Abstract
State Medicaid programs and other state health agencies need to monitor and evaluate changes in health insurance coverage, access to care, financing, and the quality of health care delivery. The availability of new financial resources through the Patient Protection and Affordable Care Act is accompanied by raised expectations for such accountability. While state agencies often contract with universities on an ad hoc basis for specific policy projects, fourteen states have established formal state-university partnerships so that their analytic and technical needs can be addressed more readily. After a brief overview of these partnerships, this article provides examples of their projects, which most often affect Medicaid policy, including work on program eligibility, provider payments, and optional benefits. State-university partnerships are working on policy-relevant projects that influence decision making. Like the variation in Medicaid programs across the country, no two partnerships are alike. They thrive in a mix of structures, using different means of contracting, and with varied degrees of data access. All partnerships are interested in building a national network to share innovative practices and projects, spawn comparative policy studies across states, and support the development of new state-university partnerships.
Editor's Note: Thanks to funding from the Blue Shield of California Foundation and the Robert Wood Johnson Foundation, JHPPL has begun the coordination of an Engaged State Health Reform Research Network to bring together people from different backgrounds (practitioners, stakeholders, and researchers) involved in state-level health reform implementation to inform and extend health reform across the United States. A network website will document implementation projects across the country, workshops will be held, and JHPPL will publish essays under this new section based on findings emerging from network participants. All essays in the section will be published open access.
—Colleen M. Grogan
Introduction
Partnerships and collaborations between state health agencies and their in-state universities are not new, but they are growing, particularly as states monitor their Medicaid programs ever more closely. Whether or not a state chooses to expand its eligible population as allowed in the Patient Protection and Affordable Care Act (ACA), all states grapple with a growing low-income population and Medicaid spending that constitutes one of the largest items in their state budgets. Medicaid and other state agencies are also responding to new federal grant opportunities through the ACA that promote changes in health care financing and the organization of care. The availability of new financial resources is accompanied by raised expectations for state monitoring of the impact of changes in health insurance coverage on access to care, financing, and the quality of health care delivery.
Medicaid and other state agencies are increasingly meeting these responsibilities by partnering with public universities in their state (Coburn 1998; Coburn et al. 2007. While many state health agencies contract with university experts on an ad hoc basis to conduct specific policy-related projects, some states have established a more formal partnership with their in-state university to facilitate an ongoing collaboration. These state-university partnerships are typically created so that anticipated and unanticipated policy research and technical assistance can be addressed more readily while developing in-state policy experts and resources.
We used a snowballing outreach effort in which leaders of partnerships we knew were asked to identify additional partnerships. As a result, we identified sixteen state-university partnerships in various stages of planning across the United States. We interviewed and surveyed the university centers' directors and also interviewed many of the partnering state officials between October 2012 and October 2013. We collected information about how the partnership was structured and functioned, the nature of the state work, and examples of policy projects and their impact.
Overview of Current Partnerships
State-university partnerships currently exist in California, Florida, Georgia, Iowa, Maine, Maryland, Massachusetts, Michigan, New Hampshire, New Jersey, Ohio, Oregon, South Carolina, and Wisconsin. Two additional states, Connecticut and Minnesota, are in the planning stages but have not yet established any contracts.
Maine was the first state to establish its partnership, in 1989. Eight states followed in the 1990s, and five additional partnerships have been created since 2000. Most partnerships exist between a state agency and a single in-state public university, with the exception of California and Ohio, where multiple campuses of the university system contract with the state agency. Because public universities are publicly financed and administered organizations, state agencies can enter into interagency agreements with them and bypass the contract procurement process required with private institutions.
State Government Partners
All fourteen states' universities primarily collaborate with the Medicaid agency in their state, but in eleven states the universities also work with other public agencies on state health issues. Seven states also have relationships with their state legislature, sometimes under separate contract. Four of the partnerships work with Medicaid agencies outside their own state. The prime example is Oregon, which has organized a members-only network of states sharing Medicaid value-based purchasing studies including comparative drug effectiveness studies.
Means through Which Partnerships Can Influence Policy
We identified five ways in which universities assist state health policy makers through these partnerships. University leaders identified program evaluation as the most common activity they perform in conjunction with these partnerships (32 percent of their state projects). For example, states are required by the Center for Medicare and Medicaid Services to arrange for independent evaluations of Medicaid waivers and other discretionary funding opportunities. Universities provide a wide range of expertise for this multifaceted work.
A second application of the university's expertise in these partnerships is for policy research. Policy research often includes systematic literature reviews and analyses of existing data that are used to help states consider policy options. Partnerships report that policy research makes up approximately one-quarter of their work together.
Universities are called on to help design state health programs (15 percent of partnership activity) as part of a state plan amendment or a federal waiver. For example, a university may help design new payment models for long-term care or for acute care through an accountable care organization (ACO).
States also turn to their university partner for technical assistance (15 percent of projects) in areas where they do not have adequate staff capacity for operational functions, such as the building of an agency's data infrastructure or a case-mix adjustment tool.
Finally, 10 percent of the work undertaken by universities is educational, from coordinating student internships within Medicaid agencies to providing regular health policy seminars for state legislators.
State-University Partnerships' Impact on State Health Policy
All state-university partnerships work with Medicaid agencies — and in some states exclusively with their Medicaid agency — so that most of the examples of policy impact center on the policy options that Medicaid administrators have available to them. Medicaid leaders seek to manage their programs through choices they make in eligibility for coverage, provider payments, and optional benefits. We provide examples below in each of these areas.
Eligibility for Coverage
The Hilltop Institute at the University of Maryland, Baltimore County is working with Maryland's Department of Health and Mental Hygiene (DHMH) on Medicaid and Children's Health Insurance Plan (CHIP) expansion. Maryland passed the Kids First Act in 2008, using information from the state's personal income tax forms to identify children who may be eligible for Medicaid/CHIP but remain uninsured, and mailed families information about the two programs and how to enroll. In this policy-related work, the Hilltop Institute helped revise the tax form language with the comptroller's office and DHMH (Idala 2010), analyzed the state tax data (which required additional legislation to allow the comptroller to share data with the DHMH), and evaluated the strategy. The evaluation is partly funded by the Robert Wood Johnson Foundation through support for the Hilltop Institute (Spicer et al. 2012. While Maryland's Kids First Act faced several implementation challenges, other states should be interested in its findings, as the ACA now authorizes data sharing between federal agencies (including the Internal Revenue Service) and state health programs, opening up cost-effective opportunities to target outreach to potential Medicaid beneficiaries.
Wisconsin's Population Health Institute (UWPHI), based at the University of Wisconsin, works with both the Medicaid agency and the state legislature. A good case example of affecting health coverage was shared by UWPHI's Health Policy Group. About a decade ago, the group addressed questions about coverage for undocumented women, who were ineligible for Medicaid coverage for prenatal care, while the newly born child, as a US citizen, would be Medicaid eligible. The UWPHI's issue brief reviewed the Medicaid costs and complications of delivery, as well as subsequent care for the child (Friedsam 2004). Such analytic support led to a provision that extended coverage for prenatal care to undocumented women in the following year's state budget (Wisconsin State Legislature 2005). The UWPHI attributes this level of policy impact, in part, to its many public briefings for lawmakers and legislative and executive agency staff, as well as ad hoc, invitation-based dialogues and forums — all of which disseminate findings to a broad stakeholder audience and create a culture for evidence in the policy-making process (University of Wisconsin 2013).
Provider Payments
Partnerships have also contributed to Medicaid provider payment policies. For example, work done by the Center for State Health Policy (CSHP) at Rutgers University helped shape New Jersey's Medicaid ACO model. State legislation creating the Medicaid ACO program designates CSHP to support implementation and evaluation of the demonstration, and the center has developed metrics for the demonstration (DeLia and Cantor 2012). The CSHP is now providing technical assistance to Medicaid staff working on final regulations and will be supporting the implementation of ACOs in multiple communities, as well as evaluating their quality of care and cost effectiveness (Yedidia, Lontok, and Cantor 2013).
Universities can also support reforms in provider payment policy by working within state government to build technical capacity that otherwise would not be available. For example, the University of New Hampshire's Institute for Health Policy and Practice helped its state's Department of Health and Human Services (DHHS) and Department of Insurance (DOI) as they developed a statewide all-payer claims database (APCD). The institute has since worked with the DOI as it develops and maintains a Health Cost website that shows costs (based on paid claim information) for specific procedures and for specific providers for each distinct carrier. More recently, the Institute for Health Policy and Practice received a Robert Wood Johnson Foundation payment reform grant to create a set of metrics from APCD and clinical data for an ACO project in New Hampshire. Project participants cover approximately 25–30 percent of Medicaid and commercially insured persons in the state. The university designed the ACO attribution methodology and the evaluation measures, with input from insurance carriers, policy makers, and providers, based on data from the APCD. The participating providers are able to look at data such as cost per member per month for each organization, as well as a state comparator for nonproject participants.
Optional Benefits
University research can also help states decide which optional Medicaid benefits to cover by conducting research on their effectiveness. The Oregon Health and Science University's Center for Evidence-Based Policy works in coalition with state Medicaid programs and other partners across seventeen states to provide recommendations on comparative effectiveness of medications (through the Drug Effectiveness Review Project, DERP) and data-driven use of procedures and other technologies (through the Medicaid Evidence-Based Decisions Project, MED). The results often drive states' Medicaid benefit decisions. For example, Oklahoma's Medicaid agency requested a review of evidence on the benefits and risks associated with terbutaline pumps to prevent preterm labor. The center's MED project concluded the terbutaline pump had minimal effect on preterm labor while also causing substantial risks. The MED project also reviewed subsequent studies presented by manufacturers of the pump and found them to be of poor quality. The Oklahoma Medicaid agency used the results of the MED project's studies to convince the legislature to forego the creation of a mandate to cover the pump under Medicaid. In another case, the center's DERP project demonstrated that drugs in the most commonly prescribed group of antidepressants (selective serotonin reuptake inhibitors, SSRIs) are roughly equivalent in efficacy but variable in cost. This finding supported decisions in some states to adopt Medicaid formularies covering the lowest-cost SSRI first. The center's leadership believes its work “creates a more functional marketplace” for Medicaid as a purchaser, as it provides state Medicaid programs objective information about drugs and technologies to better guide their benefit decisions.
Policy research conducted by the California Medicaid Research Institute (CAMRI) has also been used to guide Medicaid benefit decisions. CAMRI was asked by the state's Medicaid program to rapidly assess the feasibility and costs of a newly legislated benefit: a home medication dispensing machine. Lobbyists had convinced state legislators that it could save $140 million annually by preventing adverse medication errors among those at risk. CAMRI, working with its partners at the University of California at Davis, prepared an evidence-based literature review of these devices and reexamined the cost model to assess potential savings. The CAMRI team questioned the validity of the cost savings estimate and concluded that the state would likely incur substantial new costs to cover the device that would not be offset by a reduction in claims for emergency visits and hospital and nursing home admissions. California's Medicaid leaders were able to use this evidence to successfully repeal the law, including the pilot program, its evaluation, and the benefit.
Challenges Partnerships Face in Addressing State Health Policy
Policy-relevant work of the type described above cannot wait for prolonged contracting, delayed access to state databases, or the right grant opportunity. State contracting can be a complex process, and the fourteen partnerships vary in the ways they contract with their state partners. Many partnerships write separate contracts for each of their projects — which are seldom executed quickly. Partnership leaders in these arrangements report that it is difficult to complete short-term projects (less than six months long) given the time needed to finalize individual contracts. However, having an overarching agreement can reduce the contracting time; six of the nine partnerships with such agreements report that they could complete a short-term project easily.
Data-sharing challenges exist in the majority of partnerships. Most partnerships write separate data agreements for each project and consequently report substantial study delays. The exception is when universities have direct access to state data by virtue of helping build the databases or have agreements to receive regular data transfers (in four partnerships).
State Medicaid agencies have limited resources for research and program evaluation. However, these activities are eligible for a minimum of 50 percent federal contribution provided that the state can demonstrate to CMS its financial contribution to the work and it identifies the work as an element of its program administration (Preston 2013). States can leverage their ability to draw federal matching dollars for Medicaid administration by attracting funds from private foundations, their legislatures, other state agencies, and universities themselves. However, adding an external funder to a state-university partnership may add complexity that requires additional time, negotiations, and reporting requirements.
Finally, university center directors, including those with long-lasting partnerships, discussed the challenge of working in the state political environment. Since officials in all the states were involved in setting the partnership's agenda, university teams are often constrained in what they can study, particularly when they are using the state's databases. State research and analytic projects are steered largely by the health and health care agenda of a regularly changing set of elected and appointed state officials. Priorities and personalities can change rapidly. Partnerships that have demonstrated an ability to be successful over time have built trusting relationships that typically go much deeper than just between the leadership at the university and the state. If the work of the partnership can evolve beyond valued but discretionary to instead become integral and necessary to the operating of the state agency, then a change in political leadership is less of a threat.
Conclusion
State-university partnerships function in various ways with respect to their contractual arrangements, the types of assistance the university provides to policy makers, and the policy focus of their work. Their successes in bringing evidence to the policy-making process suggest that these partnerships are a model worthy of greater attention, particularly for states where such a partnership is under consideration.
Collaboration to strengthen and expand these partnerships is timely not only because their role in state policy may expand under the ACA but because they could also collectively support the needs of third-party stakeholders, such as federal policy makers. All sixteen of the partnerships interviewed expressed interest in building a national network to share innovative practices and projects, spawn comparative policy studies across states, and support the development of new state-university partnerships. Support from foundations and federal agencies to enable sharing of best partnership practices among these state-university models could improve their efficiency and effectiveness while allowing them to evolve into a network that could address policy questions that could best be answered using a multistate sample or perspective.