America’s system for providing health care to the poor has its flaws, as anyone who has ever sought medical assistance under Medicaid can attest. But among its troubles, two have stood out. For physicians, Medicaid has traditionally paid poorly ––doctors received around half as much for seeing a Medicaid patient as for an otherwise identical Medicare patient. And for its clients, the key issue was simply finding doctors willing to take them into their practices. From a simple economics perspective, one might expect that both of these problems were two sides of one coin, and that solving them could be accomplished by raising the government reimbursement rates to doctors for Medicaid patients to match what Medicare pays. But economics is rarely simple.
A team of researchers from Wharton, the University of Pennsylvania’s Perelman School of Medicine and the Washington, D.C.-based Urban Institute went in search of facts. When, for two years, starting in 2013, those payments were equalized, they measured the effect on how easy it was for Medicaid patients to enroll with new physician practices. What they found surprised even the researchers.
The full study is detailed in a paper titled, “Appointment Availability after Increases in Medicaid Payments for Primary Care,” which was published in the New England Journal of Medicine. The paper was authored by Daniel Polsky, a Wharton health care management professor and a professor of medicine at the Perelman School; Michael Richards, a post-doctoral research associate at Penn’s Leonard Davis Institute of Health Economics; Simon Basseyn, an MBA/MD candidate at Penn and Wharton and a fellow at the Leonard Davis Institute; Karen V. Rhodes, a professor of emergency medicine at the Perelman School, and Genevieve M. Kenney, Douglas A. Wissoker and Stephen Zuckerman of the Urban Institute.
Polsky, who is also executive director of the Leonard Davis Institute, recently spoke with Knowledge at Wharton about the findings.
An edited transcript of the conversation appears below.
Fundamental Access Issue
Medicaid underpays doctors, and when Medicaid underpays doctors, it is hard for patients who have Medicaid insurance to get the care they need. Either the providers that typically give care to low-income patients are struggling with their finances, or they go out to doctors in the community and a lot of them do not want to see the patients [if] they do not get paid a lot of money. As part of the Affordable Care Act, [the federal government] decided to pay Medicaid doctors more. They paid them the same amount that they pay [for] Medicare patients — and that doubled the amount they got paid for [providing] those patients primary care.
So the question is: Are there more doctors willing to see these Medicaid patients when they are paid twice as much as they were before? And as an economist, it seems like a pretty obvious answer to that question: If you get paid more, you are more likely to see [those] patients. But this was a critical policy question, because this was a policy put in place for just two years to see how well it worked. And the hope was that doctors would engage the system at a time where Medicaid was expanding. There were more patients available.
So what we did was we called up these practices. We had field workers who pretended they were patients call, and they said, “Can I get a new patient appointment?”And we just marked down whether they got an appointment or not. It was as simple as that. Half the time, the scripts included patients who had Medicaid insurance and half the time, the control group [said they] had private insurance. We did this exact same study in 2012 before this policy went into effect, and then again this past spring in 2014, right in the middle of the implementation of this policy.
“There is a debate about why doctors do not see Medicaid patients, and what this [research] suggests is that if you pay them reasonably, they show up to see these patients.”
We saw about an eight percentage point jump in the percentage of [Medicaid] patients who were getting appointments. And what was really compelling about our findings was there is great variation between states in terms of how much the pay increased, because Medicaid pays different amounts in different states — it’s a state-run program. For some states the jump [in reimbursement rates] was double — a 100% jump. In others, it was maybe 30% or 40%. And in those states where the jump was double, the increases in appointment rates were about 15 percentage points. And in those states where it was much lower, it was about five percentage points. So you saw this much greater increase when they got paid even more.
When we looked in the private insurance group as a control to [see if] there was something else going on, we saw no change at all. So [those were] some pretty compelling results. [They] really suggested that this policy worked to increase appointment availability with primary care physicians when you pay them more money in the Medicaid system.
This is a very timely research for policymakers. It was a two-year policy that ended in January 2015, so policymakers were really interested in whether it worked.…There is a debate about why doctors do not see Medicaid patients, and suggests is that if you pay them reasonably, they show up to see these patients. It is really about payment. And it is not the fact that there may not be enough providers. I mean, there are some areas where there may not be enough providers overall, but the solution — if physicians are not seeing Medicaid patients — is to pay them adequately.…The results seem to show this makes a difference.
When we started this study, there was a lot of hesitation because it was the very beginning of this policy implementation. As I said, it was a two-year policy. It was starting up in 2013 and that was when we were getting ready for our second round of data collection. Implementation was delayed by six months. They had real trouble figuring out exactly how to pass through payments if there is capitated managed care [i.e., a system where health care providers are paid a set amount per period for each person assigned to them, regardless of whether that person ever requires care]. It is a lot easier when it is a fee-for-service than when it is capitated. So there were struggles with implementation. And a lot of people said, “Well, the payment is temporary, and so doctors are not going to change their patterns for a temporary policy.”The other thing that we had heard is that a lot of physicians were not really calling up the Medicaid office and saying, “You know, I really want to be in this program.”So a lot of the anecdotal evidence was pointing toward there really being no action.
So at the end of the summer, we took our first peek at the results and we were all very surprised to see such a large, unambiguous affect. Really, the surprise was that we found something — particularly when the rhetoric at the time was that this policy really was not having an impact.
“I think we can expect that the gains that we saw in the last two years in terms of more doctors participating in Medicaid will probably go back down.”
Doctors, Patients—and Politicians
I think the most practical implication [is for] policy makers. A lot of discussion at the state level [involved politicians] trying to decide if they wanted to extend this pay bump policy, and what they were saying at the time was, “You know, there is no evidence yet. So we’re not going to make a decision because there is no evidence.”But now we have evidence. So I think the real practical implication for people who are on the front lines of deciding whether doctors should be paid more for this specific policy is that, you know, there is evidence. And evidence-based policy is an important way to make policy, and this policy seems to work.
But I think there is a broader point that can be made that is important for physicians who run businesses: Physicians are not immune to the same economic incentives as anyone else running a business.…When you are talking about health, there might be some moral obligation to see everyone who comes in your office, but you cannot have an office that can turn the lights on unless you can pay the bills. And when you do not pay [doctors] enough to turn the lights on to see the patients — when the reimbursement is like $30 or $40 for a half hour of your time — it is really hard for some of these doctors to justify that in their day. They are just trying to have a business where they bring in more money than they spend. So I think what this suggests is …physicians are running businesses, too. And they should not be set up against different rules than other businesses.
There is a lot of debate in Congress right now — I would not say a lot of debate, but this policy of what we call the Medicaid rate bump was introduced in the President’s budget. Just a couple weeks ago, in February, he proposed to Congress that they extend the pay bump for another two years and make some modifications. And there is a lot that is going on in Washington to make some decisions about various aspects of health policy. It is all very, very political, but from a practical end, this is about a $6 billion a year policy to pay Medicaid providers for primary care at the same rate as they pay Medicare doctors. And [that’s] really a drop in the bucket when you think of the trillion dollar Medicaid/ACA bill. And so I think the practical implication is that we know there are some effects — some important effects — for patients getting access to providers, having continuity of care, and this provides some evidence that there is a benefit of this policy. Now the politicians have to decide if that payment — that expense — is worth it for the gain that you get.
The Next Level
The focus here was on primary care physicians in Medicaid. That is what the policy was aimed at. But in Medicaid it is really difficult in some areas not only to get access to primary care providers, but if you have a need for specialty care it could be even more difficult. The payment gaps for specialty care can be even more severe. So the implications extend beyond primary care. If you pay doctors more, they are more likely to extend appointment availability to patients without insurance, and so I think it would extend beyond primary-care providers to specialty-care providers in particular areas where that access is a real issue. [The research] suggests that paying specialty providers more money would help address that problem.
“When you are talking about health, there might be some moral obligation to see everyone who comes in your office, but you cannot have an office that can turn the lights on unless you can pay the bills.”
There was an article in The New York Times in December when this two-year policy ended. It was always a temporary policy, and so payment rates doubled two years ago, but when they went back down January 1, the headline was “Doctors Get Pay Cuts of 50%.”Will they still see patients? I think …what goes up is probably going to come down. What will happen now that payment rates have gone back down? I think we can expect that the gains that we saw in the last two years in terms of more doctors participating in Medicaid will probably go back down. We will go back in the field and measure it, but I think for the impact on Medicaid patients, I think it is an important finding.
Correcting Public Misperceptions
I think that …there is a perception that there are fewer doctors that see Medicaid patients, maybe because doctors do not want to see Medicaid patients. What this research suggests is that if you pay them reasonably, they will show up to see these patients. It is more about the payment rates than it is something about this patient population.
How the Study Stands Apart
This is the first time we have been able to look at the relationship between payment rates and physician appointment availability or access — participation in the Medicaid program using experimental methods. Other people have looked [at the] data, [and found that] those states that have higher payment rates tend to have more doctors that participate versus [those with lower payments]. But that could be for a number of reasons …or that Medicaid sets up payment rates that way, and it is just an observation. But to be able to go into the field, pose as patients actually seeking appointments, measure real behavior using this audit methodology in this experimental way where we randomized the scripts to people having Medicaid or private insurance, we have controlled for other aspects of access. That really sets us apart.…I think just that audit methodology is very compelling for people on the street. If you say “I ran a regression analysis and my P value was whatever,”somehow you have already lost half the crowd. So it was both a clean experimental design and I think a very compelling easy-to-relate-to finding.
Well, this paper that I was telling you about is part of a much larger project — a larger data collection effort — which is tracking how available physicians are for seeing patients around the insurance expansions of the Affordable Care Act. In the past couple of years, there has been a large group of people — 10 million to 15 million people — who did not have insurance before, and now have insurance and will be demanding to see doctors in ways that they have not before, because they have insurance to cover those visits. The fear is that there will not be enough doctors to meet this increased demand for care. We did our baseline measures in 2012 before insurance expanded. In a number of different domains, we are studying whether there is sufficient capacity in the physician workforce to meet this increased demand. Are we going to see a squeeze on the Medicaid side?
For private insurance, one of the questions that people are curious about is whether, [for] those people [who] have always had insurance: Will their care remain the same? But if their doctors are so busy because there are more patients, maybe their care will not be the same, and we will be able to measure that. You know, we will have these baseline data to know what things were like before, and we can monitor changes. We are going to look at changes — look at the access issues that exist in health insurance exchanges.
There are a lot of questions we are still pursuing with this larger ongoing work. We are hoping to go back in the field to collect a third round of data in the next six to 12 months.