Policy makers and researchers are eager to learn the effects of the Patient Protection and Affordable Care Act of 2010 (ACA) and its many provisions, but to date, they have been frustrated by the dearth of robust evidence on the ACA's true impacts on important health care and patient outcomes (e.g., access to primary care services). The present limitations of evidence, often a consequence of delays and inconsistencies in the law's implementation, have begun to affect policy making in the ACA's wake. In this article, we consider the debates among state and federal policy makers about whether to extend the ACA's so-called fee bump provision, whereby Medicaid fees for primary care services were increased to 100 percent of Medicare levels during 2013 and 2014. We describe the difficulties state Medicaid programs have experienced in implementing the fee bump, as well as how the resulting evidence gap and the broader political context have shaped the deliberations. To conclude, we identify policy alternatives and other factors policy makers should consider when deciding whether to extend or reinstitute the fee bump in the coming years.
Editor's Note: JHPPL has started an ACA Scholar-Practitioner Network (ASPN). The ASPN assembles people of different backgrounds (practitioners, stakeholders, and researchers) involved in state-level health reform implementation across the United States. The newly developed ASPN website documents ACA implementation research projects to assist policy makers, researchers, and journalists in identifying and integrating scholarly work on state-level implementation of the ACA. If you would like your work included on the ASPN website, please contact web coordinator Phillip Singer at email@example.com. You can visit the site at http://ssascholars.uchicago.edu/jhppl/.
JHPPL seeks to bring this important and timely work to the fore in Report on Health Reform Implementation, a recurring special section. The journal will publish essays in this section based on findings that emerge from network participants. Thanks to funding from the Robert Wood Johnson Foundation, all essays in the section are published open access.
—Colleen M. Grogan
One of the many contentious debates brought about by the Patient Protection and Affordable Care Act of 2010 (ACA) concerns whether to extend the law's increase in fees paid for primary care services in Medicaid. This provision, commonly known as the “fee bump,” raises these services' reimbursement rates to 100 percent of Medicare levels during the years 2013 and 2014 through a 100 percent federal-match subsidy. With 2015 and the expiration of these increased fees rapidly approaching, state and federal policy makers have been evaluating whether and how to extend the fee bump.
In an ideal world, these discussions would be informed by early returns on the policy's effectiveness. It would be important to know how the fee changes have affected costs and access to primary care for Medicaid beneficiaries. However, multiple administrative delays, mounting frustrations with the implementation process, and confounded data have significantly complicated efforts to extract meaningful insights from experiences to date. As a result of this limited evidence base, making this decision required a leap of faith.
A primary goal of the ACA was to increase access to health care services, largely through major expansions of state Medicaid programs in 2014 and beyond. One of the key lessons from the 2006 Massachusetts health care reform law, on which the ACA was modeled, is that increases in insurance coverage do not ensure increases in access if provider capacity is strained (Long 2010). For this reason, the ACA's architects took several measures to strengthen the US medical care workforce. Among the law's broader measures were to increase funding for federally qualified health centers and to enhance medical education capacity (Kaiser Family Foundation 2013). Another measure, the fee bump, specifically targeted primary care physicians' (PCPs') willingness to participate in Medicaid (Miller 2013; Sommers, Swartz, and Epstein 2011), which has been declining over time (Decker 2012, 2013).
The fee bump, like other reimbursement-based efforts to increase PCP participation in Medicaid, had important limitations known to the provision's authors. Most recent studies of fee increases have found that their average effects on PCP participation are only small to moderate (Berman et al. 2002; Coburn, Long, and Marquis 1999; Fanning and de Alteriis 1993; Perloff, Kletke, and Fossett 1995; Perloff et al. 1997; Wilk 2013; Zuckerman et al. 2004, though effects will be more significant for some physicians than for others (Wilk 2013). Thus, quite substantial fee increases may be required to achieve meaningful improvements in PCP participation.
Yet the magnitude of the ACA's fee increases is quite large in some states but very small in others. This variation emerged because the legislation provided for the federal government to fund fee increases up to Medicare levels from each state's fees as of July 1, 2009 (prior to the ACA's passage), which themselves varied dramatically (CMS 2012b). Similar variation in Medicaid fees across states — ranging from 33 percent to 135 percent of Medicare fees — was documented for the year 2012 by Stephen Zuckerman and Dana Goin (2012); we present this variation in figure 1. In states where the fee increases are smaller, the fee bump may have little effect.
Many studies of fee incentives have also identified multiple states with relatively high fees but low participation rates and vice versa, making it unclear whether low fees are a principal impediment to PCPs' participation in Medicaid. Among the other reasons researchers have identified are delays in payment, burdensome administrative processes, the perception that Medicaid patients may be more difficult to care for or to refer for specialist care, organizational barriers, and worries about Medicaid beneficiaries churning on and off Medicaid rolls (Cunningham and Nichols 2005; Cunningham and O'Malley 2009; Davidson 1982; GAO 2011; Hadley 1979; Long 2013; Wilk 2013).
In addition, PCP participation in Medicaid is only one of several factors that determine Medicaid beneficiaries' access to care. For many Medicaid beneficiaries, the most significant determinants of access to care may include the capacity of other safety-net health care providers, proximity to physician offices, access to transportation, and the availability of care providers with sufficient spoken language capabilities or cultural sensitivities (Adams, Gavin, and Benedict 2005; Fossett and Peterson 1989; Fossett et al. 1992; Kim, Norton, and Stearns 2008; Long and Coughlin 2001–2; Syed, Gerber, and Sharp 2013; Weech-Maldonado et al. 2001. These factors could be expected to change (at most) marginally in response to primary care fee increases.
Despite these limitations, the writers of the ACA focused on increasing Medicaid fees for primary care physicians as a means of increasing Medicaid beneficiaries' access to primary care because it could be addressed in federal legislation more easily than other issues and perhaps because it was thought relatively straightforward to implement. The provision was made temporary to help keep the law's overall cost projections down and because evidence that the fee increase would work was limited.
Difficulties in Implementing the Fee Bump
Like many of the ACA's provisions, the fee bump has been more difficult to implement than anticipated, let alone to monitor and evaluate. One reason is that the ACA's many high-priority provisions were very demanding of Medicaid agency staffs' limited capacity. As a result, Medicaid agencies could not dedicate significant resources to finalizing their implementation plans as far in advance as they would have liked. In particular, although the fee bump's planned effective date was January 1, 2013, any significant implementation steps taken in advance of the federal Department of Health and Human Services' (HHS) formal guidance could well have been contradicted. To the frustration of many state policy makers, this guidance was not issued until November 1, 2012 (CMS 2012a), just two months before the fee increase was to go into effect.
Moreover, despite its length, the ACA left open important questions about many of its provisions. While the law clarified several components of the fee bump, including which services would be considered primary care (evaluation and management or vaccination services) and how the elevated fees would be paid for, most states still grappled with two key questions: First, which providers would be eligible for the fee bump? Second, what would it mean to pay increased fees for primary care services through the managed care organizations (MCOs) currently administering most Medicaid benefits?
On the first question, HHS clarified in its final rule who would be eligible, including nurse practitioners (CMS 2012b). However, the guidance also noted that non-board-certified PCPs would need to be offered an opportunity to attest that at least 60 percent of their services were primary care. States hurried, often with the assistance of administrative contractors, to establish such self-attestation systems for PCPs, but this added delays to the implementation process.
As for how to administer the fee bump through MCOs, HHS provided guidance only on several peripheral issues and concluded that the fine details would need to be negotiated between each Medicaid agency and its MCOs. How the ACA's prescribed fee-for-service fee increases should be translated into payment increases in the more complicated reimbursement systems MCOs often have in place, such as capitated or partially capitated systems, could be determined only through contract amendments.
HHS's decision reflects the adage that “if you've seen one state's Medicaid program, then you've only seen one state's Medicaid program.” Due to the considerable variation across states in the magnitude of the role played by MCOs in the Medicaid program overall (from 0 percent in Alaska, New Hampshire, and Wyoming to 100 percent in Idaho, South Carolina, and Tennessee [see fig. 2]), in the mix of established reimbursement systems they employ, and in state contracting procedures, HHS could not provide further guidance.1 States were left with unprecedented terms to negotiate. Because Medicaid's contracts with MCOs are proprietary, how much progress states have made in their negotiations is unclear.2 Anecdotal evidence suggests that MCOs in many states did not pay enhanced Medicaid fees to their participating PCPs until late spring or summer 2014. Thus, in states where Medicaid benefits are administered largely or entirely through managed care organizations, unsuccessful negotiations with MCOs could have neutralized virtually all of the provision's effects.
After working through many of these issues, more than forty states took the first formal step toward implementing the fee bump by March 2013, submitting to the Centers for Medicare and Medicaid Services (CMS) applications to amend the state plans that govern their Medicaid programs. Under the terms of their shared financing agreements with the federal government, state plan amendments would be required for any such fee increase, with some states requiring legislative approval (Families USA 2012). Some states, such as California, experienced further delays at this point in the process, depending on how long negotiations with MCOs ran before the application was submitted, how quickly these applications could be prepared and approved internally once green-lighted, and how quickly CMS could review and approve the applications. Reviewing state plan approval documents, we found that only one-third of states applying for a state plan amendment had received approval by June 1, 2013, though this fraction increased to more than 90 percent by October 1, 2013 (AAFP 2013).
How the Evidence Available to Legislators Was Affected
These compounding delays and the compressed time frame of state regulators' work to implement the fee bump had important consequences for subsequent legislative debates. While HHS clarified that CMS would approve state plan amendments requesting retroactive reimbursement of physicians for services rendered as of January 1, 2013 — even if they were not approved until several months later — states could not guarantee the fee bump would go into effect at any point before undertaking these administrative processes. Uncertainty and delays likely discouraged some PCPs considering increasing their participation in Medicaid during 2013, though how much increases in PCP participation were slowed as a result is unclear.
A related issue is that these state plan amendments and (in some states) their associated legislative approval processes can be highly opaque to many PCPs, especially those in small practices. Such practices may make decisions about whether to accept more Medicaid patients only on a quarterly or annual basis and only after the full details of their states' state plan amendments are released. Thus, it may take time for this information to be disseminated, digested, and acted on, particularly given the wide variety of other developments reshaping health care in America.
Furthermore, no PCP practice's payer mix can change overnight. Thus, even the most current data available to legislators about Medicaid beneficiaries' access to care almost certainly do not reflect the full, steady-state effects on PCPs' Medicaid participation of the fee bump. Anecdotal evidence from interviews with state policy makers and leaders at the American Medical Association Advocacy Resource Center and the National Association of Medicaid Directors suggests that some states are observing increased Medicaid participation by PCPs as a result of the fee bump, while others are seeing no effect. However, because of the many delays and implementation problems associated with the fee bump, no estimate of its effects on PCP participation in Medicaid has been current or comprehensive.
Early Policy Debates
Federal and state policy makers, keenly aware that PCPs would follow their deliberations closely before deciding whether to increase their participation in Medicaid, have begun debates on whether to extend the fee bump in 2015 and later years.
At the federal level, the American Medical Association and other physician advocacy organizations made a priority of lobbying Congress to extend the fee bump. They even included among their proposals extending the fee bump for as short a period as only six months so that the policy would sunset at the end of many states' fiscal years in June. Doing so would have given state legislators extra coverage to work out their annual budgets with fee bump provisions.
Congress faces two broad political realities in more permanently increasing Medicaid reimbursement rates for primary care services. The first is the great difficulty of passing legislation of any kind due to increased polarization in Washington, DC. Congress is on pace to pass fewer bills than at any point in decades (DeSilver 2013). Second, the ACA remains unpopular, with only 35 percent of Americans saying that they have a favorable view of the reform and with three-fourths (74 percent) of Republicans opposing the law (Hamel, Firth, and Brodie 2014). It is very unlikely that the Republican-controlled House of Representatives, which has voted more than forty-five times to repeal the ACA, will vote to extend part of the law. The best hope of congressional action for a sustained fee increase may be as a smaller component of a larger deal, thereby disconnecting the vote from this particular issue.
In the absence of congressional activity, many states have striven to identify feasible and politically palatable ways to extend the fee bump. State proposals have varied meaningfully, including temporary extensions of the fee bump, permanent extensions, and expansions of the set of services eligible for increased fees to include specialty care. Some states considered passing extensions through budgetary processes rather than legislation, to avoid floor votes on ACA-related issues. States choosing this path would have paid for the increase on their own rather than using federal funds. In many states the fee bump has been a nonstarter altogether because of significant budget concerns. Medicaid is already the largest or second-largest line item in most state's budgets (NASBO 2012), and the costs associated with fee bump extension legislation are substantial in some states.3 Opponents of the ACA have also argued that such state-driven fee bump extensions represent hidden costs of federally supported Medicaid expansions, thereby seeking to undermine the political feasibility of both provisions.
Robust evaluations and evidence regarding the effects of the fee bump would contribute significantly to these and future debates on physician fees in Medicaid. Without such information, more ideologically driven arguments have become the focus. In this challenging political climate, many states have experienced great difficulties implementing other major elements of the ACA, such as health insurance exchanges (Jones, Bradley, and Oberlander 2014) and Medicaid expansions (Jacobs and Callaghan 2013). Similarly, in only six states have advocates passed legislation extending the fee bump for 2015: Alabama, Colorado, Iowa, Maryland, Mississippi, and New Mexico (Galewitz 2014). That so few states' advocates have succeeded in this way to date bodes ill for other efforts to extend the fee bump or to reinstitute increased fees in later years.
What Should Be Done for 2015 and Beyond?
Difficulties with implementing the Medicaid fee bump mean that federal and state authorities will not have reliable and complete data about its effects in the near future. As such, whether states should extend the increase into 2015 and later years is unclear. However, since the debates proceed despite the absence of good data, policy makers weighing the alternatives would be well advised to acknowledge three important concerns.
First, since the principal goal of the fee bump provision is to increase Medicaid beneficiaries' access to primary care services, achieving this goal may be possible through other means. While fee increases have not been shown to dramatically increase physician participation in Medicaid, efforts to reduce payment delays, administrative burdens, or the risks of beneficiaries churning off Medicaid may affect PCP participation in Medicaid as much or more. It may also be possible to achieve improvements in access through a more selective application of reimbursement-based incentives, directing those incentives to types of care providers more likely to respond to them. Moreover, efforts focusing on other determinants of patient access to care (e.g., MCOs' provider network restrictions, transportation, language barriers) have significant potential to improve Medicaid beneficiaries' access to care.
Second, as long as any fee bump extension is temporary in nature, those PCPs who are not currently participating may be unlikely to open up their panels to Medicaid patients. They may worry about the financial hardship to come when the fee bump sunsets and reimbursement rates return to prior levels. Where nonparticipating physicians stay out of Medicaid despite the fee bump, such temporary extensions amount to little more than a few extra dollars in the pockets of the PCPs who have cared for Medicaid beneficiaries all along. Therefore, it may be more sensible to amend Medicaid reimbursement systems in ways that are predictable and sustainable over periods considerably longer than two years.
And third, across more than half of the United States, Medicaid expansion has already begun. Since Congress likely cannot be counted on to enact fee bump–related legislation anytime soon, the pressure brought on by these expansions will be felt most strongly by expansion states. With no real prospects for finding “clean” estimates of even the earliest actual effects of the fee bump, policy makers in only six states (including two — Alabama and Mississippi — that have not expanded their Medicaid Programs) have taken the leap of faith that maintaining increased fees will lead to increased participation in Medicaid among PCPs. And they have done so despite persistent concerns that other factors, like the administrative burdens and delays in payment, may still make Medicaid a less attractive payer to most physicians than Medicare or private insurers. Assuredly, this decision has not been an easy one. We hope clear evidence can be drawn from the experience of these six states to inform Medicaid fee policy in future legislative sessions.
We thank John McDonough and Colleen M. Grogan for their helpful comments.
1. Seventy-four percent of Medicaid beneficiaries were enrolled in a managed care plan as of July 1, 2011 (CMS 2011), though enrollment varied from 0 percent to 100 percent across states. Not all states administer Medicaid managed care programs through third-party contractors: Oklahoma administers its own managed care program, for example. For such states, implementing the fee bump may have been easier because doing so would not have required negotiations with external contractors.
2. A selection of “model contracts” from 2008 and 2011 is made available through the New York State Department of Health website (NYSDH 2013).
3. As shown in figure 1, Medicaid fees for primary care services as a fraction of Medicare fees for the same services varied substantially across states in 2012; variation in 2009 was comparable (Zuckerman and Goin 2012). As a result, the corresponding magnitude of the fee bump also varies significantly across states.