Although it is too early to assess the effect of ACOs and other integrated delivery systems on safety net hospitals, such initiatives—if widely adopted—could have a profound impact on safety net hospitals by restructuring the delivery system to emphasize and incentivize primary care, care management, and care coordination with specialists. By design, such restructuring is intended to reduce the need for hospital inpatient and emergency department care. Already, hospitals and health systems are seeking greater alignments and partnerships with primary care providers, both to form ACOs and to provide a buffer against the expected changes in the way care is delivered and paid for in the community (Witgert and Hess 2012). Competition for patients will increase as more private and public purchasers of care require or incentivize patients to use some form of integrated delivery system.
In many communities, it will be crucial for safety net hospitals to participate in and even lead integrated systems in order to compete for patients and be included in health plan networks. A number of safety net hospitals have already developed or are a part of integrated care systems. For example, Cambridge Health Alliance (a major safety net hospital in the Boston area) provides primary care, pharmacy, and behavioral health care for Medicaid and uninsured patients (Witgert and Hess 2012). Denver Health is another fully integrated system that includes community health centers, other primary care providers, a Medicaid plan, and even public health functions. Some integration efforts are community wide, such as the Camden Coalition of Healthcare Providers in New Jersey and the Medical Home Network in Chicago.
Another way of integrating care is through local programs that seek to provide and manage a comprehensive set of services for low-income, uninsured people as if they have insurance coverage. With safety net hospitals as core components of the provider networks, the Health Advantage Program in Indianapolis, Boston Medical Center’s HealthNet, and the Healthy San Francisco program create broad delivery systems with community health centers and others in an effort to help direct patients to the most appropriate services; such systems may give these communities a leg-up on identifying and helping transition uninsured people into coverage.
Some safety net hospitals will be well positioned to form or participate in ACOs and other integrated systems, especially for their Medicaid patients. Safety net hospitals that have operated managed care plans have experience in taking on the financial risk of providing care to Medicaid patients. In addition, many safety net hospitals have already developed partnerships and collaborations with other providers in the community as part of previous efforts to coordinate and integrate care for the uninsured and Medicaid populations (Cunningham et al. 2012; Hall et al. 2011). For example, Los Angeles County operates an extensive safety net system consisting of three acute care hospitals, a rehabilitation hospital, and multiple primary care and specialty care sites as well as a Medicaid managed care organization, LA Care. It is partnering with other safety net hospitals, community health centers, and private practice physicians to form a regional ACO for Medicaid patients, with a goal of incorporating patients with other types of insurance in the future (Felland et al. 2013).
There are a number of barriers to being part of integrated systems that some safety net hospitals will encounter, however. For example, some safety net hospitals may believe that forming integrated care systems is inconsistent with their mission. Or doing so might require a legal change to their mission that could be politically difficult with their key constituencies (Witgert and Hess 2012). Similarly, questions about governance—who owns the system and who is represented on the board—could be an issue for some safety net hospitals, when the hospital and ACO organization have conflicting requirements for the composition of board membership (Shortell and Weinberger 2012).
Funding for infrastructure, staffing, and training to support integrated care may also be a barrier for some safety net hospitals (Ku et al. 2011). Infrastructure includes not only the physical facilities to expand capacity and purchase new equipment, but also upgrades to health information technology necessary to coordinate care with other providers and monitor utilization of services by patients as well as key quality indicators. Many safety net hospitals lack the access to capital, margins, or cash reserves needed for such infrastructure improvements and will need assistance from both public and private sources. As part of the American Recovery and Reinvestment Act (ARRA) of 2009, the Health Information Technology for Economic and Clinical Health (HITECH) Act includes provisions to support and incentivize providers to adopt and use health information technology to improve health care quality and care coordination; this includes extra support for some types of safety net providers such as Critical Access Hospitalsand other small rural hospitals (Gold et al. 2012; Heisey-Grove et al. 2012). Establishment of Regional Extension Centers through HITECH prioritizes engagement with safety net providers in order to reduce disparities in health care that may arise from the “digital divide” (Heisey-Grove 2012). Some states have used Section 1115 Medicaid Waivers to support development of the infrastructure needed for integrated care. For example, California’s “Bridge to Reform” Medicaid waiver included Delivery System Reform Incentive Payments (DSRIP) to safety net hospitals for such purposes (Harbage and Ledford 2012).
In addition, the risk profile of many patients who use safety net hospitals may make inclusion of the hospital or many of its patients less attractive to an ACO, which may wish, for example, to include only private and Medicare patients, who have a better risk profile as well as higher reimbursement than Medicaid patients. Risk-adjustment methodologies used to define spending targets may or may not adequately account for the high clinical risk of Medicaid patients; it will be more difficult to account for other aspects of being socially disadvantaged, such as language barriers, social isolation, and the lack of other critical social services that could make it harder to realize cost savings and improvements in quality of care (Lewis et al. 2012). For the most vulnerable patients, coordination of care with other social services will be necessary to realize the cost and quality benefits of integrated care systems. Some safety net hospitals, such as Wishard Health Services in Indianapolis, have invested in housing units for their homeless patients as a way of meeting patients’ primary care and social services needs while saving the hospital money by reducing their use of the emergency department (Katz et al. 2011).
Failure to participate in or adapt to the new, integrated delivery systems could have significant consequences for safety net hospitals and their patients. Their ability to retain existing patients and compete for newly insured patients could be undermined if health plans (both private and Medicaid) require or strongly incentivize enrollees to use providers who are part of ACOs or other integrated delivery systems. In addition, to the extent that integrated delivery systems improve the quality of care to patients—such as through better access to primary and specialty care, fewer emergency department visits, and fewer inpatient stays for ambulatory-care-sensitive conditions—these improvements will not accrue to patients of safety net hospitals that do not participate in integrated delivery systems.