, not that I know of. It’s not enforced in that way. It’s not required. It’s not a requirement that if you’re enrolled, you have to see patients.
Miller: How much of this just, in the end, boils down to reimbursement rates? In other words, if Medicaid paid more to physicians for their various services, do you think you’d see a lower rate of so-called ghost providers?
Zhu: It’s simple but relatively unanswered, empirically. There’s a lot of data out there that suggests that if you increase… And they took advantage of other policies in, for example, the ACA marketplace, primary care reimbursement went up for a small period of time. And so researchers have studied that and they study the effect of increasing reimbursement on provider participation and actually, the data out there is fairly mixed.
There is some data that suggests that if you raise rates for providers that you do increase access to care, and there’s others that suggest that you don’t. I think what is really interesting and important to note is that there have been a lot of policy efforts – particularly in behavioral health, because of the crisis that we’re facing in behavioral health and because of the access gaps – to go ahead and increase reimbursement rates for behavioral health services, and this is something that providers themselves say is an important factor.
What we don’t know is whether those rate increases that have been implemented across many, many states in the U.S. have actually had an effect on provider participation in bringing both new providers into the system and sustaining the core groups of providers that are taking care of an outsized proportion of Medicaid patients.
Miller: So if there’s some question about the role that increasing reimbursements would play in this, what other factors or leverage at the state or federal level could be pulled to increase actual participation among doctors?
Zhu: I talked earlier about things like utilization management, prior authorization documentation burden, but certainly reimbursement is one lever. I think overall though, there needs to be, and I think our research suggests this, interestingly enough, we find that participation in Medicaid isn’t static. So most physicians stayed in the same participation category across our study period, but we actually saw pretty meaningful movement at the margin.
About 20% of ghost physicians began seeing Medicaid patients, and about a quarter to a third of what we call peripheral physicians, people who saw one to 10 Medicaid patients a year, began seeing more Medicaid patients year to year. So to me, what that suggests is that there’s actually latent capacity that you could potentially activate with the right incentives and policy and operational support.
Perhaps the levers that we think about could still work, but we need to design policy to differentiate between the persistent ghost providers, the ones that, no matter how much you will try to engage them, they don’t want to participate in Medicaid, and that’s fine, that’s their choice, versus the physicians who want to participate in Medicaid but don’t have the adequate support.
And using tools like longitudinal claims, using measures that actually assess this over time, and understanding the intensity of participation and the patterns of participation in Medicaid is going to be really important for developing and targeting policy, in my view.
Miller: We’ve been talking about the provider side, the doctor side of Medicaid, but our country and our state, we’re all about to go through a gigantic change in the other piece of this equation – the number of people who are on Medicaid itself. With the impending federal cuts, the number of Medicaid recipients is expected to plummet all over the place. How have you been thinking about those cuts to come, as somebody who researches access to healthcare in the U.S. more broadly?
Zhu: Yeah, this isn’t obviously a focus of this particular study, but we are obviously aware that states are really entering a period of fiscal pressure and Medicaid, and when budgets tighten, which they are bound to do, often the policymaker focus is to tighten utilization, to reduce policies implementing, for example, rate changes to reduce spending in some way.
I think what our study is really highlighting is that approach could potentially backfire if it’s weakening the already relatively small group of clinicians who are delivering
Medicaid care. I think we need to be really focused on understanding where money is spent and to what effect in the Medicaid program so that we can retain the programs that are working.
The other point behind that question would be, I think that coverage losses and patient churn off Medicaid is not just affecting patients, but it also affects the stability of the care delivery system. When people are moving on and off coverage, practices are going to be facing more unpredictable demand and unpredictable revenue streams, and that instability can make it harder for clinicians to continue to maintain Medicaid panels or invest in the staffing that’s needed to serve this population.
So I think in general, I am, as a researcher in this field, really worried about the changes coming to Medicaid, and I think doing work like this and understanding what are the areas of priority that need to be retained despite budget cuts is going to be really important as more people confront budget issues.
Miller: Jane Zhu, thanks very much.
Zhu: Thank you for having me.
Miller: Jane Zhu is a primary care physician and an associate professor of medicine at OHSU.
“Think Out Loud®” broadcasts live at noon every day and rebroadcasts at 8 p.m.
If you’d like to comment on any of the topics in this show or suggest a topic of your own, please get in touch with us on Facebook, send an email to thinkoutloud@opb.org, or you can leave a voicemail for us at 503-293-1983.
