Fully integrated managed care models

The Arizona Long Term Care System, Minnesota Senior Health Options, the Wisconsin Partnership Program, and the Program for All-Inclusive Care for the Elderly (PACE) are all well-established programs that primarily serve the dual eligible population on a statewide basis, or in the case of PACE, nationally.

Arizona Long Term Care System (ALTCS)

In 1989, Arizona created the Arizona Long Term Care System (ALTCS), a mandatory managed care program under which contractors provide a range of acute and long-term care services for individuals who are Medicaid eligible and at risk of institutionalization (McCall 1997). Coverage for medical care includes doctor’s office visits, hospitalization, prescriptions, lab work, and behavioral health services. ALTCS is targeted to individuals who fall into any one of the following categories: age 65 or over, blind, or disabled. The individual must be in need of ongoing long-term care. Approximately 48% of ALTCS participants live in their own homes or an assisted living facility and receive needed in-home services. ALTCS uses a formula for determining case load, and the Arizona Health Care Cost Containment System (AHCCCS) sets a specific rate for case management services separate from the capitated rate that covers health care services. ALTCS requires case managers to be in contact with clients living in the community at least once every 90 days, which differs from most plans that do not stipulate such a high frequency of contact.

ALTCS pays contractors prospectively on a capitated, per member, per month basis. Using utilization trend data and medical cost inflation factors, the blended rate is established by taking into consideration acute care, home and community-based care, institutional care, behavioral health care, and administrative costs. Two different capitation rates are established, based upon whether or not members have Medicare benefits (CMS 2007). Some of the eight contracted programs also have special established payment arrangements for disease-specific conditions (CMS 2007).

Minnesota Senior Health Options (MSHO)

The Minnesota Senior Health Options (MSHO) program began as a Medicaid 1115 demonstration in 1997 and was designed to test the cost and quality-related outcome effectiveness of an integrated Medicare/Medicaid service benefit for dually eligible seniors in four counties in Minnesota. Participating health plans were converted to Medicaid 1915 waivers in 2003 and became Medicare Advantage (MA) Special Needs Plans in 2006. The program continues to operate under Medicare Section 402 demonstration authority to allow for payment differences from other MA plans (CMS 2007; Malone 2004). Over 35,000 members are enrolled in MSHO (CMS 2007); of those, approximately 38% reside in nursing homes. Another 32% meet the criteria for nursing home placement but are being served in the community through home and community based services. MSHO contracts with non-profit health organizations to provide a full range of Medicaid acute, behavioral health, home and community-based services and also the full range of Medicare Part A, B, and D benefits (CMS 2007). The model’s centerpiece is its use of care coordination as the tool to improve quality and ensure efficiency. Each member is assigned a care coordinator, who works closely with both the primary care physician and the beneficiary to coordinate all medical and social service needs.

Wisconsin Partnership Program (WPP)

The Wisconsin Partnership Program (WPP), also a managed care plan for dually eligible individuals, began in 1995 with funding from the Robert Wood Johnson Foundation. WPP was converted into an integrated Medicare/Medicaid waiver program in 1999. It was designed to serve nursing-home eligible people over the age of 65 and adults (age 18+) with physical disabilities. Interdisciplinary teams coordinate all primary, acute, mental health, and long-term care services, with a nurse practitioner serving as the primary coordinator.

Capitation contracts are awarded to health organizations by the state to provide or arrange for the provision of all Medicaid-covered primary and acute care services, community based long-term care services, and nursing facility services. The Medicaid payment is 95% of the weighted average of payments made to fee-for-service participants, resulting in the State realizing an automatic 5% savings. In addition, there is an annual case mix adjustment to the rates. Medicare payments are based on hierarchical condition category (HCC) methodology, in addition to a “frailty adjuster” to account for the frailty of the population being served (CMS 2007).

Program for all-inclusive care for the elderly

OnLok Senior Health Services in San Francisco was the first program in the country to test a combined Medicare and Medicaid capitated approach to meeting the health and social support needs of nursing-home eligible, community-dwelling elders. The Program for All-Inclusive Care for the Elderly (PACE) was modeled after OnLok and was authorized under the Balanced Budget Act of 1997 (White et al. 2000).

PACE provides and coordinates the entirety of medical and social services for its members using an interdisciplinary team approach composed of a primary care physician, nurses, social workers, rehabilitation therapists, dieticians, and direct care workers. The team develops the care plan and delivers all services (acute care, medical, social, and when necessary, nursing facility care) in a seamless, ongoing and integrated manner. PACE centralizes its services around an adult day center.

PACE is a capitated benefit for older adults (age 55 or older) who, based on an assessment, are deemed to require a nursing-home level of care. PACE members can be eligible for Medicare or Medicaid or both. Medicare recipients who are not eligible for Medicaid pay monthly premiums equal to the Medicaid capitation amount. PACE organizations receive capitated Medicare payments from the Centers for Medicare and Medicaid Services and Medicaid payments from the state. At this time, the Medicare rate includes the “frailty adjuster” which accounts for the relative frailty of the member population. The Medicaid rate is reimbursed below a traditional fee-for-service equivalent. PACE providers assume full financial risk for participants’ care without limits on amount, duration, or scope of services.

Community-based socially oriented models

This analysis also looked to evidence from programs that are traditionally categorized as socially oriented, long-term care programs, but that also formally include coordination with at least some integration of medical health services provided to the recipients. Two programs met the criteria for inclusion, the Ohio PASSPORT program and the Wisconsin Family Care Program.

Ohio passport program

The Ohio PASSPORT Program (Pre-Admission Screening System Providing Options and Resources Today) is a Medicaid 1915(c) waiver established in 1984. It is the oldest Ohio waiver program and one of the largest waiver programs in the country. PASSPORT’s goal is to provide long-term services and supports to nursing-home eligible individuals to help them remain safely in their own homes. PASSPORT has two arms: first, a point-of-entry system that offers pre-admission telephone screening to determine Medicaid eligibility, assess care needs, and provide information about available service options, and second, a home care system, which includes case management and monitoring services. Once a person is determined eligible, a case manager is assigned to that person and works with the individual to craft an individualized care plan. The case manager primarily coordinates long-term support services but also arranges for some medical aspects of the person’s care (medical equipment purchasing and transportation for medical appointments). In order to facilitate integration of the medical and social support dimensions, the person’s physician must be made aware of and agree to the care plan created. Certainly, there is room for growth in the level of coordination with the health care needs of the older adult, since the current level of integration is minimal. The ground work for effective care coordination is in place, however, and lends itself to an enhanced connection with health care needs.

In support of the goal of making the program cost effective, Ohio requires that the PASSPORT program for each participant must not exceed 60% of the cost of comparable nursing home care. The Ohio Department on Aging with oversight by the Ohio Department of Job and Family Services established contracts with 13 agencies to administer the PASSPORT program. The administrative agency is not permitted to be the home care service provider and must contract out for services. In addition to the services provided in the PASSPORT care plan (primarily personal care), participants receive traditional medical Medicaid services.

Wisconsin Family Care Program (WFC)

The Family Care Program in Wisconsin (WFC, a combined 1915c and 1915b waiver) began in 1998, authorized to serve people with disabilities (developmental and physical) and frail elders. Its goals are to provide consumers with choices in services/residences, improved access to services, improved quality related to outcomes, and cost effectiveness. WFC has two organizational components: aging disabilities resource centers (ADRCs) and managed care organizations. The ADRCs serve as a point of entry into the long-term care system and provide an array of telephone-based and in-home services.

Managed care organizations (MCOs) offer the Family Care Benefit for both institutional and community dwelling individuals. The benefit does not include coverage for medical services (unlike ALTCS and MSHO) but does blend funding streams to offer consumers an individualized and flexible long-term care benefit. MCOs receive a capitated per person payment to manage and purchase services for their members. The Family Care benefit requires MCOs to use an interdisciplinary team approach to assessment and coordination.

Although medical services are not covered by the Family Care Benefit, these services are coordinated by the Family Care interdisciplinary team. Nurses on the interdisciplinary team coordinate the care directly with the medical providers and also may accompany the member to appointments. Any recommendations made by medical practitioners are brought back to the team, and the team then assists the member in following the recommendations.

Patient-centered medical homes

An evolving model of coordinated care is the patient centered medical home. The primary care physician practice serves as the central point for the delivery of services to its patients, and care is delivered in a holistic, person-centered way. Although many of these models are primarily medically focused, two have explicitly integrated medical care and long-term social support services: Community Care of North Carolina and the Vermont Integrated Care Pilot Program.

Community Care of North Carolina

Community Care of North Carolina (CCNC) began in 1998 with its foundations in the Medicaid program. The program is structured as an enhanced fee-for-service model with a designated dollar amount per member/per month to support care management activities. To expand the model to recipients of Medicaid and dually eligible individuals with chronic conditions and long-term care needs, CCNC has established a chronic care program that targets Medicaid eligible beneficiaries who are elderly, blind or disabled. This program also targets a defined subset of high risk/high utilization individuals for comprehensive case management. The model is structured so that primary care physicians are supported by a network of other community providers, including hospitals, health centers, and community social service departments (Steiner et al. 2008). CCNC currently includes 14 non-profit networks with over 3,000 physicians managing approximately 900,000 Medicaid recipients (80% of North Carolina’s Medicaid population). Case management is a core function of the networks, and case managers, employed by the networks, follow a standardized protocol to identify at-risk patients, assess the patients’ needs, and determine the level of intensity case management required.

Using case management and information technology systems as vehicles, the Chronic Care Program includes services such as disease management, prevention strategies for avoidable emergency department (ED) visits, hospital admissions, and readmissions (Wilhide & Henderson 2006). While CCNC does not explicitly address long-term care, it maintains a community focus and includes mental health referrals, social case management, family/caregiver involvement, and collaboration with community providers.

Vermont Blueprint Integrated Pilot

The Vermont Blueprint Integrated Pilot (BIP) shares similarities with the CCNC model, as it utilizes the patient-centered medical home as the source of primary care, with local interdisciplinary “Community Care Teams” composed of nurse coordinators, public health prevention specialists, Office of Vermont Health Access care coordinators, social workers, dietitians, and community health workers. This ambitious program does not restrict itself to older adults, although its focus on chronic conditions such as arthritis, heart disease, and diabetes ensures that older adults, who have a disproportionate amount of chronic illness compared to their younger counterparts, stand to benefit from these pilots. A linchpin of this system reform is the establishment of patient-centered integrated care models and the integration of health care, public health, and supporting social services to support population health. As with CCNC, the community-based nature of the program provides for social support and mental health services, but lacks an explicit focus on long-term care.

Financial integration is to be achieved through the mandatory participation of the three major commercial insurers and Medicaid, with a sliding care management fee linked to ten NCQA Patient-centered Medical Home criteria and plans for participation on the national level by Medicare. Finally, the program has a strong focus on the incorporation of health information technology as a means to improve the quality of care and facilitate a thorough evaluation of outcomes.

Other models of care coordination

Two additional programs are included in this analysis: the Veterans Administration Geriatric Evaluation and Management Program and Hospice (as a Medicare benefit). They are grouped for convenience as neither fits neatly in the preceding categories.

Geriatric Evaluation and (Case) Management (GEM)

Geriatric Evaluation and (Case) Management (GEM) is a widely evaluated model of case management and health care service delivery. The Veteran’s Administration (VA) began implementing the model with positive results in 1976. Although the VA has taken the lead and is most commonly associated with it, GEM has spread rapidly in the United States and abroad. GEM can be implemented in an inpatient unit, outpatient clinic, or primary care setting. It is based upon a multidimensional bio-psycho-social assessment conducted by an interdisciplinary team (most commonly comprised of a social worker, nurse practitioner and geriatrician). Using the Comprehensive Geriatric Assessment, the team creates a care plan that then coordinates medical/health, rehabilitation, education, and social service interventions with the aim of improving the quality of life for targeted individuals in a cost efficient manner.

Hospice

The Medicare Hospice Benefit is intended to provide compassionate and cost-effective care for Medicare beneficiaries with incurable advanced illnesses. Medicare’s very large expenditures on dying beneficiaries, combined with federal funding pressures, have given new prominence to end-of-life care. Hospice provides a combination of services designed to address not only the physical needs of patients, but also the psychosocial needs of patients and their caregivers. The team of providers engaged in hospice care includes doctors, nurses, home health aides, social workers, clergy/spiritual counselors, therapists, and volunteers.

Hospice has been a Medicare benefit for almost three decades, starting in 1979 when the Health Care Financing Administration (now CMS) invited and supported 26 hospice care demonstrations. Hospice became a Medicare benefit (temporarily) in 1984, and permanently in 1986. In 1985, Medicaid programs were given the option to cover Hospice as a benefit. Many private insurance plans also include a hospice benefit. Under Medicare, Hospice is a capitated benefit with daily prospective (and regionally adjusted) rates established for routine home care, continuous home care (24-hour care), inpatient respite care, and general inpatient care. Medicaid uses the same categorization. Physician services are reimbursed separately.

Table 4.2 Program Outcomes

Analysis of outcomes and program effectiveness

For all of the programs under analysis, the respective evaluations report changes in utilization patterns that demonstrate effective approaches to integrating health and long-term care services: decreases in hospitalizations, readmissions, and/or ED use, and decreases in nursing home utilization and costs. The methodology used for measuring these outcomes differed by program: some used a control group, others compared outcomes to another state’s utilization statistics and expenditures for a similar population, and yet others compared the community-based program costs to projected nursing home costs. Table 4.2 summarizes the outcomes found for each program, followed by brief descriptions of the most significant studies and findings for each program. Of the programs selected for inclusion, evaluation outcomes for Vermont’s Blueprint Integrated Pilots are not yet available, although a comprehensive evaluation is in progress and an outline of this evaluation will be provided. We have chosen to include this program based on the innovative aspects of its program design, the robust use of information technology to measure outcomes, and its potential as a model of the future to address the care needs of older adults through a PCMH model.

Compared to traditional Medicaid in New Mexico, ALTCS beneficiaries were found to use significantly less institutional care and more, but not significantly, ambulatory care. When compared to New Mexico, ALTCS beneficiaries had more home visits (case management and evaluation); however, the number of hospital days per thousand, per year was 22% lower (McCall 1997). Over a five year period (1989–1993) a comparison of traditional Medicaid in New Mexico to Arizona Long Term Care System (ALTCS) found ALTCS total costs, including medical, long-term support services, and administrative costs, to be 16% less. If only medical costs were considered, an 18% savings was achieved, totaling almost $290 million. The savings increased over time (McCall 1997). Community-dwelling Minnesota Senior Health Options Plan (MSHO) patients had significantly fewer preventable hospitalizations and ED visits when compared to patients in two control groups (Kane 2004). MSHO enrollees living in a nursing home had significantly fewer ED room visits, hospital admissions, shorter lengths of stay when hospitalized, and fewer preventable hospitalizations than control group enrollees (Kane 2004).

For people who had physical disabilities, The Wisconsin Partnership Program demonstrated a statistically significant decrease in hospital admissions when compared to a similar population. Lower rates of preventable emergency services were also achieved (CMS 2007). For specific diseases/conditions, the rate of hospital admissions for diabetes, congestive heart failure (CHF), bacterial pneumonia, and chronic obstructive pulmonary disease diminished compared to the year prior to enrollment. The length of stay during an admission for these conditions decreased 18.6% for diabetes and 71.6% for CHF (Landkamer 2005).

PACE achieved lower rates of hospital use, nursing home admissions, ED visits, and mortality rates and better quality of life than the control groups (MacAdam 2008). Ambulatory service rates were higher than the control group (MacAdam 2008). In a study by Abt Associates, PACE was found to show lower total overall costs – with lower Medicare and higher Medicaid costs than compared to the control group (White et al. 2000).

With regard to qualitative measures, Ohio’s PASSPORT program successfully targeted the population in need of their services, based on specific eligibility criteria, such as financial need and functional limitations. Further, it was found that case management function was a linchpin of the program’s success, and that the assessment process adequately covers consumer needs and contributes to an appropriate service plan. Overall, an independent evaluation of the program found that PASSPORT “is a cost-neutral, effectively targeted, quality-oriented, thoroughly monitored, consumer-responsive care program” (Ciferri 2007).

On the cost savings front, the home and community-based care options proved less expensive than nursing home care. In 2006, the average Medicaid cost per person for nursing home care was $55,751 (including health care expenditures, medication and long-term supportive services); the average cost for a PASSPORT client was $23,703 (including traditional Medicaid health care expenditures, medication and PASSPORT long-term supportive services). Even when considering all public sources of funding (SSI, food stamps, housing, HEAP), the costs of caring for a PASSPORT recipient were less than for a similar beneficiary living in a nursing home (Ciferri 2007).

The stated goals of the Wisconsin Family Care Program (WFC) are to provide “Choice, Access, Quality and Cost-Effectiveness.” Qualitatively, members demonstrated a high level of satisfaction (92% of the survey respondents) with the care management and overall quality of services provided. WFC was also successful in supporting the desire of members to gain access to their residential choice (almost always home and community), with 82.4% of Family Care Members residing in their preferred living arrangement (Wisconsin Department of Health and Family Services [WDHFS] 2009).

In an independent assessment, researchers found that the overall Medicaid costs were lower for Family Care beneficiaries in the four non-Milwaukee county managed care organizations and within each of the target groups. Costs were also lower than in a comparison group when looking at frail elders in Milwaukee County. Researchers determined that the savings were related to two factors: controlling service costs and indirectly, favorably affecting beneficiaries’ health and abilities to function (thus needing fewer services). For all but one of the counties studied, the average monthly long-term care costs were significantly less than those in comparison groups. The only group where this was not the case was in the developmentally disabled population in Milwaukee County. Although home health costs (home health care, personal care, and supportive home care) increased over the study period, the increase was significantly slower in the Family Care beneficiaries group than in the control group, except in Milwaukee County. Inpatient hospital costs significantly decreased over the study period for the Family Care beneficiaries in all counties, even though the costs for this group averaged more than the control group at baseline (WDHFS 2007).

The Community Care of North Carolina (CCNC) currently has an evaluation underway of its Chronic Care Program, which is specifically targeted to older adults. This evaluation will incorporate outcomes that are related to global costs/utilization, outreach and enrollment, and case management, as well as outcomes related to specific chronic diseases. In the meantime an evaluation of CCNC’s Asthma Disease Management Initiative has demonstrated cost effectiveness related to ED use and hospital admission rates, compared with children in a control group (CCNC 2003). Researchers estimated a $3.5 million dollar savings resulting from the asthma program and $2.1 million savings with the diabetes management program. In estimating the impact of the whole CCNC program, Mercer (2009) found when comparing what the access model would have cost in SFY04, without any concerted efforts to control costs, the program saved approximately $60 million in SFY03, $124 million in SFY04, and $231 million in SFY05 and SFY06.

The Vermont Blueprint Integrated Pilot, first implemented in 2008, has an evidence-based evaluation in progress, which will be based on variety of data sources, including electronic medical records for individual patient care, medical claims from insurers and Medicaid, and public health surveys, among other sources. Uses of the data will cover a broad range of outcomes focused on clinical results, resource utilization, health care expenditures, and overall quality improvement. From the program’s inception, health information technologies have been incorporated into the planning both as a tool for quality improvement and as a means to measure outcomes. Thus far, the Vermont pilot programs have been well received by patients, physicians and staff, while the formal evaluation is being conducted. Vermont legislation mandates statewide expansion by July of 2011, and the estimated cost savings is $115 million per year in five years from implementation (Connecticut Health Policy Project 2010).

In one of the longest-running programs, VA Geriatric Evaluation and Management, GEM patients were less likely to use the ED as compared to those in a control group (Engelhardt et al. 2006; Boult et al. 1994, 2001). In a 24-month period, results showed that GEM patients incurred significantly lower costs (primarily attributable to fewer hospital days of care) than patients receiving routine care (Engelhardt et al. 2006).

Turning to Hospice, one study found that when a social worker was involved in hospice care, there were a lower number of hospitalizations per patient; lower home health aide costs, lower nursing costs, lower labor costs, and lower average pain management cost per patient (Reese & Raymer 2004). In another evaluation performed on the Advanced Illness Coordinated Care Program (AICC), a study commissioned by the National Hospice and Palliative Care Organization, found that patients with one of sixteen diagnoses receiving hospice services cost Medicare less than those with the same diagnosis not receiving hospice (Ascribe 2004).

Common elements of successful programs

Despite the varied degree of integration and system of origin, the care coordination approaches described in this paper have several common elements: targeted intervention, in-person assessment/meetings, comprehensive initial assessment, a team approach to care coordination, family involvement, and focus on community-based services (Table 4.3).

Table 4.3 Common Components of Models Analyzed

Targeted intervention

Most of the programs target a specific population or segment within it. Hospice is specifically designed for patients in the final stage of a terminal illness. GEM is offered to a select group of medically complex, frail veterans, although some programs around the country have broadened their scope in order to promote health and prevent disease. While Community Care of North Carolina serves almost the entire Medicaid population of North Carolina, the case management intervention is based on an assessment that assigns a status/level of intensity (Heavy, Medium, Light, Very Light, and Deferred). Vermont’s BIP also serves a broad swath of the population, although the program is targeted toward individuals with chronic care needs and utilizes Wagner’s Chronic Care Model as a building block. The Wisconsin Family Care Benefit requires that the participant have at least one chronic condition that is expected to last more than 90 days. The other models involve specific eligibility criteria, in particular requiring that an individual need a nursing home level of care in order to qualify for community-based services and care coordination that can prevent institutionalization.

Face-to-face contact and comprehensive assessment

All of these models utilize an assessment, either performed by a designated care coordinator or on the team level. The initial assessment is conducted in person. Most care coordination models also involve telephone contact, including periodic follow-up calls to monitor the care plan and ensure satisfaction, among other tasks. As noted above, the initial assessment is often conducted in the person’s place of residence (own home or a residential care facility). Providing in-home assessment allows the care coordinator to directly observe the person’s environment, identifying and possibly alleviating barriers the individual may have in receiving services. The PACE assessment is conducted at the adult day center (the epicenter of the model), Community Care of North Carolina assessments are done at the medical home, and GEM’s assessments are conducted in outpatient clinics.

Team approach and family involvement

Every program uses an interdisciplinary team approach to coordinating care and/or creating a care plan with the person. A team approach to assessment and care planning recognizes the specialized knowledge and assessment skills provided by different disciplines and encourages multidisciplinary solutions. The Arizona Long Term Care System and the Minnesota Senior Health Options employ a small team (case manager and primary care physician), with the case manager serving as a liaison to the physician. Other programs also use one member of the team as the primary coordinator of services – in GEM, primarily a social worker, in Hospice, a nurse. This central source of coordination decreases the likelihood of miscommunication, gaps in service, and other forms of fragmentation. By contrast, PACE and the Wisconsin Partnership Plan use the entire interdisciplinary team to fulfill the care coordination function.

Direct involvement of the primary care physician, although it was not a universal practice, has positive impacts on rate and length of hospitalizations, cost, and nursing home admissions based upon outcome data from Community Care of North Carolina, PACE, the Wisconsin Partnership Program, GEM, and Hospice. In each instance, the primary care physician does not serve as the care coordinator; instead, this role was most often filled by a nurse or social worker. At least two models involve the family (where appropriate) as a member of the team in the decision making, assessment and coordination of care (Minnesota Senior Health Options and PACE). Some programs also provide supportive services for family caregivers, including support groups and training (PACE, Hospice, and GEM).

Focus on home and community-based services to avoid institutionalization

Focus on home and community-based services that help delay or avoid institutionalization is a key design element in all of the programs. The Arizona Long Term Care System provides incentives for providers to prevent premature institutionalization. Similarly, both the Wisconsin Partnership Program and the Wisconsin Family Care Program stress the importance of early education, early assessment, and early and correct service provision to prevent hospitalization, delay institutionalization, and avoid unnecessary services. PASSPORT is designed specifically for frail homebound seniors with the goal of keeping the senior living independently in the community, rather than place the individual in a nursing home. GEM targets individuals with higher rates of service use such as ED visits, in order to decrease these over time. Hospice strives to keep the person as stable as possible in the home and often uses as a desirable outcome measure, death at home rather than in the hospital.

Table 4.4 Care Coordination Team

Notes: APN = advanced practice nurse; RN = registered nurse; SW = social worker

Care coordinator characteristics

As summarized in Table 4.3, the integrated care coordination programs use registered nurses and/or social workers as care coordinators, working as part of an interdisciplinary team. In a few cases, the whole team is responsible for performing care coordination. Most programs require bachelor’s level (or higher) education as a minimum requirement for care coordinators. The care coordinators are employed by a variety of different entities, ranging from the Veteran’s Administration for GEM to community agencies that are part of the networks in Community Care of North Carolina to the Office of Vermont Health Access in the Vermont Blueprint Integrated Pilot.

Other programs of interest

In order to establish an evidence base for care coordination models that integrate the health and long-term care needs of older adults, we have made deliberate choices to include programs that offer representative examples of integrated care and have undergone evaluations that describe key indicators with respect to hospital utilization, use of institutional care and cost outcomes. These are certainly not the only measures that can be used to evaluate the success of a given program, although they do provide significant information about positive results and are closely allied with policy concerns about preventable hospital admissions, rebalancing of long-term care to home and community-based alternatives, and the potentially huge increase in the cost of care brought on by an aging population.

In selecting ten programs that meet the criteria previously outlined, it is important to note that there are a number of other programs that seek to integrate health and long-term care, and are doing so with positive results; in some cases these programs are using different evaluation techniques that are particularly relevant to the health and long-term care issues of older adults. As one example, Missouri has made significant strides towards integrating medical care with community-based social support for older adults through its Aging in Place Program (AIP) which has improved on a state-funded HCBS program called Missouri Care Options, a program designed to provide an alternative to institutionalization. The AIP model uses a nurse care coordinator who performs a comprehensive assessment and creates a care plan for the participant that coordinates the interventions of physician, nurse, and other health providers with the Missouri Care Options services of personal care and homemaking. The most current evaluation of AIP utilizes a quasi experimental design to compare outcomes between older adults who reside in nursing homes with those who received services in the AIP program. The clinical outcomes measured were activities of daily living, cognitive function, depression, incontinence and pressure ulcers. By integrating home care with medical services, the AIP program generated positive results for this set of clinical outcomes, which are of particular relevance to the health of frail and vulnerable older adults (Marek et al. 2005).

Another model of integrated care that has generated interest among health and aging professionals is the Senior Health and Clinic (SHC) model, which is used to establish primary care clinics with a geriatric focus on serving older adults with multiple chronic conditions and social support needs. The first Senior Health and Wellness Clinic, a geriatric outpatient clinic in Eugene, Oregon, has been in continuous operation since 2000. Many elements of the SHC model are based in the principles and design of the Chronic Care Model (CCM), which stipulates a true interdisciplinary care approach. The interdisciplinary team is typically composed of a geriatrician, nurse practitioner, social worker, nurse, and dietitian. Ad hoc members may include a chaplain, physical therapist, and home health nurse. Senior Health and Wellness Center patients had lower average Medicare charges, as a result of same or reduced utilization of outpatient, hospital, and ED use (John A. Hartford Foundation 2007). The SHC model has also led to improvements in health-related quality of life (HQRL) outcomes (Stock et al. 2008). A hindrance to the expansion of the model, although it is designed for export to any state or region, has been its reliance on Medicare payments – it has no waiver authority or state-supported alternative reimbursement structure – which means that a split billing method has to be employed in order to support the delivery of coordinated, integrated care (Silow-Carroll et al. 2006).

Conclusions and policy implication

The historic passage of H.R.3590 – the Patient Protection and Affordable Care Act – is good news for those who believe that care coordination is an essential part of improving the flawed delivery of health and long-term care services to older adults. This act contains provisions that specifically target the development and implementation of models of care that have care coordination principles built in to their design. The continuing dissemination of these models is supported by a number of measures. The establishment of the Centers for Medicare and Medicaid Innovation Center (CMI) has a directive to test innovative payment and service delivery models that improve or maintain quality while controlling costs. Preference will be given to models that promote care coordination by transitioning health care providers away from fee-for-service based reimbursement; models that utilize geriatric assessment and comprehensive care plans to coordinate the care (including through interdisciplinary teams) for individuals with multiple chronic conditions; and models that establish community-based health teams to support small-practice medical homes by assisting the primary care practitioner in chronic care management. The bill also provides for the establishment of a CMS Coordinated Health Care Office that will specifically target dual-eligibles. Some of the goals of this office will be to more effectively integrate benefits under the Medicare and Medicaid programs, improve the quality of health care and long-term services, and support state efforts to align medical and long-term care services for dual-eligibles.

A focus on care coordination has over 30 years of history to build on, with care coordination strategies as a centerpiece of programs serving Medicaid beneficiaries, beneficiaries dually eligible for Medicare and Medicaid, selected veterans in programs administered by the Veterans Administration, and other targeted populations of older adults. As described in this chapter, these types of programs have been essential to the development of models that bridge the medical care and long-term social support service domains and are supported by the evidence for their efficacy. At the same time, many of the problems that have prevented system reform are still entrenched and will require constant vigilance and effort to overcome. One problem is the ongoing medical bias of care coordination, as is often the case with the patient-centered medical home, or the relative weakness of socially oriented models in achieving a meaningful integration of medical care. As mentioned in the introduction, the silos of care that separate medical, long-term care and social supports continue to present roadblocks to change. Dissemination of programs to reach meaningful numbers of the older adult population is also an ongoing problem, and will require fundamental changes to reimbursement mechanisms in order to support models of care that do not fit in to the traditional fee for service system that rewards acute, episodic care. The programs that are the most effective in achieving a full integration of care also tend to be the most time consuming and complex to implement – so more streamlined, flexible ways of achieving full integration will need to be explored.

The findings of this study regarding the components of care coordination that lead to positive outcomes are an important lesson of our analysis: the fundamental role of targeting; the importance of face-to-face contact between the care coordinator and the recipient of care as well as between the care coordinator and the primary care physician; and the central role of assessment and a comprehensive care plan in guiding care. This analysis also highlights the importance of community-based services in the broad array of cost-related outcomes achieved by these diverse programs. Overall, the programs evaluated here showed reductions to hospitalization, ED visits, nursing home admissions, and the total cost of care. These more wide-ranging outcomes were possible because ongoing linkages to social service and other community-based resources help to improve or stabilize the medical conditions of many patients. Care coordination models that address the social support dimensions and link them to medical care are generally designed so that broad-based interdisciplinary teams have the capability to address both the treatment of chronic disease and associated functional limitations that define long-term care needs.

Evidence in support of care coordination and integrated care will have to continue to be developed in order to persuade policy makers and health and social service professionals that integrated models of coordinated care in fact achieve the desired combination of improvements to quality and either cost neutrality or cost savings. It will be important for newer models of care to incorporate a well-thought out evaluation plan from the beginning of their development, and for evaluation experts to agree upon the most relevant and important outcomes. If integrated models of care are able to consistently produce positive results across different settings and populations, this will greatly facilitate the broad dissemination of models of care that effectively target the needs of older adults.

References

American Association of Retired Persons Public Policy Institute. (2008) Valuing the Invaluable, The Economic Value of Family Caregiving, 2008 Update. Available at: http://assets.aarp.org/rgcenter/il/i13_caregiving.pdf.

Agency for Healthcare Research and Quality. (2009) Innovations Exchange. Agency for Healthcare Research and Quality, Rockville, MD. Available at: http://www.innovations.ahrq.gov/ evidencerating.aspx.

APS Healthcare Inc. (2005) Family Care Options for Long-Term Care: Family Care Independent Assessment: An Evaluation of Access, Quality and Cost Effectiveness for CY 2003–2004. Available at: http://dhs.wisconsin.gov/ltcare/ResearchReports/IA.pdf.

Ascribe Newswire. (2004) Hospice Costs Medicare Less and Patients Often Live Longer, Research Shows. Ascribe Newswire: Medicine, 9/22/2004, p 2–4. Available at: http://search.ebscohost.com/login.aspx?direct=true&db=heh&AN=14533297&site=ehost-live.

Berenson, R., & Howell, J. (2009) Structuring, Financing and Paying for Effective Chronic Care Coordination. A Report Commissioned by the National Coalition on Care Coordination (N3C). Available at: http://www.urban.org/UploadedPDF/1001316_chronic_care.pdf.

Boult, C., Boult, L., Murphy, C., et al. (1994) A controlled trial of outpatient geriatric evaluation and management. Journal of the American Geriatrics Society, 42, 465–470.

Boult, C., Boult, L., Morishita, L., et al. (2001) Randomized clinical trial of outpatient geriatric evaluation and management. Journal of the American Geriatrics Society, 49, 351–359.

Brown, R. (2009) The Promise of Care Coordination: Models That Decrease Hospitalizations and Improve Outcomes for Medicare Beneficiaries with Chronic Illness. Available at: http://www.socialworkleadership.org/nsw/Brown_Full_Report.pdf.

Centers for Medicare and Medicaid Services. (2007) Long-Term Care Capitation Modes: A Description of Available Program Authorities and Several Program Examples. Available at: http://www.cms.hhs.gov/IntegratedCareInt/Downloads/LTC_Capitation.pdf.

Centers for Disease Control and Prevention. (2003) Public health and aging: trends in aging—United States and worldwide. Morbidity and Mortality Weekly Report, 52, 101–106.

Ciferri, W., Kunkel, S., McGrew, K., et al. (2007) Program Evaluation of PASSPORT: Ohio’s Home and Community Based Medicaid Waiver. Available at: http://sc.lib.muohio.edu/handle/2374.MIA/62.

Connecticut Health Policy Project (2010) Development of a Vermont Pilot Community Health System to Achieve the Triple Aims. Available at: http://www.cthealthpolicy.org/webinars/20100223_jhester_webinar.pdf.

CCNC: Community Care of North Carolina. Community Care of North Carolina, Raleigh, NC. Available at: http://www.communitycarenc.com.

Community Care of North Carolina. (2003) Progress Report: Asthma Initiative. Community Care of North Carolina, Raleigh, NC. Available at: http://www.communitycarenc.com.

Engelhardt, J.B., McClive Reed, K.P., Toseland, R.W., et al. (2006) Effects of a program for coordinated care of advanced illness on patients, surrogates, and healthcare costs: a randomized trial. The American Journal of Managed Care, 12, 93–100.

The John A. Hartford Foundation. (2007) Geriatric Interdisciplinary Teams in Practice: Senior Health and Wellness Clinic Model. Available at: http://www.jhartfound.org/ar2007html/pdf/Hart07_SENIOR_HEALTH_AND_WELLNESS_CLINIC_MODEL.pdf

Kane, R., Homyak, P., Bershadsky, B., et al. (2004) Patterns of utilization for the Minnesota Senior Health Options Program. Journal of the Geriatrics Society 52, 2039–2044.

Landkamer, S., & Landsness, R. (2005) Integrated Health and Long-Term Care Services: The Wisconsin Partnership Program. Available at: http://ici.umn.edu/products/impact/181/.

MacAdam, M. (2008) Frameworks of Integrated Care for the Elderly: A Systematic Review. CPRN Research Report. Available at: http://www.cprn.org/documents/49813_EN.pdf.

Malone, J., Morishita, L., Paone, D., et al. (2004) Minnesota Senior Health Options (MSHO) Care Coordination Study. Minnesota Department of Human Services, Minneapolis, MN. Available at: http://www.dhs.state.mn.us/main/groups/healthcare/documents/pub/dhs_id_028242.pdf.

Marek, K.D., Popejoy, L., Petroski, G., et al. (2005) Clinical outcomes of aging in place. Nursing Research, 54, 202–211.

McCall, N. (1997) Lessons from Arizona’s Medicaid managed care program. Health Affairs, 16 (4), 194–199.

Mercer. (2009) CCNC/ACCESS Cost-Savings – State Fiscal Year 2007 Analysis. Mercer Government Human Services Consulting, Atlanta, GA. Available at: http://www.communitycarenc.com.

Mollica, R., & Gillespie, J. (2003) Care Coordination for People with Chronic Conditions. National Academy for State Health Policy, Portland, ME.

Reese, D. & Raymer, M. (2004) Relationships between social work involvement and hospice outcomes: results of the National Hospice Social Work Survey. Social Work, 49, 415–422.

Silow-Carroll, S., Alteras, T., & Stepnick, L. (2006) Patient-Centered Care for Underserved Populations: Definition and Best Practices. Economic and Social Research Institute, Washington, DC.

Steiner, B.D., Denham, A.C., Ashkin, E., et al. (2008) Community Care of North Carolina: improving care through community health networks. Annals of Family Medicine, 361–367.

Stock, R., Mahoney, E. R., Reece, D., et al. (2008) Developing a senior healthcare practice using the chronic care model: effect on physical function and health-related quality of life. Journal of the American Geriatrics Society, 56, 1342–1348.

White, A., Abel, Y., & Kidder, D. (2000) Evaluation of the Program of All-Inclusive Care for the Elderly Demonstration: Comparison of the PACE Capitation Rates to Projected Costs in the First Year of Enrollment. Abt Associates, Cambridge, MA (HCFA Contract No. 500-01-0027).

Wilhide, S. & Henderson, T. (2006) Community Care of North Carolina: A Provider-Led Strategy for Delivering Cost-Effective Primary Care to Medicaid Beneficiaries (Executive Summary). American Academy of Family Physicians. Available at: http://www.aafp.org/online/etc/medialib/aafp_org/documents/policy/state/medicaid/ncexecsumm.Par.0001.File.tmp/ncexecsummary.pdf.

Wisconsin Department of Health and Family Services. (2007) Wisconsin Partnership Program Census Graphs. Available at: http://dhfs.wisconsin.gov/WIpartnership/census.htm.

Wisconsin Department of Health and Family Services, Division of Long Term Care. (2009) Long-Term Care in Motion: 2008 Annual Report of Wisconsin’s Managed Long-Term Care Programs. Available at: http://dhs.wisconsin.gov/ltcare/ResearchReports/PDF/2008annualreport.pdf.

Stay updated, free articles. Join our Telegram channel

Join