Consensus Statement on the Upcoming Crisis in Geriatric Mental Health: Title and subTitle BreakResearch Agenda for the Next 2 Decades

The 20th century has witnessed an exponential growth in the number and proportion of Americans older than 65 years (from 3 million [4%] in 1900 to 34 million [13%] in 1997), yet this historical rise will be dwarfed by the one that will begin in the year 2011 as the first of the post-war "baby boom" generation (those born between 1946 and 1964) reach the traditionally defined "old" age of 65 years.¹ The economic² and political effects of the aging baby boom generation on the Medicare and Social Security systems has already become a focus of national dialogue, but there is another hidden challenge that has not received due attention. We can also expect an unprecedented explosion in the number of people older than 65 years with potentially disabling chronic mental illnesses.³

A national crisis in geriatric mental health care is emerging. The current research infrastructure, health care financing, pool of mental health care personnel with appropriate geriatric training, and mental health care delivery systems are extremely inadequate to meet the challenges posed by the expected increase in the number of elderly persons with mental illnesses. Deliberate and coordinated action is urgently warranted.

A workshop held in San Diego, Calif, in March 1998 discussed the current state of affairs and the upcoming crisis in geriatric mental health and recommended a research agenda. This article represents a statement of consensus among the individual participants who came from different institutions, organizations, and advocacy groups and represented various disciplines.

Aging and aging-associated disorders (eg, vascular disease) may influence expression and recognition of mental illnesses.⁴ Yet, psychiatric epidemiologic studies have frequently excluded elderly subjects.^{5 - 6} Table 1 presents the prevalence of major psychiatric disorders in elderly persons per the Epidemiologic Catchment Area (ECA) study⁶ (with comparative figures from younger adults) and the prevalence of "clinically significant" symptoms reported in some investigations. The importance of the latter is illustrated by the finding that depressive symptoms not meeting criteria for major depression are often associated with adverse outcomes in older adults.¹⁵

Per the ECA study,⁶ 13% of elderly persons would meet criteria for psychiatric disorders other than dementia. We consider this a serious underestimation secondary to factors such as misattribution (by subjects and raters) of psychiatric symptoms to cognitive impairment, physical disorders, or "normal aging"; lack of age-appropriate diagnostic criteria for substance abuse or dependence¹⁶ or schizophrenia^{17 - 19} ; and underreporting of psychologic symptoms due to forgetfulness and social stigma. We believe the "real" prevalence of psychiatric disorders other than dementia in elderly persons to be at least 25% higher—ie, about 16.3%. Additionally, about 10% of elderly persons have dementia, usually of the Alzheimer type^{20 - 21} ; noncognitive psychiatric disturbances are common in patients with dementia, with at least a third of them (about 3% of the total elderly population) exhibiting psychotic and/or depressive symptoms.²² Thus, a total of 19.6% of elderly persons would have significant psychopathological symptoms. (This figure does not include people with delirium or other "mental disorders secondary to general medical conditions" [DSM-IV], which are particularly common in the geriatric population.)^{1 ,23}

The number of Americans aged 30 to 44 years is estimated to increase from about 41.1 million in 1970 to 65.8 million in 2030, while the number of people aged 65 years and older will grow from 20.0 million to 69.4 million.¹ We believe that there will be a disproportionately greater increase in the numbers of mentally ill elderly persons. Excess mortality (from suicide or physical comorbidity) in early life among patients with schizophrenia, depression, substance dependence, and other psychiatric disorders was thought to be an important reason for a lower prevalence of serious mental illnesses in elderly compared with younger adults found in the ECA study.⁶ With an improved overall standard of living, better treatment for physical disorders, and more effective treatments for psychiatric illnesses, the high mortality in mentally ill young adults should begin to decrease, resulting in a greater number of those who will reach old age. Additionally, with the general increase in the numbers of people older than 65 years, there will be more elderly individuals in the community with the possibility of developing late-onset psychiatric disorders. Furthermore, aging baby boomers are expected to have a higher risk of depression and anxiety disorders,²⁴ as well as substance abuse,²³ than the current cohort of elderly adults. For all these reasons, we estimate that the prevalence of elderly mentally ill adults in the population will increase by at least 10% during the next 30 years—ie, by year 2030, it will be about 21.6%. Thus, the numbers of psychiatrically ill elderly persons will rise by 275%—from 4 million in 1970 to 15 million in 2030, while comparable numbers for those aged 30 to 44 years will increase by only 67% (Figure 1).^{1 ,6 ,24 - 25}

We realize that the above estimates are educated crude guesses; nonetheless, they provide some understanding of the magnitude of the challenge facing us.

For several disorders, there will be higher proportions of patients at both ends of the severity spectrum. Thus, expected advances in the ability to detect Alzheimer disease in earlier stages and to decrease the rate of decline via use of effective neuroprotective agents will result in greater numbers of patients who are more mildly affected and who will present different management problems from patients with advanced dementia. Issues regarding employment, financial planning, driving, home care, and day care will become more pressing. On the other hand, decreased mortality from severe forms of physical illness will lead to increased numbers of severely demented patients with comorbid medical conditions living longer, raising a new set of clinical, research, and ethical issues.

Older adults with mental disorders frequently consult a generalist physician and not a mental health specialist,^{26 - 27} so the primary health care setting is pivotal in the management of elderly mentally ill patients. Primary care clinicians often fail to detect and treat mental disorders and suicidality in older patients.²⁸ In one study, 55% of the internists felt confident in diagnosing depression but only 35% felt confident in prescribing antidepressants to older persons.²⁹ Three quarters of physicians thought that depression was "understandable" in older persons,³⁰ consistent with other studies,³¹ suggesting pervasive ageism. In one report³² only 2 of the 45 adults older than 50 years who committed suicide and who had been found by their primary care physicians during the prior month to have psychiatric symptoms had been given adequate treatment. Inadequate treatment included prescriptions of incorrect medicines (eg, benzodiazepines for major depression) or inadequate doses (eg, 25 mg of a tricyclic antidepressant). Another study³³ found that less than 25% of patients with moderate to severe dementia were identified by general practitioners as having dementia.

The numbers of health care professionals currently available to treat elderly mentally ill persons in the United States are inadequate. Presently there are 2425 board-certified geriatric psychiatrists³⁴ and 200 to 700 geropsychologists^{35 - 36} ; a need for at least 5000 of each specialty is projected.³⁵ Little is known about the numbers of primary care physicians, nurses, and social workers caring for older adults with psychiatric disorders.

Recent waves of deinstitutionalization for older adults with severe mental illnesses have led to either transinstitutionalization from state hospitals into nursing homes, where access to mental health treatment is reduced,³⁷ or a move into the community, where necessary support services are lacking. A sizable proportion of nonelderly persons in the criminal justice system who have mental illnesses such as schizophrenia⁶ will reach old age there.

As rates of inpatient service use by older adults have declined, psychiatric outpatient service use has climbed.³⁸ Nonetheless, community-dwelling older persons with severe mental illnesses do not utilize outpatient mental health services adequately³⁹ and are at high risk of "falling through the cracks."^{40 - 41} Community mental health organizations tend to underserve elderly persons,⁴² lack staff trained to address medical needs, and often exclude those with cognitive impairments.⁴¹ Medicare is unsatisfactory for severely mentally ill elderly patients—eg, it offers no general outpatient prescription drug coverage and limits inpatient hospital days.⁴³ Medicaid often imposes restrictions on the number of visits to health care providers.⁴³

Dramatic changes in the structure and financing of long-term care and managed care are proceeding at a rapid pace across the nation, with a virtual lack of attention to older adults with severe mental disorders. The recently developed models of managed mental health care do not assure quality medical health care, clarify the role of cognitive and psychosocial rehabilitation, or identify the optimal mix of services necessary to maintain the older patient with severe persistent mental disorders in the community as long as possible.⁴⁴ With the rush toward Medicare managed care, questions arise about the quality of health care, such as limited provider panels, precertification of hospitalization, and restrictive formularies.⁴⁵ Hence, older persons requiring intensive or long-term mental health care will not fare well in the new era of managed care.

We recommend a concerted and combined effort on the part of federal and private agencies to formulate a 15- to 25-year plan for research on mental disorders in elderly persons. The Women's Health Initiative, funded by the National Institutes of Health, that began in the latter part of the 1980s is one kind of model to follow.⁴⁶ This will need not only greater collaboration among different professional groups but also a reduced social stigma against elderly mentally ill persons. One way to improve the public image of (and support for research on) mental disorders in elderly persons is to put a "human face" on these illnesses. The dual social stigma associated with psychiatric disorders in elderly persons will persist unless the lay public becomes aware that such illnesses can and do occur in otherwise healthy people. A recent example in this regard is the inspiring story of the Nobel laureate, John Nash,⁴⁷ who had remission of schizophrenia after 25 years of the illness.

Below we present a research agenda under categories of prevention, translational research, intervention studies, health services research, and training.

The concept of prevention is not restricted to children and can be applied to elderly persons as well.⁴⁸ Examples of primary prevention include interdisciplinary collaborative trials for reducing risk factors for cerebrovascular disease and thereby preventing a form of late-onset depression with cerebrovascular etiology⁴ and routine screening of at-risk individuals for suicide, depression, dementia, or substance abuse, followed by appropriate management. To illustrate the public health value of prevention, delaying the onset of Alzheimer disease by just 5 years (eg, through the use of possible neuroprotective agents such as antioxidants) could decrease the number of people with Alzheimer disease by 50% in one generation, because many elderly individuals in the at-risk group would succumb to other diseases before manifesting dementia.⁴⁹

Examples of translating research results from bench to bedside range from studies of "successful" cognitive aging to those of biological markers of Alzheimer disease. Presently, several putative biological markers for Alzheimer disease, including apolipoprotein E and τ protein, are being studied. None of them is, however, clinically optimal. There is still a need for a more convenient, specific, and low-cost biological marker to allow a definitive diagnosis of Alzheimer disease prior to autopsy, facilitate its early detection, and reduce the demand for clinical expertise in the diagnosis, thus improving the accuracy of its diagnosis by nonspecialists.²⁵

From another perspective, age-appropriate modifications of diagnostic criteria for various psychiatric disorders (including mood disorders, schizophrenia, and substance abuse) are required to better define the phenotypes of elderly psychiatric patients for genetic and other investigations. More importantly, new conceptualizations of several mental illnesses are warranted. Much of our current perspective is based on studies of young adults (and, sometimes, institutionalized elderly persons). For example, the Kraepelinian concept of dementia praecox⁵⁰ has led to a strong prevailing notion that the onset of this illness is restricted to adolescence and young adulthood, and that it usually leads to dementia in old age. Such thinking hinders broader conceptualizations suggested by late-onset schizophrenia,^{51 - 52} as well as remission of the illness in later life. Strategies to delay the onset or to produce remission of illness are required. Attention should also be paid to studying other serious but neglected late-life mental disorders such as anxiety disorders, alcohol dependence, and personality disorders. Reports of relatively low prevalence of psychiatric illnesses among elderly persons (Table 1) have contributed to a dismissive neglect of such conditions in the past.

Issues of polypharmacy, medical comorbidity, greater risk of adverse effects, and age-related changes in pharmacokinetics and pharmacodynamics make it vital to conduct trials with elderly subjects. We recommend that Food and Drug Administration approval for psychotropic drug use in elderly persons require clinical trials in that age group. Similarly, federally funded, large-scale, psychopharmacologic investigations using quality of life and everyday functioning as outcome measures are needed in diverse geriatric populations—eg, treatment studies of depression in primary care settings⁵³ should include minor depression as well as depression accompanying physical illness. A plan similar to that recently proposed by the National Institute of Mental Health, titled "Bridging Science and Service,"⁵⁴ should be developed specifically for older adults.

Studies of nonpharmacologic interventions such as brief and focused psychotherapies, cognitive behavioral therapies, social skills training, and case management for elderly patients with mental illnesses should be standardized and tried in various settings, either alone or in combination with pharmacotherapy. Similarly, the role of psychosocial management for anxiety or depression in caregivers of patients with dementia should be evaluated.⁵⁵ The growing use of nontraditional therapies also deserves scientific evaluation.⁵⁶

Research is needed on decision-making capacity in mentally ill elderly persons.⁵⁷ A balanced approach that takes into account patients' vulnerability as well as the need to conduct appropriate research studies to improve their care is warranted.

Another area of intervention research concerns ethnic minorities, whose proportions among American elderly persons are expected to increase.¹ Ethnic and cultural factors influence manifestations of mental disorders, reporting of psychiatric symptoms, and diagnoses and treatments received.⁵⁸ Ethnic differences in pharmacokinetics and pharmacodynamics contributing to differential drug responsiveness⁵⁹ should also be explored.

At a time when managed care and long-term care are dramatically changing the face of health care in the United States, studies to assess the effect of these reforms on quality of care for elderly mentally ill adults are warranted. The role of specialty mental health services, development and implementation of practice guidelines, and assessment of outcomes deserves attention. This also applies to general medical care because physical comorbidity in elderly mentally ill persons (eg, those with schizophrenia) is often compounded by the poor general medical care available to them.⁶⁰

As the knowledge base grows more complex, the information gap between specialists and primary care continues to widen. Another unsettled issue is the role of mental health care providers in treating patients with dementia.⁴³ We believe that psychiatrists, psychologists, and other mental health professionals bring unique skills to the management of patients with dementia, particularly those with noncognitive psychological and behavioral symptoms that increase caregiver burden.⁶¹ Quality of life of nursing home patients should also be among the critical areas for services research.

Cost of mental health care for disorders such as schizophrenia⁶² may be higher in elderly persons than in younger age groups. Pharmacoeconomic investigations of the newer medications, which are safer⁶³ but more expensive than older drugs, are necessary in the geriatric population.

Training in geriatric mental health should be improved at various levels including future primary care clinicians (medical students, residents, nurse practitioners); local law enforcement officers, letter carriers, and public utility workers (through courses in helping seniors with mental illnesses living alone); and the public at large (with media campaigns and health fairs about depression, anxiety, and memory loss). At most levels of professional training, stronger financial incentives, such as increased stipends and salaries and loan forgiveness, should be provided.³⁴

Below we briefly discuss 2 specific strategies of research training for young investigators and for clinicians.

Summer Research Institute in Geriatric Psychiatry⁶⁴ is a national initiative that includes a 1-week intensive and interactive workshop on research methods for 25 to 30 selected fellows and junior faculty, followed by ongoing communication and mentoring. It provides a useful model for bridging and shortening the transition period from fellowship to first research funding, and for assuring a continued flow of new investigators. The program has been highly successful, judging from the trainees' accomplishments in terms of publications and research funding during follow-up.⁶⁵

There is often a considerable gap between clinical practice and research literature. For example, the newer antipsychotics and antidepressants are commonly prescribed to geriatric patients, but those clinical data are rarely published. This situation can be improved by training selected clinicians in the use of practical but valid outcome measures, and pairing clinicians with appropriate researchers so that reliable clinical data with large sample sizes can be published.

The numbers of elderly persons with significant psychopathological disorders are anticipated to nearly quadruple from 1970 to 2030. Presently most psychiatrically ill elderly persons are managed in the primary care setting, where the typical health care professional is inadequately trained to diagnose and treat geriatric psychiatric disorders. With the managed care market pressure favoring lower-cost, nonspecialized services, the gap between necessary treatment and available management will widen further. This problem is compounded by a long-standing neglect of geriatric issues in traditional psychiatric research, although there has been some recent progress in this area. We recommend a multipronged approach, beginning with the formulation of a 15- to 25-year plan for research on mental disorders in elderly persons. The research agenda should include prevention, translational research, intervention studies, and health services research. The focus should be on innovative strategies—eg, investigations directed toward delaying onset and enhancing remission of chronic mental illnesses. Another critical area is training of clinicians, researchers, policymakers, and the lay public in issues related to geriatric mental health. Through concerted and timely actions, the upcoming crisis could be converted into an opportunity to improve our understanding as well as management of mental illnesses in elderly persons.

Accepted for publication June 3, 1999.

This article was supported in part by grants from the National Institute of Mental Health, Rockville, Md; Department of Veterans Affairs, Washington, DC; and unrestricted educational grants from Abbott Laboratories, North Chicago, Ill; Bristol-Myers Squibb Pharmaceuticals, Wallingford, Conn; Forest Laboratories, St Louis, Mo; Janssen Pharmaceutica and Research Foundation, Titusville, NJ; Eli Lilly and Co, Indianapolis, Ind; Novartis Pharmaceuticals Corp, Summit, NJ; Pfizer Inc, New York, NY; and Zeneca Pharmaceuticals, Wilmington, Del.

This article is based on presentations and discussion at a workshop on "The Future of Mental Health and Aging," San Diego, Calif, March 11, 1998.

This article represents a statement of consensus among all the participants in their individual capacity. We thank the following for their valuable comments: Norman Abeles, PhD; Soo Borson, MD; Marc Cantillon, MD; Chris Colenda, MD, MPH; Horace Deets; Janet Duffey, RNC, MS; Laurie Flynn; Richard Glass, MD; Robert Glover, PhD; Larry S. Goldman, MD; Thomas Horvath, MD; Steve Hyman, MD; John Kennedy, MD; James Lavery; Laurie Lindamer, PhD; Rick Martinez, MD; Rodrigo Munoz, MD; Janet Pailet, JD; Tim Slone; Leon Thal, MD; and Peter Whitehouse, MD, PhD.

Corresponding author: Dilip V. Jeste, MD, University of California, San Diego, VA San Diego Healthcare System, 116A-1, 3350 La Jolla Village Dr, San Diego, CA 92161 (e-mail: djeste@ucsd.edu).