HEALTH AND LONG-TERM CARE PROGRAM INTEGRATION

The health and long-term care service delivery system is compartmentalized. Physicians, usually the primary care physician, are responsible for outpatient care. Inpatient care is usually under the management of a medical specialist, such as a surgeon, cardiologist, or neurologist. Care after the patient leaves the hospital includes home health care, rehabilitation, and skilled nursing home care. These services, although authorized by a physician, are usually more directly managed by a nurse or therapeutic specialist (e.g., a physical or occupational therapist). Extended care, whether in a nursing home or in the community, is often considered to be long-term care. Long-term care includes "skilled" services provided by a nurse or therapist, as well as the "nonskilled" care provided by a nurse's aide, personal care aide, or housekeeping services. The vast majority of nonskilled long-term care services are provided by family members, but they can also be purchased from private individuals and community agencies. Home or community-based long-term care services, if not directly coordinated by the patient or a family member, are typically planned and monitored by a social services case manager.

The distinctions among these multiple levels of care, and between skilled and unskilled care, are directly connected to the financing and reimbursement processes. The major public insurance program for health care is Medicare. This federally financed program provides coverage for physician, hospital, and skilled home health and skilled nursing home care. Extended or long-term care and unskilled care are usually excluded from Medicare reimbursement. Payment for these services comes either from the patient or the patient's family; or from public programs such as Medicaid or the Veterans Administration (for those who qualify). State governments also use social service programs to finance some unskilled care.

Historically, each level of care and each financing program has operated within its own budget. Under this arrangement, any cost savings from the substitution of lower levels of care for higher ones add costs to the program financing the lower-level services and produce savings for programs financing the higher levels of care. For example, reduced days in a hospital save Medicare costs, and may result in transfers to nursing homes, which have a lower daily cost than hospitals. Nursing home stays of up to one hundred days are likely to be paid by Medicare, but more extended stays usually are paid for privately or through Medicaid. Another example is the decision by a family to keep a member at home rather than place him or her in a nursing home. For those who qualify for Medicaid coverage, the availability of informal family care saves Medicaid costs, and pushes all the cost to the family. For the relatively few persons with private long-term care insurance, this would be a cost savings to insurance as well.

State governments have tried to encourage informal care by permitting persons eligible for Medicaid (who also qualify for nursing home admission) to receive a portion of the funds that would have gone for nursing home care for the purchase of unskilled home care. This is done within the Medicaid program under home and community-based care. This approach begins to solve the problem of limited financing for community-based, long-term care services (for those who have incomes and other assets low enough to qualify for Medicaid), but it does not integrate health care and long-term care.

Several promising approaches to acute and long-term care integration are briefly described here: the Program for All-Inclusive Care for the Elderly (PACE), the social/health maintenance organization demonstration, screening and care coordination in Medicare HMOs, the use of geriatricians and geriatric care teams, and managed care within the Medicaid program.

Program of All-Inclusive Care for the Elderly

The PACE program was initiated in 1986 as a congressionally mandated demonstration. It has since become a permanent program with authorization for several dozen sites around the United States. PACE offers complete coverage of all Medicare-reimbursed services (including hospital, physician, home health, and skilled nursing homes) and Medicaid-covered services. Medicaid coverage includes the same services reimbursed by Medicare, as well as long-term nursing home care, homemaker services, and adult day health care. The PACE program is targeted exclusively on persons whose needs make them eligible for nursing home placement. These programs are at full financial risk for all Medicare- and Medicaid-covered services used by their members. They receive a fixed monthly payment for each member, regardless of the care received. This is called a capitation payment. Under this reimbursement system the PACE sites can allocate Medicare and Medicaid revenues to whatever services, including non-health care services, deemed most appropriate. Having all services covered under a single budget permits the PACE sites to substitute lower levels for higher levels, when appropriate, and to capture the cost savings. Health and long-term care service integration is further achieved through adult day health care participation of the members, which permits the routine monitoring of functional and health status, medication management, and involvement in physical exercise and other enrichment activities. Care planning is done through multidisciplinary teams who work together rather than through professionals working in isolation.

A difficult challenge for PACE has been maintaining enrollment levels (Irvin, et al.). This is largely the result of program requirements for adult day care participation and the loss of freedom to choose one's physician (participants must use a PACE physician, who is usually a staff doctor). Several sites have been modifying their operations to place less reliance on day care and to give members the option of retaining their own physician.

While this has been a highly popular program with federal and state policy makers, there is still some question about the reimbursement payment levels, and whether the sites are receiving overpayment (Mukamel et al.). This concern arises, in part, because the sites can be selective in the referrals they accept into the program.

Social/health maintenance organization

The first generation of this Medicare demonstration, known as the S/HMO, was implemented in 1985 with the objective of adding a package of chronic care benefits to the acute services and operational structure of the Medicare HMO model. Chronic care benefits included nursing home stays (usually a maximum of thirty days), and personal care, homemaker, and case management services. There was an annual (and in some cases monthly) limit on the amount of services a member could receive. S/HMOs also offered expanded care benefits (such as prescription drugs, eyeglasses, transportation, and preventive dental care) to all members.

The S/HMO demonstration was testing the efficacy of offering and managing access to chronic care benefits (e.g., nonskilled home care), and examining how the expansion of responsibility into community-based care affected the health plan's general approach to its Medicare eligible members. Four sites participated in the initial demonstration. These first generation S/HMOs used a traditional model of outpatient and inpatient physician services and hospital utilization control, with each of these functions operated independently from the functional assessments, and chronic care benefit authorization and case management available to frailty qualified members. Contacts between physicians and case managers were limited—usually to the authorization of Medicare services (Harrington et al.).

In 1995 the Health Care Financing Administration (the agency responsible for the Medicare program) initiated the planning of a second generation model of the S/HMO. The second generation model retains the chronic care benefit package implemented in the first generation plans, and it adds several fundamental refinements intended to better integrate and coordinate primary care and the management of high risk cases. Most important of these was an attempt to develop a strong geriatric service model of care. The "geriatric" approach includes a screening program intended to identify patients at risk for high service costs and disability; timely application of primary care monitoring and treatment to reduce illness and disability; and a geriatric education and consultation program to provide specialty support for complex cases. Care management supports the primary care functions for those requiring home care, those discharged from hospitals or nursing homes, and those who are having difficulty complying with their treatment regimen. Case management efforts are closely integrated with the provision of primary care, including conferences among the various professionals. The proactive attention to clinical care and preventive services requires the definition of "risk" to include acute and chronic conditions and problems, in addition to the limitations of activities of daily living that have been more typical in the long-term care field (Kane et al.). One second generation plan became operational in 1997.

Screening and service coordination in Medicare HMOs

Surveys of the largest Medicare HMOs show that as early as 1990, plans had begun to establish procedures for identifying high risk patients, assessing and treating multiproblem patients, rehabilitating patients following acute events, reducing medication problems, and expanding benefits to include more home care and case management for nursing home patients (Kramer et al.). These activities incorporate many of the features of the S/HMO model, but without the advantage of a 5 percent higher capitation payment and with no obligation to provide long-term care benefits for those in the community.

Screening and assessment are generally limited to new enrollees or to those who have been hospitalized. High risk cases flagged via this process are referred to primary care (or geriatric assessment) for more in-depth assessments. Screening data are based on self-report questionnaires and telephone interviews. The emphasis in these instruments is given to health conditions and other factors that may be associated with hospitalization, preventable disability, and other avoidable expenditures. The screening of current members has been less formalized. It is generally based on a referral from the primary care physician or triggered by hospitalization (Pacala et al.). Management information systems are becoming more capable of capturing medical records, service encounters, and even prescription refills. Access to these data increases the likelihood of their use to identify and monitor those thought to be at risk for expensive care or avoidable complications, though there are no recent studies documenting the extent or means of implementing such monitoring.

Primary care for post-acute care patients is delivered by a combination of methods (Kramer et al.). For those returning home, most plans rely on the primary care physician. For persons in skilled nursing homes, some plans rely on the primary care physician, others have a medical director who handles all such cases, and still others use a nurse practitioner. Combinations of these approaches may occur, depending on whether the nursing home is a primary referral site for the health plan or a freestanding facility with only a few plan members. The management of other "high risk" but not hospitalized cases largely falls on the primary care physician, although many plans have developed programs targeted to selected diagnoses that include congestive heart failure, diabetes, and chronic obstructive pulmonary disease. Within disease management programs the patient's health care may be managed by a specialist who is supported by a team of professionals responsible for helping the patient monitor symptoms, diet, weight, medication use, and other treatments. The effectiveness of the screening, risk identification, and disease management approaches within health plans has not been formally reported, but experience from clinical trials and other small scale demonstrations leads to expectations of program success (Fama, et al.; Miller and Luft).

Geriatricians and multidisciplinary teams

The extent to which geriatric medicine should, and can, be integrated into the delivery of health care and long-term care is not yet resolved. Historically, some HMOs have used geriatricians as part of their primary care practitioner group but have not allowed these physicians to limit their practice exclusively to geriatric patients. Another model for practice is to have geriatricians (or a geriatric team) provide care and management to the most frail and vulnerable elderly within a system. This is implemented through screening programs, such as those noted earlier, that identify and refer new members who are frail (or current members who have become frail) into this specialty practice for ongoing primary care. Such a model requires a large elderly enrollment to generate a cost-effective practice volume. A variant on this model uses geriatricians as specialist consultants for assessment and advice with ongoing treatment. The patient remains under the care of the regular primary care physician. These two models can be used in combination (Friedman and Kane).

A further variation on the above involves the use of multidisciplinary teams (i.e., nurses, social workers, and/or other health professionals) in conjunction with the primary care physician. This model recognizes that geriatric training is generally more common among nurses and social workers, and it allows for case monitoring to occur through means other than office visits and to encompass care plans that go beyond purely medical treatment. In many cases geriatric nurse practitioners (GNPs) or adult nurse practitioners may assume responsibility for basic primary care, freeing the geriatrician's time for more complex cases. These team models can operate as components in ambulatory care clinics or as adjuncts to the home care program. Such teams are perhaps more common in hospital inpatient and nursing home settings. Under these inpatient circumstances the team likely replaces the patient's primary care physician until the patient returns to the community.

One area where GNPs have been used to good advantage is in providing primary care to nursing home patients. A corporation, EverCare, has developed a cost-effective managed care approach to capitating the acute care of nursing home patients, using GNPs as the major source of primary care. The underlying premise of this approach is that closer attention to the nursing home residents' primary care needs will reduce the use of more expensive hospital care (Kane and Huck). EverCare is being demonstrated at six sites across the country. As it has been implemented, some changes have been necessary. The participating physicians are recruited from those in the community who are already active in nursing home care. Because sufficient numbers of GNPs are not always available, adult nurse practitioners are sometimes used instead.

The relative value of any of these approaches is largely untested, but as a practical matter the choices made by a health plan and medical groups more generally are constrained by the size of their Medicare enrollment and by the limited availability of geriatricians.

Medicaid managed care

Many federal agencies, state governments, providers, and foundations are interested in using managed care for integrating acute care and long-term care. However a number of unresolved practical questions have slowed progress in this area. Two of the most important issues are discussed here. First, combining acute care and long-term care under a single entity raises questions about auspices. Many proponents of long-term care fear that a merged authority will be dominated by a medical mentality and that important social dimensions of long-term care will receive less attention. A second issue has to do with the experience base upon which to establish payment rates for both acute and long-term care services. PACE is the only program that has used a capitation payment blending funds from Medicare and Medicaid. In contrast to this, most state long-term care programs adjust Medicaid payments on the basis of expected Medicare reimbursement. Contractors in these situations are responsible for obtaining the Medicare portion of their payment. On the other hand, if Medicaid is capitated, Medicare usually remains fee-forservice—creating an incentive to shift costs to the fee-for-service payer. If acute care for both Medicaid and Medicare is capitated, then long-term care is usually fee-for-service—creating no incentive for capitated plans to keep members out of the long-term care system because Medicaid will pay for long-term stays. Data systems that can accumulate the full cost (across both Medicare and Medicaid) of the long-term care population or those at risk of entering the long-term care system are only beginning to be implemented.

Recognizing these important knowledge gaps, states have approached acute and long-term care integration in an incremental manner. A handful of states (Arizona, Florida, Minnesota, and Wisconsin) have enrolled elderly persons in Medicaid managed care programs that include varying levels of long-term care coverage (Holahan et al.). Four programs are briefly profiled. An initial common characteristic of these, and the other programs, is that none includes a direct role for the state regarding Medicare reimbursement.

Arizona. Arizona is the only state with a statewide managed care program for persons needing long-term care (Arizona Long Term Care System, ALTCS). The Arizona Health Care Cost Containment System contracts with managed care organizations for the coverage of all Medicaid enrollees. A Medicaid recipient is enrolled in ALTCS only if he or she meets Arizona's long-term care criteria. Through ALTCS the state pays managed care organizations (private companies in the Phoenix and Tucson areas, and county governments in the less populous counties) a capitated rate that covers the full range of community-based long-term care and nursing home care. ALTCS contractors are also responsible for the primary and acute care needs of their members, but this program covers only Medicaid costs; Medicare costs are reimbursed on a fee-for-service basis (McCall and Korb).

Florida. Florida has undertaken demonstration projects in three counties with two managed care organizations to examine the effects of managed Medicaid programs for elderly persons. Only recipients who meet the state's criteria for nursing home care are eligible. During the demonstration only those age sixty-five and older were eligible.

In these projects Medicaid is capitated (similar to the Arizona ALTCS program). State law prohibits the program from enrolling members in a Medicare risk contract. The benefits include in-home care, day care, transport services, supplies, and home adaptations.

Minnesota. Minnesota's Prepaid Medical Assistance Program enrolls elderly persons. It includes some community-based long-term care and up to ninety days of nursing facility care. Any additional long-term care is paid on a fee-for-service basis. Minnesota has initiated a waiver to Medicaid regulations so that it can establish the Senior Health Options Project. This program will integrate a full range of Medicare and Medicaid services for older persons who are dually eligible, regardless of whether they need long-term care. Nursing facility liability under this program will be limited to 180 days, but the rate structure provides incentives to purchase community care.

Wisconsin. The Wisconsin Partnership Program serves persons eligible for nursing facilities. The program currently operates as a Medicaid prepaid health plan (which is partially capitated). Medicare is billed by the providers on a fee-for-service basis. Program planners are seeking federal approval for full capitation of both Medicaid and Medicare services.

Robert Newcomer

See also Case Management; Day Hospitals; Long-Term Care; Medicaid; Medicare; Multidisciplinary Team; Veterans Care.

BIBLIOGRAPHY

Fama, T.; Fox, P.; and White, L. "Do HMOs Care for the Chronically Ill?" Health Affairs 14, no. 1 (1995): 234–243.

Friedman, B., and Kane, R. "HMO Medical Directors' Perceptions of Geriatric Practice in Medicare HMOs." Journal of the American Geriatric Society 41, no. 11 (1993): 1144–1149.

Harrington, C.; Lynch, M.; and Newcomer, R. "Medical Services in Social Health Maintenance Organizations." The Gerontologist 33, no. 6 (1993): 790–800.

Holahan, J.; Zuckerman, S.; Evans, A.; and Rangarajan, S. "Medicaid Managed Care in 13 States." Health Affairs 17, no. 3 (1998): 43–63.

Irvin, C. V.; Massey, S.; and Dorsey, T. "Determinants of Enrollment Among Applicants to PACE." Health Care Financing Review 19, no. 2 (1997): 135–153.

Kane, R. L., and Huck, S. "The Implementation of the EverCare Demonstration Project." Journal of the American Geriatrics Society 48, no. 2 (2000): 218–223.

Kane, R. L.; Kane, R. A.; Finch, M.; Harrington, C.; Newcomer, R.; Miller, N.; and Hulbert, M. "S/HMOs, the Second Generation: Building on the Experience of the First Social Health Maintenance Organization Demonstrations." Journal of the American Geriatrics Society 45, no. 1 (1997): 101–107.

Kramer, A.; Fox, P.; and Morgenstern, N. "Geriatric Care Approaches in Health Maintenance Organizations." Journal of the American Geriatrics Society 40, no. 10 (1992): 1055–1067.

McCall, N., and Korb, J. "Utilization of Services in Arizona's Capitated Medicaid Program for Long Term Care Beneficiaries." Health Care Financing Review 19, no. 2 (1997): 119–134.

Miller, R., and Luft, H. "Does Managed Care Lead to Better or Worse Quality of Care? A Survey of Recent Studies Shows Mixed Results on Managed Care Plan Performance." Health Affairs 16, no. 5 (1997): 7–25.

Mukamel, D. B.; Temkin-Greener, H.; and Clark, M. L. Health Care Financing Review 19, no. 3 (1998): 83–100.

Pacala, J.; Boult, C.; Hepburn, K.; Kane, R. L.; Kane, R. A.; Malone, J.; Morishita, L.; and Reed, R. "Case Management of Older Adults in Health Maintenance Organizations." Journal of the American Geriatric Society 43, no. 5 (1995): 538–542.