Fact check: Medicaid’s doctor participation rates – USA TODAY
In seeking to repeal and replace the Affordable Care Act, Health and Human Services Secretary Tom Price has said “one out of every three physicians in this nation aren’t seeing Medicaid patients.” House Speaker Paul Ryan said “more and more doctors just won’t take Medicaid.”
It’s a common criticism of the Medicaid program — that the doctor participation rate is lower than the rate for Medicare beneficiaries or the privately insured. The implication is that Medicaid patients cannot access care and that it has gotten worse since the Affordable Care Act expanded the health care program for the low-income and disabled.
But experts say that implication is misleading:
The most recent figures from the Centers for Disease Control and Prevention’s National Center for Health Statistics are for 2013, showing the percentage of physicians accepting new Medicaid patients was 68.9%, while 84.7% accepted new privately insured patients and 83.7% accepted new Medicare patients. That’s based on a national survey of more than 4,000 office-based physicians.
Price made his claim twice during a March 15 CNN town hall on the since-failed Republican health care bill, which would have cut federal Medicaid spending and phased out the Affordable Care Act’s expansion of eligibility, with enhanced federal funding.
Medicaid, which has nearly 69 million enrollees (with another 5.5 million on the Children’s Health Insurance Program), is operated with state and federal matching funds. For the newly eligible under the ACA expansion, the federal government paid 100% of the cost from 2014 to 2016. That went down to 95% in 2017 and will be a 90% payment by 2020.
The expansion, which 31 states, plus Washington, D.C., chose to implement, extended eligibility beyond the traditional enrollees — low-income families, pregnant women, children, the disabled and the elderly — to all individuals under age 65 who earn up to 138% of the federal poverty level (about $16,643 a year for an individual).
That includes individuals like Brian Kline, who works a retail job and was diagnosed with cancer last year. He told Price during the CNN town hall: “Medicaid expansion saved my life and saved me from medical bankruptcy.” He asked, “Why do you want to take away my Medicaid expansion?”
Twice in his response, Price cited the doctor participation statistic. Price said Medicaid may have worked for Brian, but it didn’t work for everyone. “Now, I know that folks don’t want to hear this, but the fact is that, again, one out of every three physicians in this nation aren’t seeing Medicaid patients. And they should. Again, if we want to be honest, we ought to ask the question to ourselves and our society why and fix those challenges that exist in the program,” he said.
Earlier this year, House Speaker Paul Ryan claimed Medicaid’s doctor participation rate had decreased, saying “more and more doctors just won’t take Medicaid.” But the available data don’t support that.
“There isn’t an ongoing monitoring system that measures this trend over time. There are occasional studies,” Benjamin D. Sommers, an associate professor of health policy and economics at Harvard University whose research includes Medicaid policy and health care access, told us in an email. “It’s true that a lower share of doctors take Medicaid than private insurance or Medicare, but I haven’t seen any trend for that.”
Experts consistently referred us to the work of Sandra Decker, a co-author on the NCHS’ latest report, for participation rates. Decker used the same data source — the National Electronic Health Records Survey, which the 2015 report describes as “an annual, nationally representative survey of office-based physicians affiliated with the National Ambulatory Medical Care Survey” — for a July 2013 report published in the journal Health Affairs. That report similarly found 70.1% of physicians accepted new Medicaid patients, using data from the physician surveys for 2011 and 2012 combined. The figure for only primary care physicians was 66.8%.
A September 2012 NCHS report suggests little to no change in participation rates for “generalist physicians” over a decade. “Between 1999–2000 and 2009–2010, no significant change was observed in the percentages of generalist physicians who said they would accept new patients with private insurance, Medicare, Medicaid, or self-payment,” the report said.
Specifically, 69% of generalist physicians accepted new Medicaid patients in 1999-2000 and 65% accepted them in 2009-2010. The report showed a higher acceptance rate among specialty physicians: 71% in 2009-2010, but a drop of 7 percentage points from the 1999-2000 level. Smaller reductions were reported for specialty care for Medicare patients and the privately insured.
Participation rates also vary by specialty, with pediatricians more likely than internal medicine doctors to accept new Medicaid patients, for instance, according to the Health Affairs report on 2011 and 2012 surveys. Psychiatry — with more than half of psychiatrists saying they wouldn’t accept new Medicaid patients in 2011-2012 — is “a big problem,” Julia Paradise, associate director of the Program on Medicaid and the Uninsured at the Kaiser Family Foundation, told us.
Earlier this year, Ryan’s office had referred us to figures from Merritt Hawkins, a consulting firm on physician staffing, but more recent results from those studies also show no drop in participation rates. The firm called about 1,400 physician offices in five specialties in 15 markets, asking to make a new patient appointment and if the doctor accepted Medicaid. The results: 53% said they accepted new Medicaid patients in 2017, 45.7% in 2014, 55.4% in 2009 and 49.9% in 2004.
It’s a more limited measure than the NCHS reports, and researchers didn’t try to call the same doctors’ offices year to year. But the results show the rate hovering around 50%.
A broader study by Merritt Hawkins and the Physicians Foundation found higher percentages of Medicaid acceptance. It surveyed 17,236 physicians and found 63.7% accepted all Medicaid patients and another 20.3% accepted limited numbers of Medicaid patients in 2016. The survey results for 2014 were similar.
The lower doctor participation rate for Medicaid, as compared with Medicare or private insurance, is largely tied to lower reimbursement rates. “Research suggests strongly that there’s a positive correlation between provider payment rates in Medicaid and participation rates in Medicaid,” KFF’s Paradise said in a phone interview. “And there’s a lot of variation in state payment rates.”
While the most recent report from the National Center for Health Statistics found that 68.9% of physicians said they accepted new Medicaid patients, that figure is nationwide. The participation rate was “significantly higher than the national average in 25 states,” and “significantly lower” than the average in five states, the report said.
The variation went from a low of 38.7% in New Jersey (where primary care reimbursement rates are 48% of Medicare rates) to a high of 96.5% in Nebraska (where the primary care reimbursement is 75% of Medicare). Montana, with a 90% physician participation rate, pays the same rate as Medicare for primary care, while California, with a 54.2 participation rate, pays 42% of the Medicare reimbursement rate.
Why the variation in what state programs will pay doctors? There are federal requirements on what groups of people qualify for Medicaid, but states have leeway to expand that eligibility. Gail Wilensky, the head of the Medicare and Medicaid programs during the George H.W. Bush administration and now a senior fellow at Project HOPE, a health training and humanitarian organization, told us that reimbursements tend to correlate to how much a state had expanded eligibility.
“The states that restricted their coverage to the mandatory coverage and only minor deviations, meaning they had a smaller population covered, tended to have reimbursement rates that were close to Medicare,” she said. “And the states like New York and New Jersey, which had very expansive eligibility coverage in Medicaid tended to have among the lowest reimbursement rates.”
In order to limit spending, Wilensky explained, states can limit optional eligibility for Medicaid or optional benefits — or they can limit what they’ll pay. So if a state expands eligibility, the only option is to cut the reimbursement rate. “It is one of the policy options at their disposal to try to limit expenditures, although sometimes it has had unintended consequences of driving people to the emergency rooms for much of their care.”
“It’s why most states now have used managed care contracts to provide health care” to Medicaid enrollees, Wilensky said.
Paradise, of the Kaiser Family Foundation, agrees that “a majority” of Medicaid beneficiaries are in managed care plans.
While those networks may be limited, beneficiaries have a list of participating doctors, just like those on private insurance plans.
A lower reimbursement rate isn’t the only aspect of Medicaid that can deter some doctors from accepting patients. “What I’ve seen both in speaking to physicians and some research that’s been done, it’s not just the lower pay, it’s the lower pay on top of the paperwork burden and sometimes delays in payment,” Wilensky said.
An April 2011 Kaiser Family Foundation report echoes that. It found, based on a nationally representative mail survey of 1,460 primary care physicians, that 90% of the doctors who take no or only some new Medicaid patients said low payment was the reason. About 75% also cited delayed payment and billing requirements, and 60% cited a “high clinical burden.” The Medicaid population, which includes those with disabilities, tends to have poorer health than the privately insured.
So how would one increase the doctor participation rate? Obviously, Wilensky said, the program could pay doctors more. Beyond that, paperwork requirements could be reduced and payments to doctors could be made more quickly. “I think it would help,” she said.
Access to Care
The question is how much of a problem this 70% participation rate is. The experts we interviewed cautioned us that the participation rate for Medicaid was a limited measure.
The supply of doctors and their geographic distribution are important factors, Paradise said. Access to care could be better in an area with a high supply of doctors but low participation, compared with an area with few physicians but high participation rates.
And there are other caveats. “It depends not just on whether they participate, but how much they participate,” Wilensky said. “You have people who have a couple patients and you have people whose practice is primarily Medicaid.”
A 2016 paper in the Journal of Health Politics, Policy and Law also said that “measures focusing exclusively on physicians ignore the growing role played by nonphysician providers, particularly in underserved settings” and that it wasn’t clear “whether having a large number of providers each treating a small proportion of low-income adults is preferable to greater concentration among providers with particular expertise caring for low-income populations.”
One of the authors of that paper, Harvard’s Benjamin Sommers, told us via email: “When a politician says, ‘Doctors don’t take Medicaid,’ they are implying that patients with Medicaid can’t access care. But there have been numerous studies of the ACA’s Medicaid expansion that consistently show large improvements in access to medical care from Medicaid expansion.”
Sommers was the lead author for one such study, published in JAMA Internal Medicine in 2016, that looked at survey data for three states and found Medicaid expansion “was associated with significantly increased access to primary care,” as well as “fewer skipped medications due to cost,” among other factors.
A KFF analysis of data released in 2015 from the National Health Interview Survey — a sample of about 35,000 households — shows that 74% of adults with Medicaid coverage had seen a doctor in the previous 12 months, while 69% with private insurance had done so. Eighty-five percent with Medicaid were satisfied with their health care, while 87% with private insurance said so.
“Medicaid beneficiaries actually fare as well as privately insured people on very important measures on access to care,” Paradise said.
A 2014 Urban Institute study of 2012 NHIS data on low-income adults found that “under 5 percent of adults with established Medicaid coverage [those with insurance for the previous 12 months] and 2 percent of low-income adults with [employer-sponsored insurance] report that they had trouble finding a general doctor or provider in the past year.” The gap was higher for those with new Medicaid or private insurance — defined as those who had been uninsured at some point in the previous 12 months. Using 2011 and 2012 data, the study found 11.3% of these new Medicaid enrollees reported difficulty in finding a general care provider, while 6.2% of new private enrollees reported difficulty.
The vast majority — 85% — of both low-income adults with established Medicaid coverage and those with established employer coverage said they had a usual source of health care. The Medicaid population was more likely to say a health clinic or center was that usual source of care.
A June 2016 issue brief from MACPAC, which advises Congress on Medicaid policy, described similar findings for children with Medicaid or CHIP coverage. It said that those children “are more likely than those with private coverage to report difficulties accessing medical care,” such as finding a doctor who accepts their insurance. But Medicaid and CHIP children “are as likely to have a usual source of medical care as privately insured children even when controlling for age, race and ethnicity, income level, and special health care needs status.”
That brief also used data from NHIS as well as the household component of the Medical Expenditures Panel Survey, which draws from a subsample of those who participated in the NHIS.
All of these analyses show that the uninsured do worse — in terms of both access to and usage of care, and affordability — than those with Medicaid or private insurance.
A July 2015 Government Accountability Office report — which relied on GAO’s past reports, documentation from the Centers for Medicare & Medicaid Services and interviews with CMS officials — also echoed other research: “Medicaid enrollees report access to care that is generally comparable to that of privately insured individuals and better than that of uninsured individuals, but may have greater health care needs and greater difficulty accessing specialty and dental care.” The report mentioned mental health care as a specialty area of concern.
GAO said that Medicaid enrollees have “a higher rate of mental health conditions” than the privately insured. Officials in six states that expanded Medicaid under the ACA “generally reported that Medicaid expansion had increased the availability of mental health treatment for newly eligible adults,” but there were still concerns about access “due to shortages of Medicaid-participating psychiatrists and psychiatric drug prescribers.”
One would logically expect access and participation issues to worsen with the Affordable Care Act’s Medicaid expansion. More than 11 million newly eligible adults enrolled in Medicaid through March 2016. But some research on the issue hasn’t found that to be the case.
A University of Pennsylvania study looked at how a temporary increase in payment rates under the ACA affected the ability of Medicaid enrollees to get new-patient appointments at doctors’ offices that participated in Medicaid. Under the ACA, the federal government in 2013 and 2014 boosted reimbursement rates for primary care doctors, who already treated Medicaid patients, bringing them up to Medicare reimbursement rates.
For the study, published in the New England Journal of Medicine in February 2015, trained staff members called primary-care offices in 10 states, seeking appointments for both routine and urgent care, with some saying they had Medicaid and others saying they had private insurance. The results: There was an increase in appointment availability from 2012 to 2014, for Medicaid callers from 58.7% to 66.4%. For private-insurance callers, the availability stayed the same at 86%.
But that increased availability continued, even after the federal payment bump ended, according to 2016 data gathered by the researchers. “Appointment availability is much higher than before health care reform,” one of the authors, Daniel Polsky, executive director of the University of Pennsylvania’s Leonard Davis Institute of Health Economics, told us.
The latest research, published in JAMA Internal Medicine, found that appointment availability with primary care physicians for new Medicaid patients increased by 5.4 percentage points between 2012 and 2016. And “both Medicaid patients and the privately insured experienced a one-day increase in median wait times” for an appointment, the authors wrote in a March 2017 issue brief.
Fifteen states continued the boost in payments, fully or partially, after the federal payment increase ended. However, one state — Alaska — already had higher reimbursement rates than Medicare. And only one of those 15 states — Iowa — was among the states studied by Polsky and his co-authors.
The higher appointment availability for Medicaid callers in 2016 was a “surprising result,” the authors said.
“As an economist, you’d think – more demand for limited supply with fixed prices, that you’re going to see a decline,” Polsky told us. “But we didn’t see it.”
Another study, by MACPAC, which interviewed eight states on their experiences, concluded that “there is not enough evidence to definitively determine whether the payment increase had an effect on provider participation or enrollee access to primary care in Medicaid.”
KFF’s Paradise told us that she hasn’t seen evidence that access to care or physician participation has dissipated under the ACA Medicaid expansion so far. “We have not seen that in the work we’ve done,” she said.
“I think it’s common sense to watch what happens when there’s a coverage expansion and more demand on the health care system and capacity in the system,” she said. But “what the data suggest is access continues to be robust in the program,” with some areas of concern, such as access to psychiatric care.