Chapter 5
Older Adults and Mental Health

Chapter Overview

Overview of Mental Disorders in Older Adults

Depression in Older Adults

Alzheimer’s Disease

Other Mental Disorders in Older Adults

Service Delivery

Other Services and Supports



Service Delivery

Overview of Services

New perspectives are evolving on the nature of mental health services for older adults and the settings in which they are delivered. Far greater emphasis is being placed on community-based care, which entails care provided in homes, in outpatient settings, and through community organizations. The emphasis on community-based care has been triggered by a convergence of demographic, consumer, and public policy imperatives. In terms of demographics, approximately 95 percent of older persons at a given point in time live in the community rather than in institutions, such as nursing homes (U.S. Department of Health and Human Services, Administration on Aging, and American Association of Retired Persons [U.S.DHHS, AoA & AARP], 1995). Of those living in the community, approximately 30 percent, mostly women, live alone (U.S. DHHS, AoA & AARP, 1995). Most older persons prefer to remain in the community and to maintain their independence. Yet living alone makes them even more reliant on community-based services if they have a mental disorder.

Service delivery also is being shaped by public policy and the emergence of managed care. The escalating costs of institutional care, combined with the recognition of past abuses, stimulated policies to limit nursing home admissions and to shift treatment to the community (Maddox et al., 1996). Mental disorders are leading risk factors for institutionalization (Katz & Parmelee, 1997). Therefore, to keep older people in the community, where they prefer to be, more energies are being marshaled to promote mental health and to prevent or treat mental disorders in the community. In other words, treating mental disorders is seen as a means to stave off costly institutionalization—resulting either from a mental disorder or a comorbid somatic disorder. An untreated mental disorder, for example, can turn a minor medical problem into a life-threatening and costly condition. Problems with forgetting to take medication (e.g., with dementia), developing delusions about medication (e.g., with schizophrenia), or lowering motivation to refill prescriptions (e.g., with depression) can increase the likelihood of having more severe illnesses that demand more intensive and expensive institutional care. Therefore, promotion of mental health and treatment of mental disorders are crucial elements of service delivery.

The delivery of community-based mental health services for older adults faces an enormous challenge. Services for older adults are insufficient and fragmented, often divided between systems of health, mental health, and social services (Gatz & Smyer, 1992; Cohen & Cairl, 1996). Under these three systems, services include medical and psychosocial care, rehabilitation, recreation, housing, education, and other supports. Yet although every community has an Administration on Aging to assist with services for older adults generally, there is no administrative body responsible for integrating the daunting array of services needed specifically for individuals with severe mental illnesses. Similar problems are encountered with coordinating services for children, as discussed in Chapter 3. Local mental health authorities and systems of care have been effective in coordinating care for some groups of adults, but no special administrative mental health entities exist for older adults. The fragmentation of service systems for older people in the United States stands in contrast to the United Kingdom and Ireland, where governmental authorities coordinate their care (Reifler, 1997). Older adults eventually may benefit from the local mental health authorities developing in the United States, but thus far these authorities have been focused on services for other adults. Because of ethnic diversity in the United States, systems of care must also deal with the special needs of older Americans who have limited English proficiency and different cultural backgrounds.

The following section describes the nature and settings in which older people receive mental health services. It concentrates on primary care, adult day centers and other community care settings, and nursing homes. A recurrent theme across these settings is the failure to address mental health needs of older people. Selected issues in financing of services for older adults are discussed briefly at the end of this section, but most of the issues related to financing policy (e.g., Medicare, Medicaid) and managed care are discussed in Chapter 6.

Service Settings and the New Landscape for Aging

Demographic, consumer, and public policy imperatives have propelled tremendous growth in the diversity of settings in which older persons simultaneously reside and receive care (Table 5-2). Care is no longer the strict province of home or nursing home. The diversity of home settings in suburban and urban communities extends from naturally occurring retirement communities to continuing care retirement communities to newer types of alternative living arrangements. These settings include congregate or senior housing, senior hotels, foster care, group homes, day centers (where people reside during the day), and others. The diversity of institutional settings includes nursing homes, general hospitals (with and without psychiatric units), psychiatric hospitals, and state mental hospitals, among others. In fact, the range of settings, and the nature of the services provided within each, has blurred the distinction between home and nursing home (Kane, 1995).

Across the range of settings, the duration of care can be short term or long term, depending on patients’ needs. The phrase, “long-term care,” has come to refer to a range of services for people with chronic or degenerative illness or disabilities who require support over a prolonged period of time. In the past, long-term care was synonymous with nursing home care or other forms of institutional care, but the term has come to apply to a full complement of institutional or community-based settings.

Table 5-2. Settings for mental health services for older adults*



Group homes

Retirement communities

Primary care and general medical sector

Outpatient therapy

Community mental health centers


Nursing homes

General hospitals with psychiatric units

General hospitals without psychiatric units

State mental hospitals

Veterans Affairs hospitals

*Two other settings (not included in this table) are board and care homes and assisted living facilities. These are residential facilities that serve as a bridge between community and institutional settings and have elements of each.

Within the continuum of services, one new perspective—conceived as the landscape for aging— strives to tailor the environment to the needs of the person through a combined focus on health and residential requirements (Cohen, 1994). Whether at home, in a retirement community, or in a nursing home, this health and home perspective is deemed to be crucial to achieving high quality of life for older adults. Over the past 30 years, improvements in the health side of this perspective have occurred, but the home part has lagged. The challenge is to stimulate an interdisciplinary collaboration between systems of care and consumers.

One important area for an interdisciplinary approach is the extent to which a given setting fosters independent functioning versus dependent functioning, an issue influencing mental health and quality of life. Though certainly not a goal, some settings inadvertently foster dependency rather than independence. Nursing homes and hospitals, for example, are understandably more focused on what individuals cannot do, as opposed to what they can do. Yet their major focus on incapacity (the nursing and health focus) runs the risk of overshadowing function and independence (the home and humanities focus). In other settings, the balance between dependence and independence shifts in the other direction, with the risk of nursing and health needs being inadequately addressed. In recent years, the emphasis has been on “aging in place,” either at home or in the community, rather than in alternate settings.

The landscape for aging is a construct within which to examine the depth and breadth of human experience in later life (Cohen, 1998b). A health and humanities focus across this landscape offers a design for dealing with mental health problems as well as with health promotion to harness human potential. The landscape for aging, with its health and humanities orientation, is a construct designed to stir new thinking in research, practice, and policy. It also defines a clear need for new mental health services’ development and delivery, training, research, and policies to address the range of sites, each with its own unique characteristics and growing populations. The service systems, however, have yet to embrace a broader view.

Primary Care
Primary care6 represents a pivotal setting for the identification and treatment of mental disorders in older people. Many older people prefer to receive mental health treatment in primary care (Unutzer et al., 1997a), a preference bolstered by public financing policies that encourage their increasing reliance on primary, rather than specialty, mental health care (Mechanic, 1998). Primary care offers the potential advantages of proximity, affordability, convenience, and coordination of care for mental and somatic disorders, given that comorbidity is typical.

The potential advantages of primary care, however, have yet to be realized. Diagnosis and treatment of older people’s mental disorders in the primary care setting are inadequate. The efficacious treatments described in the depression section of this chapter are not being practiced, particularly not in primary care and other general medical settings. As documented earlier, a significant percentage of older patients with depression are underdiagnosed and undertreated. The concern about inadequate treatment of late-life depression in primary care is magnified by growing enrollment in managed care.

Primary care is generally not well equipped to treat chronic mental disorders such as depression or dementia. It has limited capacity to identify patients with common mental disorders and to provide the proactive followup that is required to retain patients in treatment. To ensure better treatment of late-life depression in primary care, there is heightening awareness of the need for new models for mental health service delivery (Unutzer et al., 1997a). New models of service delivery in primary care include mental health teams, consultation-liaison models,7 and integration of mental health professionals into primary care (Katon & Gonzales, 1994; Schulberg et al., 1995; Katon et al., 1996, 1997; Stolee et al., 1996; Gask et al., 1997). For example, the intervention developed by Katon and colleagues introduced a structured depression treatment program into the primary care setting. The program included behavioral treatment to inculcate more adaptive coping strategies and counseling to enhance compliance with antidepressant medications. Patients were randomized in a controlled trial comparing this structured depression program with usual care by primary care physicians. The investigators found patients participating in the program to have displayed better medication adherence, better satisfaction with care, and a greater decrease in severity of major depression (Katon et al., 1996).

Models that integrate mental health treatment into primary care, while thus far designed largely for depression, also may have utility for other mental disorders seen in primary care. Nevertheless, primary care is not appropriate for all patients with mental disorders. Primary care providers can be guided by a set of recommendations for appropriate referrals to specialty mental health care (American Association for Geriatric Psychiatry, 1997).

Adult Day Centers and Other Community Care Settings
Over the past few decades, adult day centers have developed as an important service delivery approach to providing community-based long-term care. Adult day centers, although heterogeneous in orientation, provide a range of services (usually during standard “9 to 5” business hours), including assessment, social, and recreation services, for adults with chronic and serious disabilities. They represent a form of respite care designed to give caregivers a break from the responsibility of providing care and to enable them to pursue employment. Over the past 30 years, adult day centers have grown in number from fewer than 100 to over 4,000, under the sponsorship of community organizations or residential facilities. A large national demonstration program on adult day centers showed that they can care for a wide spectrum of patients with Alzheimer’s disease and related dementias and can achieve financial viability (Reifler et al., 1997; Reifler et al., in press). There also is evidence that adult day centers are cost-effective in terms of delaying institutionalization, and participants show improvement in some measures of functioning and mood (Wimo et al., 1993, 1994).

There are several approaches to delivering services in adult day centers. There is no research evidence that any one model of service delivery is superior to another. For example, a social model has been developed by Little Havana Activities & Nutrition Centers of Dade County (Florida). The Little Havana “Senior Center” provides mental health, health, social, nutritional, transportation, and recreational services, emphasizing both remedial and preventive services. The center focuses on the predominantly Cuban population of South Florida. Yet much more research is needed to demonstrate the relative effectiveness of different models of adult day services (Reifler et al., 1997).

Beyond adult day centers, other innovative models of community-based long-term care strive to incorporate mental health services. Few have been evaluated and none implemented on a wide scale. These models include the social/health maintenance organization (S/HMO) (Greenberg et al., 1988), On Lok Senior Services Program, and life care communities or continuing care retirement communities (Robinson, 1990b). These new features of the landscape of aging show promise, but there is insufficient evidence of cost-effectiveness and generalizability of these models, particularly the mental health component. Perhaps the lack of a research base and limited market account for the slow pace of their proliferation in the United States.

Nursing Homes
Most older adults live in the community and only a minority of them live in nursing homes; of the latter, about two-thirds have some kind of mental disorder (Burns, 1991). The majority have some type of dementia, while others have disabling depression or schizophrenia (Burns, 1991). Despite the high prevalence of people with mental disorders in nursing homes, these settings generally are ill equipped to meet their needs (Lombardo, 1994).

Deinstitutionalization of state mental hospitals beginning in the 1960s encouraged the expanded use of nursing homes for older adults with mental disorders. This trend was enhanced by Medicaid incentives to use nursing homes instead of mental hospitals. But the shift to nursing homes was not accompanied by alterations in care. In 1986, the Institute of Medicine issued a landmark report documenting inappropriate and inadequate care in nursing homes, including the excessive use of physical and chemical restraints (IOM, 1986). This subsequent visibility of problems prompted the passage in 1987 of the Nursing Home Reform Act (also known as the Omnibus Budget Reconciliation Act of 1987). This legislation restricted the inappropriate use of restraints and required preadmission screening for all persons suspected of having serious mental illness. The purpose of the screening was to exclude from nursing homes people with mental disorders who needed either more appropriate acute treatment in hospitals or long-term treatment in community-based settings. Preadmission screening also was designed to improve the quality of psychosocial assessments and care for nursing home residents with mental disorders. Nursing home placement is appropriate for patients with mental disorders if the disorders have produced such significant dysfunction that patients are unable to perform activities of daily living.

To meet the legislation’s requirements, nursing homes must have the capacity to deliver mental health care. Such capacity depends on trained mental health professionals to deliver appropriate care and treatment. Unfortunately, prior to and even after passage of the Omnibus Budget Reconciliation Act of 1987, Medicaid policies discouraged nursing homes from providing specialized mental health services, and Medicaid reimbursements for nursing home patients have been too low to provide a strong incentive for participation by highly trained mental health providers (Taube et al., 1990). The emphasis on community-based care, combined with inadequate nursing home reimbursement policies, has limited the development of innovative mental health services in nursing homes. Major barriers persist in the delivery of appropriate care to mentally ill residents of nursing homes.

Services for Persons With Severe and Persistent Mental Disorders
Older adults with severe and persistent mental disorders (SPMD) are the most frequent users of long-term care either in community or institutional settings. SPMD in older adults includes lifelong and late-onset schizophrenia, delusional disorder, bipolar disorder, and recurrent major depression. It also includes Alzheimer’s disease and other dementias (and related behavioral symptoms, including psychosis), severe treatment-refractory depression, or severe behavioral problems requiring intensive and prolonged psychiatric intervention. Although these groups of disorders have different courses of illness and outcomes, they have many overlapping clinical features, share the common need for mental health long-term care services, and are frequently treated together in long-term care settings (Burns, 1991; Gottesman et al., 1991; American Psychiatric Association, 1993). It is estimated that 0.8 percent of persons older than 55 years in the United States have SPMD (Kessler et al., 1996).

As a result of the dramatic downsizing and closure of state hospitals in past decades, 89 percent of institutionalized older persons with SPMD now live in nursing homes (Burns, 1991). However, institutions are expected to play a substantially smaller role than community-based settings in future systems of mental health long-term care (Bartels et al., in press). First, the majority of older adults with SPMD presently live in the community (Meeks & Murrell, 1997; Meeks et al., 1997) and prefer to remain there. Second, experience with the Preadmission Screening and Resident Review mandated by the Omnibus Budget Reconciliation Act of 1987 has been mixed. It may have slowed inappropriate admissions to nursing homes, restricted inappropriate use of restraints, and reduced overuse of psychotropic medications, but it did not otherwise improve the quality of mental health services (Lombardo, 1994). Furthermore, states’ opposition to what they perceived to be Federal government interference in local health care policy and a general trend toward deregulation subsequently curtailed Federal nursing home reform. Finally, the growing costs of nursing home care are stimulating dramatic reforms in reimbursement and policy, including state mandates to limit Medicaid expenditures by decreasing nursing home beds and Federal reform by Medicare to implement prospective payment for nursing home services (Bartels & Levine, 1998). To accommodate the mounting number of individuals who have disorders requiring chronic care, future projections suggest the greatest growth in services will be in home and community-based settings (Institute for Health and Aging, 1996), increasingly financed through capitated and managed care arrangements.

Older adults with SPMD are high users of services (Cuffel et al., 1996; Semke & Jensen, 1997) and require mental health long-term care that is comprehensive, integrated, and multidisciplinary (Moak, 1996; Small et al., 1997; Bartels & Colenda, 1998). The mental health care needs of this population include specialized geropsychiatric services (Moak, 1996); integrated medical care (Moak & Fisher, 1991; Small et al., 1997); dementia care (Small et al., 1997; Bartels & Colenda, 1998); home and community-based long-term care; and residential and family support services, intensive case management, and psychosocial rehabilitation services (Aiken, 1990; Robinson, 1990a; Schaftt & Randolph, 1994; Lipsman, 1996). With adequate supports, older persons with SPMD can be maintained in the community, sometimes at lower cost, and with equal or improved quality of life in comparison with institutions (Bernstein & Hensley, 1988; Mosher-Ashley, 1989; Leff, 1993; Trieman et al., 1996).

However, current mental health policies have left many older persons with SPMD with decreased access to mental health care in both community and institutional settings (Knight et al., 1998). Community-based mental health services for older people are largely provided through the general medical sector, partly due to poor responsiveness to the needs of older people by community mental health organizations (Light et al., 1986). Yet reliance on the general medical sector also has not met their needs because of its focus on acute care (George, 1992). In addition, most home health agencies provide only limited short-term mental health care. The long-term care programs that exist primarily aid older adults with chronic physical disabilities or cognitive impairment but fail to address impairments in mood and behavior (Robinson, 1990a). An additional barrier is that the majority of community-residing older adults do not seek mental health services, except for medication (Meeks & Murrell, 1997), despite continued need (Meeks et al., 1997). Those without family support generally live in nursing homes, assisted living facilities, and board and care homes. These three are forms of residential care that offer some combination of housing, supportive services, and, in some cases, medical care. In short, more resources must be devoted to programs that integrate mental health rehabilitative services into long-term care in both community and institutional settings.

Financing Services for Older Adults

Financing policies furnish incentives that favor utilization of some services over others (e.g., nursing homes rather than state mental hospitals) or preclude the provision of needed services (e.g., mental health services in nursing homes). Details on financing and organizing mental health services, with a special focus on access, are presented in Chapter 6. Selected issues germane to older adults are addressed here.

Historically, Federal financing policy has imposed special limits on reimbursement for mental health services. Medicaid precluded payment for care in so-called “institutions for mental diseases,” Medicaid’s term for mental hospitals and the small percentage of nursing homes with specialized mental health services. This Medicaid policy provided a disincentive for the majority of nursing homes to specialize in delivering mental health services for fear of losing Medicaid payments (Taube et al., 1990). Under Medicare, the most salient limits were higher copayments for outpatient mental health services and a limited number of days for hospital care. Medicare’s special limits on outpatient mental health services were changed over the past decade, resulting in significantly increased access to and utilization of such services (Goldman et al., 1985; Rosenbach & Ammering, 1997). The concern, however, is that the gains made as a result of policy changes easily could be eroded by the shift to managed care (Rosenbach & Ammering, 1997).

Increased Role of Managed Care
Projections are that 35 percent of all Medicare beneficiaries will be in managed care plans by the year 2007, amounting to approximately 15.3 million people (Komisar et al., 1997). Although the managed care industry has the potential to provide a range of integrated services for people with long-term care needs, managed care’s awareness of and response to chronic care are rudimentary (Institute for Health and Aging, 1996). Despite the potential of systems of managed health care, such as HMOs, to provide comprehensive preventive, acute, and chronic care services, their current specialized geriatric programs and clinical case management for older persons tend to be inadequate or poorly implemented (Friedman & Kane, 1993; Pacala et al., 1995; Kane et al., 1997). In addition, older patients are likely to be poorly served in primary care settings (including primary care HMOs) because of minimal use of specialty providers and suboptimal pharmacological management (Bartels et al., 1997). Further, current systems lack the array of community support, residential, and rehabilitative services necessary to meet the needs of older persons with more severe mental disorders (Knight et al., 1995). These shortcomings are unlikely to be remedied until more research becomes available demonstrating cost-effective models for treating older people with mental illness.

Carved-In Mental Health Services for Older Adults
The types of mental health services available within managed care organizations vary greatly with respect to how services are provided. In some organizations, mental health care is directly integrated into the package of general health care services (“carved-in” mental health services), while it is provided in others through a contract with a separate specialty mental health organization that provides only these services and accepts the financial risk (“carved-out” mental health services).

Proponents of carved-in mental health services argue that this model better integrates physical and mental health care, decreases barriers to mental health care due to stigma, and is more likely to produce cost-offsets and overall savings in general health care expenditures. These features are particularly relevant to older persons, as they commonly have comorbid somatic disorders for which they take multiple medications that may affect mental disorders, often avoid specialty mental health settings, and incur significant health care expenses related to psychiatric symptoms (George, 1992; Paveza & Cohen, 1996; Moak, 1996; Riley et al., 1997). Unfortunately, mental health specialty services for older persons tend to be a low priority in managed health care organizations, by comparison with medical or surgical specialty services (Bartels et al., 1997). More importantly, carved-in mental health care may have superior potential for individuals with diagnoses such as minor depression and anxiety disorders but tends to shortchange older patients with SPMD who require intensive and long-term mental health care (Mechanic, 1998). The range of outreach, rehabilitative, residential, and intensive services needed for patients with SPMD is likely to exceed the capacity, expertise, and investment of most general health care providers.

Economic factors also may limit the usefulness of mental health carve-ins in serving the needs of older individuals with SPMD. First, evidence from private sector health plans suggests that without mandated parity, insurers offer inferior coverage of mental health care (Frank et al., 1997b, 1997c). Furthermore, if providers or payers compete for enrollees, there is strong incentive to avoid enrollees expected to have higher costs from mental health problems (e.g., older persons with SPMD). To avoid such discrimination, equal coverage of mental health care would have to be mandated through legislation on mental health parity or through specialized contract requirements with managed care organizations.

Carved-Out Mental Health Services for Older Adults
Proponents of mental health service carve-outs for older persons argue that separate systems of financing and services are likely to be superior for individuals needing specialty mental health services, especially those with SPMD. In particular, advocates suggest that carved-out mental health organizations have superior technical knowledge, specialized skills, a broader array of services, greater numbers and varieties of mental health providers with experience treating severe mental disorders, and a willingness and commitment to service high-risk populations (Riley et al., 1997). From an economic perspective, since competition is largely over the carve-out contract with the payer (generally a public organization or an employer), there is less incentive to compete on risk selection, and risk adjustment becomes unnecessary. In addition, mental health carve-out organizations may be better equipped to provide rehabilitative and community support mental health services necessary to care for older persons with SPMD. Finally, growth of innovative outpatient alternatives could be stimulated by reinvestment of savings by the payer from any decrease in inpatient service use.

Unfortunately, research is lacking on outcomes and costs for older persons with SPMD in mental health carve-outs. A carve-out arrangement could lead to adverse clinical outcomes in older patients due to fragmentation of medical and mental health care services in a population with high risk of complications of comorbidity and polypharmacy. Also, from a financial perspective, the combination of physical and mental comorbidities seen in older adults, especially those with SPMD, may reduce the economic advantages of carved-out services (Bazemore, 1996; Felker et al., 1996; Tsuang & Woolson, 1997). If the provider cannot appropriately manage services and costs associated with the combination of somatic and mental health disorders, anticipated savings may not materialize. Furthermore, fragmentation of reimbursement streams would likely complicate the assessment of cost-effectiveness or cost-offsets. For example, apparent savings of mental health carve-outs under Medicare actually may be due to shifting costs when an individual is also covered under Medicaid. In this situation, Medicaid may cover prescription drugs, long-term care, and other services that are not paid for by Medicare. In order to offer true efficiencies, Medicare mental health carve-outs need to find a way to bridge the fragmentation of financing care for older persons.

Outcomes Under Managed Care
There do not appear to be any studies of mental health outcomes for older adults under managed care. In general, the available research on mental health outcomes for other adults consistently finds that managed care is successful at reducing mental health care costs (Busch, 1997; Sturm, 1997), yet clinical outcomes (especially for the most severely and chronically ill) are mixed and difficult to interpret due to differences in plans and populations served. Several studies suggest that outcomes under managed care for younger adults are as favorable as, or better than, those under fee-for-service (Lurie et al., 1992; Cole et al., 1994). In contrast, others report that the greater use of nonspecialty services for mental health care under managed care is associated with less cost-effective care (Sturm & Wells, 1995), and that older and poor chronically ill patients may have worse health outcomes or outcomes that vary substantially by site and patient characteristics (Ware et al., 1996). A recent review of health outcomes for both older and younger adults in the managed care literature (Miller & Luft, 1997) concluded that there were no consistent patterns that suggested worse outcomes. However, negative outcomes were more common in patients with chronic conditions, those with diseases requiring more intensive services, low-income enrollees in worse health, impaired or frail elderly, or home health patients with chronic conditions and diseases. These risk factors apply to older adults with SPMD, suggesting that this group is at high risk for poor outcomes under managed care programs that lack specialized long-term mental health and support services. To definitively address the question of mental health outcomes for older persons under managed care, appropriate outcome measures for older adults with mental illness will need to be developed and implemented in the evolving health care delivery systems (Bartels et al., in press).

6 Primary care includes services provided by general practitioners, family physicians, general internists, certain specialists designated as primary care physicians (such as pediatricians and obstetricians - gynecologists), nurse practitioners, physician assistants, and other health care professionals. General medical settings include all primary care settings plus all non-mental health specialty care.

7 Consultation-liaison models provide a bridge between psychiatry and the rest of medicine. In most models, a mental health specialist is called in as a consultant at the request of a primary care provider or works as a regular member of a team of health care providers.

Back to Top

Home | Contents | Previous | Next