Geriatric mental health services research: Strategic Plan for an Aging Population: Report of the...

Am J Geriatr Psychiatry 9:3, Summer 2001 191

CONSENSUS STATEMENT

Geriatric Mental Health Services Research:Strategic Plan for an Aging Population

Report of the Health Services Work Group of theAmerican Association for Geriatric Psychiatry

Soo Borson, M.D., Stephen J. Bartels, M.D., M.S.Christopher C. Colenda, M.D., M.P.H., Gary L. Gottlieb, M.D., M.B.A.

Barnett Meyers, M.D.

In November 1999, a working group of the American Association for Geriatric Psy-chiatry (AAGP) convened to consider strategic recommendations for developing ge-riatric mental health services research as a scientific discipline. The resulting consen-sus statement summarizes the principles guiding mental health services research onlate-life mental disorders, presents timely and topical priorities for investigation withthe potential to benefit the lives of older adults and their families, and articulates asystematic program for expanding the supply of well-trained geriatric mental healthservices researchers. The agenda presented here is designed to address critical ques-tions in provision of effective mental health care to an aging population and thehealth policies that govern its delivery. (Am J Geriatr Psychiatry 2001; 9:191–204)

Received October 12, 2000; accepted October 30, 2000. From The University of Washington, Dartmouth University, Michigan State University,Harvard University, and Cornell University. Address correspondence to Dr. Borson, University of Washington, Department of Psychiatry andBehavioral Sciences, Box 356560; Seattle, WA 98195-6560.

Copyright � 2001 American Association for Geriatric Psychiatry

At the dawn of the new century, two seminal reportsmark a turning-point in the national agenda for re-

search and mental health services in America: “BridgingScience and Service,” from the National Institute ofMental Health,1 and “Mental Health: A Report of theSurgeon General.”2 A third, the “Consensus Statementon the Growing Crisis in Geriatric Mental Health,”3 spe-cifically highlights the rising prevalence of mental dis-orders in older people and proposes a research programto meet this emergent need. All three reports recom-mend systematic development of a vital program of ge-riatric mental health services research to ensure thetranslation of advances in clinical science and therapeu-tics into better care for older adults.

BACKGROUND

The science of geriatric mental health services researchbegan during the 1960s and ’70s with exploration ofthe impact of mental health reforms on the fate ofchronically mentally ill patients discharged from statehospitals into the community.4,5 In the 1980s, new stud-ies confirmed findings of preliminary studies reportingthat prevalent geriatric mental disorders were under-treated in both mental and general health care set-tings.6,7 In the last decade, geriatric mental health ser-vices research has emerged as a national priority, asscientifically based understanding of the origins and

Mental Health Services Research

192 Am J Geriatr Psychiatry 9:3, Summer 2001

TABLE 1. Contrasting characteristics of clinical and health services research

Clinical Research Health Services Research

Subjects Patients; may involve families and caregivers Populations, services, or systems; mayinvolve providers, payers, andpolicymakers

Focus Patients and disorders Models of service deliveryInterventions Treating disorders Changing organizational structure, process,

or reimbursement methodOutcomes Efficacy Effectiveness

treatment of late-life mental disorders have dramaticallyimproved clinicians’ ability to diagnose and treat olderpatients. Patients with the most common mental dis-orders of late life, including depression, dementia, andemotional and behavioral disorders that cross conven-tional diagnostic boundaries, as well as those with lesscommon but severe mental disorders such as schizo-phrenia, can now be reliably identified and treated.However, successful translation of findings from preva-lence studies and controlled treatment trials into thediverse settings in which older adults receive their carehas lagged far behind. Mental health services researchbridges these gaps by identifying barriers to effectivedissemination of new knowledge, testing ways to over-come them, and evaluating the effectiveness and costsof interventions and service models in real-world set-tings.

The Nature of Mental Health Services Research

Mental health services research complements con-ventional clinical research but differs in emphasis in sev-eral important ways (Table 1). Familiar clinical psychi-atric research models describe and delineate clinicalsyndromes, develop new therapies, and assess thera-peutic efficacy. Clinical research studies typically userestrictive experimental designs to discover whether aspecific treatment works in defined patient samplesand/or to compare it with alternative treatments. Thesestudies are typically blinded or otherwise controlledand use limiting inclusion and exclusion criteria. In con-trast, mental health services research examines the ef-fectiveness of mental health care in a population andemphasizes the capacity of service systems to delivercare in the heterogeneous populations encountered inactual service settings. It may also include assessmentof outcomes defined not only by clinical parameters,but also by function, treatment acceptability, serviceutilization, and costs. Second, health services researchincorporates a variety of methodological approaches to

address the organization, financing, and delivery of carein broad populations and diverse settings. It includesrandomized services trials comparing the effectivenessof alternative treatment interventions, as well as prac-tice-based studies using the methods of clinical epide-miology, quality improvement, practice- based changestrategies, and treatment dissemination. Third, servicesresearch includes studies of delivery systems and theeffects of differences in organization and financing onoutcomes. Hence, services research draws upon widelydivergent disciplines and methodologies. Its goals arebroader than those of clinical research, and its scientificmethods are chosen to inform decisions about healthcare organization and policy. Multiple stakeholders—providers, patients, policy-makers, purchasers of healthcare commodities, and insurers—have important inter-ests in decisions regarding access to and effectivenessof treatments for older patients with mental disorders.In this respect, services research is uniquely positionedto bridge science and health policy in response toemerging public health challenges.1

The Challenge of an Aging Population

The aging of the population presents one of themost significant international challenges to health carein this new century (Figure 1). Within a decade, 13 ofevery 100 Americans will be over the age of 65, andmore than 1 in every 100 will be over age 85. The mostrapid growth in the older population, occurring in theoldest-old, demands special attention to this group athigh risk for mental disorders because of multiple func-tional impairments, medical disorders, and vulnerabili-ties to diverse physiological and psychosocial stressors.This group is also at highest risk for neurodegenerativedementias, with their associated psychiatric morbidity.Research on mental health and aging underscores thespecial service needs of persons with coexisting medi-cal comorbidity, physical frailty, cognitive changes, anddiminished access to care. The Surgeon General’s Re-

Borson et al.


FIGURE 1. United States population growth 1940-201087

6.8

0.3 0.5

9.2

11.3

0.97 1.2

12.5

1.91

13.3

1940

132 million

2010

298 million

1990

249 million

1980

227 million

1960

180 million

Total U.S. Population

14

12

10

8

6

4

2

0Perc

enta

ge o

f U.S

. Pop

ulat

ion

>85 years of age >65 years of age

port on Mental Health2 highlights the explosion inknowledge of effective treatments for late-life mentaldisorders over the last decade while proposing solu-tions to problems in implementation of these findingsin clinical settings. The escalating need for effective ge-riatric mental health treatments and services supportsa compelling argument for expanding the pool of cli-nicians and researchers and a research and trainingagenda that includes health services among its criticalcomponents.3

The present document proposes such an agenda formental health services research specific to the geriatricpopulation. It is not intended to provide a comprehen-sive overview of the broad array of possible researchdirections in the field of geriatric mental health servicesresearch or to highlight those few areas that alreadybenefit from an existing research literature or ongoingbroad-based research initiatives. Rather, priority targetsare identified that aim to close the existing knowledgegaps in areas that have received inadequate attention todate, yet have high potential to improve the mentalhealth of older adults.

PRIORITIES FOR GERIATRIC MENTALHEALTH SERVICES RESEARCH

Priorities may be conceptualized from several perspec-tives within the field of mental health services research(Table 2): 1) high-risk target populations in need of ser-vices; 2) key service delivery settings and providergroups; 3) translation of research findings into clinicalpractice by describing—and changing—provider and

consumer behavior; 4) the needs, preferences, and in-volvement of consumers and families in decision-mak-ing and treatment; and 5) service delivery systems, in-cluding the effect of organizational and financial factorson access, outcomes, and costs.

Priority Populations

Priority populations for geriatric mental health ser-vices research are those with high prevalence, risk,chronicity, or costs of mental disorders, for which ex-isting treatment models and services are poorly definedor inadequately implemented. This perspective targetsthree overlapping groups: 1) older persons with co-morbid disorders; 2) the oldest and frail elderly; and 3)older adults with chronic mental disorders.

Comorbidity. Comorbidity, or the coexistence of twoor more conditions at least one of which is a psychiatricillness, is the norm among older persons with mentaldisorders.8 Chronic medical disorders, cognitive disor-ders due to neurodegenerative disease or medical fac-tors, and medication or alcohol misuse commonly oc-cur with psychiatric disorders such as depression oranxiety. Chronic medical disease is an established riskfactor for mental disorders,9 and patients with severelong-lasting mental illness, like other aging persons, fre-quently develop major medical problems (e.g., elderlypatients with schizophrenia and diabetes or chroniclung disease).8 The prevalence of psychosis, depression,or behavioral symptoms or disorders in patients withAlzheimer’s disease and other dementias is well known,but a growing number of individuals with severe andpersistent mental illness experience worsening cogni-tive function with age10 that affects strategies for care.Also, comorbid psychiatric symptoms or disorders arecommon among individuals with alcohol or prescrip-tion, over-the-counter, or street-drug misuse and abuse,although little is known about the characteristics ofsuch “dual disorders” in older adults and what servicemodels may be effective for treating them.11

A major challenge for dealing effectively with co-morbid disorders in older adults derives from fragmen-tation of services and financing across different healthcare delivery systems (“vertical fragmentation”). Pa-tients may require services by specialty mental health,primary medical, aging non-mental health, and/or sub-stance abuse treatment providers working in very dif-ferent locales and settings that lack defined pathways



TABLE 2. Priorities for geriatric mental health services research

Populations Settings Issues Training Funding

ComorbidityFrailty in the oldest-oldSevere mental illness

Spectrum of long-term care

Primary andspecialty care

Treatment effectivenessProvider behaviorPreferences and decision-

makingDiscontinuities in careStructure and behavior of

health systemsQuality of national databases

Supply of investigatorsDefining core

competenciesIncreasing training

opportunitiesCollaborative funding

initiatives

ResourcesNIMHNIAFoundationsMedicare, HRSAdemonstrationprojects

AHRQSAMHSA

Note: NIMH�National Institute of Mental Health; NIA�National Institute on Aging; HRSA�Health Resources and Services Administration;AHRQ�Agency for Healthcare Research and Quality; SAMHSA�Substance Abuse and Mental Health Services Administration.

for integrating care into a coherent whole.12,13 Researchis needed to identify effective service models that ad-dress the complexity of clinical needs and providers en-countered by older adults.14 Moreover, the continuityof care over time is also frequently disrupted (“horizon-tal fragmentation”) for older adults with chronic im-pairments. Research specifically targeting cognitivelyimpaired older adults is required to assess use of mentalhealth services and continuity of care, to define howexplicit partnerships between providers and proxy de-cision-makers (usually family members) affect long-termoutcomes, and how impaired elderly persons livingalone without family to help can best be identified andcared for. There is evidence that underutilization ofmental health services and discontinuities in care con-tribute to untoward outcomes, such as higher rates ofrehospitalization in dementia patients with depres-sion,15 increased utilization of more expensive emer-gency and inpatient care resources in depressedmedically ill patients,16,17 and entry into long-terminstitutional care.18

Frailty and the “oldest-old.” Frailty results from thecombined effects of advanced age, cognitive impair-ment, diminished physiological and psychological re-serve, and identifiable mental and medical disorders invarious combinations,19–21 and it increases risk for un-recognized and/or undertreated psychiatric as well asmedical complications.22,23 Frail elderly persons can betargeted by use of simple tools,24 and several ap-proaches to improving health outcomes and reducingrisks in this population have been tested in randomizedcontrolled trials.25–28 However, the cognitive, behav-ioral, and mood components of frailty have been inad-equately examined in predictive and intervention re-search.29 The oldest-old who need everyday assistance

and those with psychiatric symptoms are at high riskfor institutionalization.30,31 A tactical approach to lon-gitudinal research on improving care of frail elderly per-sons builds on the concept of “sentinel events” thatbring an at-risk individual to clinical attention and allowdiscovery of broader needs at a time when provision ofwell-targeted interventions can optimize overall out-comes. Sentinel events may be emergency room visits,32

hospital admissions, falls, fractures, subnutrition, dehy-dration, or the development of delirium, stroke, or myo-cardial infarction, anxiety, depression, or aggressive be-havior. Neglect of the opportunities provided byrecognizing the sentinel nature of these events leads toavoidable chronicity and changes in levels of care andthe unmet need for specialized follow-up and care whenchronicity is inevitable. Longitudinal research is neededto define the unique contribution of psychiatric com-ponents of frailty to risks and outcome of sentinelevents and to determine whether early identificationand intervention improve clinical and family outcomes,prolong successful adaptation to community living, andlead to more efficient and appropriate use of health andsocial care. Frail elderly patients, overrepresentedamong the oldest-old, have been largely neglected inmental health services research and have just begun tobe a focus for change in general health care sys-tems.33–35

Severe mental illness. Older adults with severe mentalillnesses (schizophrenia; delusional disorder; bipolardisorder; and severe, recurrent, or treatment-refractorymajor depression) comprise a growing population thataccounts for disproportionate service use and costs.36,37

A substantial mental health service research literaturedocuments the effectiveness of specific treatment inter-ventions and services for young adults with severe men-

Borson et al.


tal illness, yet little is known about effective servicemodels for older adults. Such gaps in knowledge un-derscore the need for research that focuses on this high-risk group.3,38–40 Addressing the needs of older personswith severe mental illness for rehabilitation and long-term mental health care in the community will requiredevelopment and testing of innovative service and fi-nancing models. These models should adapt successfulrehabilitation concepts and services for younger per-sons, while recognizing that the older population posesunique challenges with respect to horizontal and verti-cal fragmentation of services, the high prevalence ofmedical and cognitive comorbidity, and the complexinteraction of long-term mental health and generalhealth care needs.41,42

Service models developed for chronic medical con-ditions may also hold promise for older adults with se-vere mental disorders. For example, specialized chronicdisease management models designed to streamline ser-vice delivery and reduce adverse outcomes for highlyprevalent chronic conditions such as diabetes43 havedefined core components of chronic disease care thatcan be generalized across diagnoses and applied to psy-chiatric disorders.44 Explicit behavioral outcomes arethe key measures of effectiveness in chronic-care mod-els regardless of diagnosis. Behavioral outcomes appro-priate for older adults will require conceptual devel-opment and adaptation for those with cognitive andmotivational deficits. Finally, any examination of the ef-fectiveness of intensive or supported interventionsmust consider the associated costs of innovative servicedesigns, while also documenting potential savings ac-crued through prevention of adverse outcomes andcostly institution-based services.

Priority Service Delivery Settings

The spectrum of long-term care settings. Althoughthe great majority of older individuals with mental dis-orders live in the community, many require varyingforms and levels of assistance that are currently beingprovided by families, advocacy groups, the aging ser-vices network, public and private care agencies, anddiverse health care systems.12,13,31 The costs of thesedisparate components are generally supported by mul-tiple payment mechanisms that are tracked over timeby different, often incompatible accounting systems,and their administrative fragmentation makes it difficultto assess outcomes or collect accurate research data.

Interdisciplinary models of community-based geriatricmental health care have the potential to provide the fullspectrum of services required by frail elderly patients,45

pointing the way for large-scale trials in at-risk popula-tions. Models of community-based long-term care canprovide innovative alternatives to institutional care,31,46

although cost-effectiveness studies suggest that servicesmust be carefully targeted in order to demonstrate su-perior efficiency.47–49 Health services research isneeded to test the effectiveness of interventions andservice models for older adults with mental disorderswho are discharged from hospitals and emergencyrooms after treatment for sentinel events and those re-ceiving home health care, social case management,adult daycare, assisted-living services, and long-termnursing home care. The costs of care in the various com-ponents of the long-term care spectrum are largelyborne by families and public service funds. Moreover,the services provided usually include mental health careas an afterthought, if at all.

In nursing homes, general models for providingmental health services have yet to be tested, and em-pirical efficacy and cost studies have been limited tospecific subpopulations.50 The assumption that nursinghome placement for mentally impaired elderly personsreflects avoidable failures of less-costly services has ledto an emphasis on studies of patient placement into less-restrictive settings; most of these studies lack robustclinical-outcome indicators or longitudinal designs.51

Older adults with mental disorders in nursing homesare a heterogeneous group that includes many individ-uals unsuited to community placement because of thecomplexity of their total care needs, as well as otherpatients who could move into the community but lackaccess to appropriate support services. A large body ofnursing home regulations that require review of psy-chotropic medication prescriptions, coupled with thehigh prevalence of psychiatric disturbances in resi-dents,52 has spawned a major expansion in nursinghome mental health consultation services.53 At thesame time, the greatest growth in long-term care is inassisted-living facilities and home- and community-based alternatives where regulations are few and mentalhealth assessments and services are not required. Ques-tions about the adequacy and appropriateness of mentalhealth services in long-term institutional care continueto engender controversy at the health-policy level, andputative fraudulent claims and other abuses54 have sto-len the limelight from serious consideration of how ser-



vices can be targeted, organized, funded, and deliveredfor improved outcomes. National surveys of nursinghome expenditures suggest that specialty mental healthproviders underserve residents, particularly those withdementia complicated by psychiatric disorders, andclinical studies indicate that unmet needs for psychiat-ric services are widespread.55–58

Primary and specialty health care settings. The ma-jority of older persons receiving any mental health careare treated by their general medical providers,59 a factthat underscores the importance of effective teachingof basic geriatric mental health skills to future primarycare physicians.60 The care of older adults with comor-bid psychiatric disorders is more costly, exclusive ofmental health services, than is care of older adults withmedical problems alone.61 Integrated models of primarycare and mental health services have been proposed asa means to address the mental health needs of primarycare patients more effectively.62,63 Accordingly, severalinitiatives sponsored by the Substance Abuse and Men-tal Health Services Administration (SAMHSA), the Na-tional Institute of Mental Health (NIMH), the Veterans’Administration (VA), and the Hartford Foundation areunderway to test the effectiveness of integrating mentalhealth services for depressed older patients into pri-mary care settings. The VA health care system is alsotesting a model of coordinated care for older personswith mental disorders or substance abuse newly dis-charged from medical and surgical services (the UP-BEAT program). These efforts hold promise for over-coming the major challenges impeding effective mentalhealth service delivery in primary care settings. Theyalso set the stage for developing testable models formanagement of other disorders, such as dementia withpsychiatric morbidity, on a broad scale. Research ex-amining the costs and effectiveness of this second gen-eration of mental health interventions and service mod-els will be a continuing priority. Other studies are alsoneeded to establish effective approaches to ensuringquality general health care for individuals primarilyserved within the specialty mental health sector, suchas older persons with chronic schizophrenia and othersevere mental illness who are at high risk for unrecog-nized and untreated medical comorbidity.41 Medicalpopulations receiving most of their care from specialtymedical providers, such as those with chronic and se-vere cardiac and respiratory diseases, are similarly athigh risk for psychiatric morbidity that has adverse

health and service-utilization outcomes,64 and they re-quire specialized intervention models that can be vali-dated in research. Targeted approaches to test the ef-fectiveness of integrated mental and general health carefor frail elderly patients are particularly important forprimary care and social-service settings. The limited de-cision-making capacity, mobility, and ability to advocatefor, organize, and monitor their own care make suchindividuals dependent on others to access the compre-hensive care they require.

TRANSLATING RESEARCH INTOCLINICAL PRACTICE: PRIORITIES FORIMPROVING EFFECTIVENESS IN REAL-

WORLD SETTINGS

Changing Provider Behavior

Dissemination and testing of research advances inclinical practice require systematic study of methods forchanging how doctors and other health and social- ser-vice providers deal with geriatric mental health prob-lems. Mental health services research has largely ne-glected the settings that care for older adults, includingprivate medical practices, primary care clinics, man-aged care organizations, nursing homes, and commu-nity-based social and health agencies. Clinical care forcommon chronic or prevalent conditions varies widelyand resists change through simple informational strate-gies and dissemination of authoritative guidelines alone,even when more intensive team approaches, such asacademic detailing and continuous quality improve-ment are used.65,66 Success rates in changing providerbehavior in managing acute, self-limited diseases aretypically higher than for chronic diseases. A recent re-view of empirical studies of provider change strategiesapplied to psychiatric disorders67 presents a more op-timistic view: success is greater when community-basedinterventions are used that enable and reinforce prac-tice change and when patient-based interventions areadded to amplify gains. The intrinsic appeal of theseapproaches warrants closer scrutiny for application toimproving the care of geriatric patients with mentalhealth problems, particularly in screening for commondisorders, managing comorbidity, increasing access,and updating treatment to conform to dynamic changein evidence-based practices. Evaluating the usefulnessof these approaches in the treatment of mental disor-

Borson et al.


ders in older adults will be facilitated by populationstudies targeting collaborative care or specialty mentalhealth services, the characteristics of practices mostlikely to respond to incentives for change, and moni-toring of carefully selected therapeutic outcomes ap-propriate to older adults. For general application in ge-riatric mental health services research, templates forassessing technologies for information transfer and pro-vider behavior change can be adapted from existing ge-riatric medical interventions to achieve limited objec-tives in defined environments and from studies in otherpatient populations. These include studies of successfulacademic detailing relevant to mental health care of el-derly patients68,69 and the schizophrenia PORT (PatientOutcomes Research Team) project.70 Additional targetsfor research include interventions and processes aimedat changing provider behavior such as decision analysisand support,32 explicit incentives for adherence to prac-tice guidelines, and system-wide interventions utilizinguniform service protocols and outcomes assessments.

Testing Treatment Effectiveness

Treatments found to be effective in controlled trialsof highly selected populations in academic research set-tings need to be implemented and tested in actual prac-tice. This is a major objective of a geriatric mental healthservices research agenda. Unique challenges for trans-lational research71 are posed by frailty, cognitive im-pairment, knowledge deficits, social isolation, limitedfinancial resources, and high prevalence of medical andpsychiatric comorbidities that may alter adherence andresponse to treatments. Moreover, interventions effec-tive for clinical problems concentrated in geriatric pa-tient populations18,72,73 often have not been tested inroutine outpatient practice settings and do not containdissemination protocols designed to reach “average” pa-tients in “average” care situations. The challenge for pa-tient-centered health services research is to ensure theinclusion of broad and appropriate subject samples inintervention studies, to demonstrate limits to general-izability where they occur, and to point to modificationsrequired for specific patient populations.

Examining Treatment Preferences and MentalHealth Care Decision-Making

Many questions about treatment preferences andmedical decision-making by patients, their families, and

physicians remain unstudied for most geriatric mentalhealth problems and are important issues for externalvalidation of interventions. Factors that facilitate (or ob-struct) acceptance of effective treatments by patientsand providers need to be identified and addressed. Neg-ative outcomes—poor adherence, poor clinical results,deteriorating general health, and declining quality oflife—are consequences of neglecting these dimensions.Also, factors that impede effective collaboration be-tween patient/family and provider must be examinedwith a particular focus on late-life mental disorders as-sociated with comorbid illnesses, where appropriatedecision-support mechanisms for both mental andmedical disorders are more complex to design and im-plement in treatment settings. This type of health ser-vices research marries the widely diverse elements oftechnology assessment and quality-assurance method-ologies with clinician/patient reasoning, health careutilities, and decision-making. Studies in this area shouldidentify, explain, and build on practice variations fromestablished recommendations and guidelines; test theimpact of geropsychiatric care managers on initiation,adherence, and continuity of care across care settings;and evaluate the costs of ensuring guideline-level carein health care systems.

Using Qualitative Research Designs

Qualitative methods that call upon the expertise ofdisciplines distant from the clinical arena and health ser-vices research per se, such as medical sociology andanthropology, have been neglected in geriatric mentalhealth research. Experiences and values that affect help-seeking, acceptance of and adherence to treatment,provider behavior, and payer policies for geriatric men-tal health care may have clinical, attitudinal, financial,geographic, and/or social and cultural sources. Quali-tative research aims to understand these sources ofhealth care disparities to improve the design of incen-tives and practical strategies for change. Qualitativestudies are time- and labor-intensive and usually limitedto small samples, but they can provide useful pilot datafor the design of intervention trials. Such exploratorymethods may be required in order to find successfulways of overcoming barriers to effective treatment, in-cluding cognitive biases among providers, patients, andfamilies; these exploratory methods may include mentalhealth utilities that examine the relative values of alter-native outcomes; clinical decision trees that incorporate



the real probability of specific treatment outcomes; andfocus groups with targeted community and providergroups.

Identifying and OvercomingDiscontinuities in Care

Barriers to access and utilization of appropriatemental health services may be more difficult to sur-mount for older than for younger adults,74 and there isevidence that ethnicity is a factor.75 Segregation of typesof care into different treatment sites and settings accen-tuates problems of stigma and access, impedes accep-tance of mental health care as a component of generalhealth care, and may reduce adherence with referralsto mental health service providers;12 older persons aremost vulnerable to the effects of such discontinuities incare. Moreover, communities differ in the complementand organization of systems of care available for olderadults and in how effectively information is dissemi-nated into the communities these systems serve. Thedevelopment and dissemination of mental health out-come measures and quality indicators that are specifi-cally designed for older patients, their providers, andother caregivers, and that are capable of spanning di-vergent systems of care, is an important focus for newhealth services research.

Understanding the Structure andBehavior of Health Systems

Analysis of existing structural, organizational, andfinancial arrangements for the delivery of mental healthservices for elderly patients is a necessary step towardtesting the effectiveness of treatments validated in effi-cacy research.1 We must understand how patients entera given treatment setting and access mental health ser-vices, what service components are offered, and howthey are paid for if proposed best practices are to beimplemented where they can do the most good. Con-ceptually, research in this area examines questions atthe interface of health systems organization and clinicalcare and may address the effect of changing financialarrangements on access or outcomes for targeted pop-ulations.

Knowledge is sparse regarding the outcomesachieved by different models of financing and deliveringmental health services to older adults under differentenrollment and payment schemes,56 and available stud-

ies often raise more questions than they answer. Forexample, a recent report on more than 80,000 olderenrollees in a managed Medicare plan found that ratesof diagnosed dementia were far lower (�1%) than inepidemiological studies and that dementia increasedthe cost of inpatient but not outpatient care.76 The datapermitted no conclusions as to whether this low de-mentia prevalence was the result of enrollment exclu-sion or perverse incentives favoring nondiagnosis ornonreporting, but suggested that outpatient care wasprobably inadequate. Preliminary program and paymentstandards for managed Medicare plans enrolling frail el-derly patients have been published.77 For the majorityof older adults not currently enrolled in managed-careplans, little is known about the impact of site of service,availability of supplemental insurance (including pre-scription drug benefits),78 and accessibility of mentalhealth services within the context of general health careand social services.

Crucial issues relevant to all forms of practice, in-cluding fee-for-service as well as managed care, havebeen raised by the growing pressures toward increasing“management” of geriatric care. Particularly importantquestions include whether mental health services arebest organized and administered via “carve-in” vs.“carve-out” payment designs or via integrated vs. seg-regated mental and general health services.79 Managed-care approaches highlight the need to define the mean-ing of “medically necessary” mental health care forolder adults and how access to necessary care can beensured for complex, long-term mental disorders suchas schizophrenia or chronic depression42 and demen-tia.80 Other important targets are the impact of cost-containment strategies on clinical and functional out-comes and the role of financial-incentive structures infacilitating or impeding care. Also, opportunities for im-proving care through research on mental health serviceneeds and delivery models in managed long-term care(such as social HMOs, PACE programs, and managedMedicaid) are both timely and likely to make significantcontributions to conceptualizing care for older personsin general.42,46 Models of financing and delivering ser-vices to older adults that have been adopted abroad,particularly those that combine medical and social mod-els in a single continuum of care,45 offer useful infor-mation for health policy in the United States. Finally,satisfaction with care is an important dimension of re-search on the effectiveness of alternative organizationaland financing schemes. This has been approached

Borson et al.


through comparative study of managed-care enrollees,those who have left a managed-care plan, and recipientsof fee-for-service care.81

Improving Established Health andAdministrative Databases

Administrative databases offer unique opportuni-ties for research on geriatric mental health care.82–84

However, most do not contain adequate mental healthmeasures in a form or depth that can be interpretedusing contemporary concepts of mental disorders. Min-imal mental health measures have been incorporatedinto the National Nursing Home Study, and Medicareclaims data do not provide reliable estimates of theprevalence and treatment of some mental disorders, inpart because of financial and other disincentives for re-porting.76,85 On the other hand, the Minimum Data Setfor nursing homes may be a reasonably reliable indica-tor of cognitive and behavioral problems in residents.86

Policy initiatives are needed to incorporate robust men-tal health indicators into all large databases that dealwith the health care of older adults, and claims-codingand storage requirements must be revised to include alldiagnoses and all forms of mental health services re-gardless of provider type or setting. These changes willgreatly enhance the usefulness of administrative data-bases for geriatric mental health research by bringingcoding of diagnoses and services into register with ac-tual clinical practice.

MECHANISMS FOR IMPLEMENTING AGERIATRIC MENTAL HEALTH SERVICE

RESEARCH AGENDA

Leveraging Existing Human andData Resources for Rapid Progress

Despite their current limitations for mental healthservices research, large databases maintained by Medi-care, Medicaid, pharmaceutical and nursing home con-sortia, and managed-care organizations offer opportu-nities for rapid progress in the generation of newknowledge, provided access and use are supported bythe necessary technical and human infrastructure.Moreover, the emergence of a small cadre of experi-enced senior investigators as mentors for the next gen-eration now makes geriatric mental health services re-

search a viable career path for young investigators. Newprograms are needed to train junior investigators in re-search methods and to facilitate transitions of clinicalresearch programs into resources capable of studyingservices questions. Nontraditional funding mechanismswill be required to support the interdisciplinary, cross-program collaboration that is essential to the conductof geriatric mental health services research, to attractthe expertise needed to carry out specific projects, andto expand investigative capacity beyond those few cen-ters having their own on-site experts.

Increasing Research Funding for GeriatricMental Health Services Within NIH

About 8 percent of the current NIMH extramuralbudget supports research on geriatric mental health dis-orders and problems (B. Lebowitz, personal communi-cation), a disproportionately small percentage relativeto that of population aging trends and health care costs(Figure 1). Although tracking systems do not currentlyidentify mental health services as distinct from clinicalresearch grants, it appears that very few funded grantsin geriatric mental health explicitly address services re-search dimensions. Enhancing the aging focus of thehealth services branch of NIMH should be a priority forcross-branch and cross-institute collaborative efforts.

Increasing the Supply of New Investigators

Systematic development of a pool of investigatorswell trained for mental health services research in el-derly populations is a sine qua non for evolving an ef-fective geriatric mental health policy for the coming de-cades. The numbers of new investigators needed tomeet the challenge of the growing crisis in mentalhealth care for the elderly population cannot be deter-mined with precision, but can be framed in geographicterms. National variations in practice and reimburse-ment patterns highlight the need for regional centers ofexcellence in mental health services research and train-ing. Such centers have the potential for individualizingthe focus of research and training to the needs of thepopulation; local practice styles; and unique sociopolit-ical, cultural, and economic environments. However,the fundamental competencies required to translate ad-vances in detection, differential diagnosis, and treat-ment into practice in health care systems are not re-stricted by locale, nor are the skills needed to apply



findings from health services research to new clinicalquestions. These skills are not found in conventionalclinical research training programs and must be nur-tured through a national plan spearheaded by NIMH.

Core competencies. Core competencies for health ser-vices research are developed through a set of linkedcareer-development activities that extend beyond theconfines of the academic medical center. Investigatorsneed formal experience at the interface between mentalhealth care and primary and specialty medical care andnonclinical service settings—in policy and economic in-stitutes, public health and community agencies, state-sponsored programs on aging and mental health, largehealth care organizations, nursing home consortia, ornetworks of senior centers and adult daycare programs.Research training should stress the acquisition of tech-nical skills for designing, implementing, and evaluatingalternative models of care that include family- and con-sumer-based interventions as well as large-scale descrip-tive studies of health care systems. Training includesspecial methods for conducting randomized services tri-als, testing the impact of service designs on patient andsystem outcomes, practice-based or dissemination re-search, and large-scale studies of service systems. It isimportant to attain familiarity with alternative paymentsystems and the ethical and practical dimensions of dis-tributing costs across multiple payer sources. Finally, ex-posure to methods and pathways by which the findingsof population-based outcome studies can be translatedinto testable treatment guidelines and health care poli-cies may require special training opportunities. Thesemay reach outside the field of academic geriatric mentalhealth to national organizations dedicated to this policydimension, such as the Association of Health ServicesResearchers (AHSR).

ORGANIZING AND FUNDING AFUNCTIONAL GERIATRIC MENTAL

HEALTH SERVICES RESEARCH ANDTRAINING PROGRAM

The number of new young investigators intending toprepare or currently training for careers in geriatricmental health services research is not known, nor is thenumber of capable senior investigators who could be-come resources in the field through development of

new competencies. Approaches to enhancing the poolof scientists qualified in geriatric mental health servicesresearch should include alternative and complementarymechanisms that recognize the need for dual compe-tency in a clinical geriatric mental health discipline andmental health services research methods. Fundingmechanisms that could integrate geriatric clinical andmental health services research may include supple-mental support for currently funded research centers tocarry out pilot services studies, transitional grants toenable established clinical or services centers to inte-grate investigators with expertise in the other disci-pline, infrastructure grants to enable programs to de-velop a critical mass of mental health servicesinvestigators, and training grants to foster the develop-ment of investigators with expertise in both geriatricclinical and mental health services research. Specificmechanisms that should be considered include

• Geriatric mental health research supplements for ex-isting centers of excellence in general mental healthservices research

• Mental health services research supplements forcenters of excellence in geriatric clinical investiga-tion to support the development of interdisciplinarytraining programs

• Request for proposals (RFPs) jointly sponsored byNIMH and the National Institute on Aging (NIA) tocreate developmental cores in mental health ser-vices research and training in Alzheimer’s diseaseresearch centers

• RFPs for T-32 and other training grant mechanismsto support postdoctoral fellowships in geriatricmental health services research

• Very early career preparation through enhancedfunding for interdisciplinary research training, dur-ing the psychiatry and psychology residency yearsand predoctoral years, in public and communityhealth, health care economics, and health care ad-ministration

• Interdisciplinary, inter-institute research traininggrants cosponsored by NIMH, NIA, and the specialtyinstitutes (e.g., the National Institute of NeurologicalDiseases and Stroke [NINDS], the National Heart,Lung, and Blood Institute [NHLBI], the National In-stitute of Diabetes and Digestive and Kidney Disease[NIDDK], and the National Institute of Nursing Re-search [NINR]) and federal administrations andagencies (e.g., SAMHSA, the Agency for Healthcare

Borson et al.


Research and Quality [AHRQ], the Health Resourcesand Services Administration [HRSA], and the HealthCare Financing Administration [HCFA]) for the de-velopment of multidisciplinary collaborative modelsof care

• Partnerships between federal research funding agen-cies and national foundations (e.g., the Alzheimer’sAssociation, MacArthur Foundation, and HartfordFoundation) for support of research training at thejunior and senior levels

• Creation of Collaborative Infrastructure SupportPrograms (CISPs) that facilitate the development ofcross-center “virtual” research programs, to maxi-mize opportunities for developing a critical mass ofinvestigators and data

SUMMARY

A pressing need exists for consolidating a mental healthservices research agenda designed specifically for an ag-ing population. Priority content areas in this servicesresearch agenda include a focus on high-risk elderlypeople (older persons with comorbid disorders, theoldest and frail elderly population, and older adults withchronic mental disorders); horizontally and vertically in-tegrated models of mental health services across thespectrum of long-term care and primary and specialtycare; translation of research findings and evidence-based practices into clinical settings; research on deci-sion-making; and systems-based health services re-search. This paper elaborates upon the rationale andpresents a detailed proposal for strategic developmentof the field. It highlights research and training targetsand proposes new funding mechanisms to support such

development. Novel interdisciplinary and interinstitu-tional mechanisms for promoting collaboration are pro-posed as a means of building upon existing nationalmental health research resources. This multifacetedstrategy will create the intellectual and organizationalinfrastructure necessary to meet the growing mentalhealth service needs of older adults in the decades to

come.

Valuable consultation was provided to the AAGP

Health Services Workgroup by Jerome Avorn, M.D.,

Chief, Division of Pharmacoepidemiology and Phar-

macoeconomics, Brigham and Women’s Hospital, and

Associate Professor, Harvard Medical School, Boston,

MA; David Blumenthal, M.D., Chief, Health Policy Re-

search and Development Unit, Massachusetts General

Hospital, and Professor, Harvard Medical School; Rich-

ard G. Frank, Ph.D., Professor of Health Economics,

Department of Health Care Policy, Harvard Medical

School; Sue Levkoff, Sc.D., Associate Professor, Division

on Aging, Harvard Medical School; Michael Smyer,

Ph.D., Dean of Graduate Studies and Associate Vice

President for Research, Boston College; and Philip

Wang, M.D., Program for the Analysis of Clinical Strat-

egies, Division of Pharmacoepidemiology and Phar-

macoeconomics, Brigham and Women’s Hospital, and

Instructor, Harvard Medical School. Dilip Jeste, M.D.,

and David Greenspan, M.D., provided helpful com-

ments on an earlier draft, and the Board of Directors

of the American Association for Geriatric Psychiatry

approved the content of this report. Expert adminis-

trative and secretarial assistance was provided by

Donna Cutillo, Nancy Monahan, and Sherry Neher.

Travel expenses for workgroup participants were

supported in part by an unrestricted educational

grant from Pfizer Pharmaceuticals.

References

1. National Advisory Mental Health Council: Bridging Science andService. #99-4353, pp 1–64

2. U.S. Public Health Service: Mental Health: A Report of the Sur-geon General. Rockville, MD, 1999

3. Jeste DV, Alexopoulos GS, Bartels SJ, et al: Consensus Statementon the Upcoming Crisis in Geriatric Mental Health: researchagenda for the next two decades. Arch Gen Psychiatry 1999;56:848–853

4. Edmonds DE: The unit approach to the rehabilitation and reset-tlement of the long-stay psychiatric patient. Occupational Health(London, UK) 1966; 18:125–133

5. Mattes JA, Rosen B, Klein DF: Comparison of the clinical effec-tiveness of “short-” vs. “long-”stay psychiatric hospitalization, II:

results of a 3-year post-hospital follow-up. J Nerv Ment Dis 1977;165:387–394

6. Goldstrom ID, Burns BJ, Kessler LG, et al: Mental health servicesuse by elderly adults in a primary care setting. J Gerontol 1987;42:147–153

7. Burns BJ, Taube CA: Mental health services in general medicalcare and in nursing homes, in Mental Health Policy for OlderAmericans: Protecting Minds at Risk. Edited by Fogel B, FurinoA, Gottlieb GL. Washington, DC, American Psychiatric Press,1990, pp 63–84

8. Katz IR, Streim J, Parmelee P: Psychiatric-medical comorbidity:implications for health services delivery and for research on de-pression. Biol Psychiatry 1994; 36:141–145

9. Borson S, Unutzer J: Psychiatric problems in the medically ill



elderly, in Comprehensive Textbook of Psychiatry. Edited by Kap-lan HI, Sadock BJ. Baltimore, MD, Williams & Wilkins, 2000, pp3045–3053

10. Davidson M, Harvey P, Welsh KA, et al: Cognitive functioning inlate-life schizophrenia: a comparison of elderly schizophrenic pa-tients and patients with Alzheimer’s disease. Am J Psychiatry1996; 153:1274–1279

11. Bartels SJ, Liberto J: Dual diagnosis in the elderly, in Double Jeop-ardy. Edited by Lehman AF, Dixon LB. New York, Harwood Aca-demic Publishers, 1996, pp 139–157

12. George LK: Community and home care for mentally ill olderadults, in Handbook of Mental Health and Aging. Edited by BirrenJE, Sloane RB, Cohen GD. San Diego, CA, Academic Press, 1992

13. Cohen D, Cairl R: Mental health care policy in an aging society,in Mental Health Services: A Public Perspective. Edited by LevinBL, Petrila J. New York, Oxford University Press, 1996, pp 301–319

14. Knight BG, Kaskie B: Models for mental health service deliveryto older adults, in Emerging Issues in Mental Health and Aging.Edited by Gatz M. Washington, DC, American Psychological As-sociation, 1995, pp 231–255

15. Kales HC, Blow FC, Copeland LA, et al: Health care utilization byolder patients with coexisting dementia and depression. Am JPsychiatry 1999; 156:550–556

16. Borson S, Barnes RA, Kukull WA, et al: Symptomatic depressionin elderly medical outpatients, I: prevalence, demography, andhealth service utilization. J Am Geriatr Soc 1986; 34:341–347

17. Saravay SM, Pollack S, Steinberg MD, et al: Four-year follow-up ofthe influence of psychological comorbidity on medical rehospi-talization. Am J Psychiatry 1996; 153:397–403

18. Mittelman M, Ferris SH, Shulman E, et al: A family interventionto delay nursing home placement of patients with Alzheimer dis-ease: a randomized controlled trial. JAMA 1996; 276:1725–1731

19. Brown I, Renwick R, Raphael D: Frailty: constructing a commonmeaning, definition, and conceptual framework. Int J Rehabil Res1995; 18:93–102

20. Chin APM, Dekker JM, Feskens EJ, et al: How to select a frailelderly population: a comparison of three working definitions. JClin Epidemiol 1999; 52:1015–1021

21. Hamerman D: Toward an understanding of frailty. Ann InternMed 1999; 130:945–950

22. “Down and dirty” medical information system identifies high-riskpatients. Data Strategic Benchmarks 1998; 2:186–187

23. Bula CJ, Berod AC, Stuck AE, et al: Effectiveness of preventive in-home geriatric assessment in well-functioning,community-dwell-ing older people: secondary analysis of a randomized trial. J AmGeriatr Soc 1999; 47:389–395

24. Boult C, Pacala JT, Boult LB: Targeting elders for geriatric evalu-ation and management: reliability, validity, and practicality of aquestionnaire. Aging (Milan, Italy) 1995; 7:159–164

25. Leveille SG, Wagner EH, Davis C, et al: Preventing disability andmanaging chronic illness in frail older adults: a randomized trialof a community-based partnership with primary care. J Am Ger-iatr Soc 1998; 46:1191–1198

26. Coleman EA, Wagner EH, Grothaus LC, et al: Predicting hospi-talization and functional decline in older health plan enrollees:are administrative data as accurate as self-report? J Am GeriatrSoc 1998; 46:419–425

27. Morishita L, Boult C, Boult L, et al: Satisfaction with outpatientgeriatric evaluation and management (GEM). Gerontologist1998;38:303–308

28. Buchner DM, Wagner EH: Preventing frail health. Clin GeriatrMed 1992; 8:1–17

29. Banerjee S, Shamash K, Macdonald AJ, et al: Randomised con-trolled trial of effect of intervention by psychogeriatric team ondepression in frail elderly people at home. BMJ 1996; 313:1058–1061

30. Morris JN, Sherwood S, Gutkin CE: Inst-Risk II: an approach toforecasting relative risk of future institutional placement. HealthServ Res 1988; 23:511–536

31. Abraham IL, Currie LJ, Neese JB, et al: Risk profiles for nursinghome placement of rural elderly: a cluster analysis of psychoge-riatric indicators. Arch Psychiatr Nurs 1994; 8:262–271

32. Colenda CC, Greenwald BS, Crossett JH, et al: Barriers to effec-tive psychiatric emergency services for elderly persons. PsychiatrServ 1997; 48:321–325

33. Patterson JA, Bierman AS, Splaine M, et al: The population ofpeople age 80 and older: a sentinel group for understanding thefuture of health care in the United States. Journal of AmbulatoryCare Management 1998; 21:10–16

34. Bierman AS, Magari ES, Jette AM, et al: Assessing access as a firststep toward improving the quality of care for very old adults.Journal of Ambulatory Care Management 1998; 21:17–26

35. Splaine M, Bierman AS, Wasson JH: Implementing a strategy forimproving care: lessons from studying those age 80 and older ina health system. Journal of Ambulatory Care Management 1998;21:56–59

36. Cuffel BJ, Jeste DV, Halpain M, et al: Treatment costs and use ofcommunity mental health services for schizophrenia by age co-horts. Am J Psychiatry 1996; 153:870–876

37. Bartels SJ, Forester B, Miles KM, et al: Mental health service useby elderly patients with bipolar disorder and unipolar major de-pression. Am J Geriatr Psychiatry 2000; 8:160–166

38. Aiken L: Chronic mental illness, in Mental Health Policy for OlderAmericans: Protecting Minds at Risk. Edited by Fogel BS, FurinoA, Gottlieb G. Washington, DC, American Psychiatric Press, 1990,pp 239–256

39. Burns BJ: Mental health services research on the hospitalized andinstitutionalized chronically mentally ill elderly, in The ElderlyWith Chronic Mental Illness. Edited by Lebowitz BD, Light E.New York, Springer, 1991, pp 207–215

40. Light E, Lebowitz BD: The elderly with chronic mental illness.New York, Springer, 1991

41. Lacro JP, Jeste DV: Physical comorbidity and polypharmacy inolder psychiatric patients. Biol Psychiatry 1994; 36:146–152

42. Bartels SJ, Levine KJ, Shea D: Community-based long-term carefor older persons with severe and persistent mental illness in anera of managed care. Psychiatr Serv 1999; 50:1189–1197

43. McCulloch DK, Price MJ, Hindmarsh M, et al: A population-basedapproach to diabetes management in a primary care setting: earlyresults and lessons learned. Effective Clinical Practice 1998;1:12–22

44. Katon W, Von Korff M, Lin E, et al: Population-based care of de-pression: effective disease management strategies to decreaseprevalence. Gen Hosp Psychiatry 1997; 19:169–178

45. Stolee P, Kessler L, Le Clair JK: A community development andoutreach program in geriatric mental health: four years’ experi-ence. J Am Geriatr Soc 1996; 44:314–320

46. Eng C, Pedulla J, Eleazer GP, et al: Program of All-inclusive Carefor the Elderly (PACE): an innovative model of integrated geriatriccare and financing. J Am Geriatr Soc 1997; 45:223–232

47. Weissert WG, Cready CM: Toward a model for improved targetingof aged at risk of institutionalization. Health Serv Res 1989;24:485–510

48. Weissert WG, Hedrick SC: Lessons learned from research on ef-

Borson et al.


fects of community-based long-term care. J Am Geriatr Soc 1994;42:348–353

49. Reifler BV, Cox NJ, Jones BN, et al: Service use and financial per-formance in a replication program on adult day centers. Am JGeriatr Psychiatry 1999; 7:98–109

50. Rovner BW, Steele CD, Shmuely Y, et al: A randomized trial ofdementia care in nursing homes. J Am Geriatr Soc 1996; 44:7–13

51. Stump TE, Johnson RJ, Wolinsky FD: Changes in physician utili-zation over time among older adults. J Gerontol B Psychol SciSoc Sci 1995; 50:S45–S58

52. Kim E, Rovner B: The nursing home as a psychiatric hospital, inPsychiatric Care in the Nursing Home. Edited by Reichman WE,Katz PR. New York, Oxford University Press, 1996, pp 3–9

53. Colenda CC, Streim J, Greene JA, et al: The impact of OBRA ’87on psychiatric services in nursing homes: joint testimony of theAmerican Psychiatric Association and the American Associationfor Geriatric Psychiatry. Am J Geriatr Psychiatry 1999; 7:12–17

54. Office of the Inspector General (OIG): Mental Health Services inNursing Facilities. Washington, DC, U.S. Dept. of Heath and Hu-man Services, 1996

55. Burns BJ, Wagner R, Taube JE: Mental health service use by theelderly in nursing homes. Am J Public Health 1993; 83:331–337

56. Shea DG, Streit A, Smyer MA: Determinants of the use of spe-cialist mental health services by nursing home residents. HealthServ Res 1994; 29:169–185

57. Smyer MA, Shea DG, Streit A: The provision and use of mentalhealth services in nursing homes: results from the NationalMedical Expenditure Survey. Am J Public Health 1994; 84:284–287

58. Borson S, Loebel JP, Kitchell M, et al: Psychiatric assessments ofnursing home residents under OBRA ’87: should PASARR be re-formed? pre-admission screening and annual review. J Am GeriatrSoc 1997; 45:1173–1181

59. Regier DA, Goldberg ID, Taube CA: The de facto U.S. mentalhealth services system: a public health perspective. Arch GenPsychiatry 1978; 35:685–693

60. Colenda CC: Essential curriculum in geriatric psychiatry for gen-eral internal medicine residency and geriatric medicine fellow-ship programs. Am J Med 1994; 97:15S–18S

61. Unutzer J, Patrick DL, Simon G, et al: Depressive symptoms andthe cost of health services in HMO patients aged 65 years andolder: a 4-year prospective study. JAMA 1997; 277:1618–1623

62. Katon W, Von Korff M, Lin E, et al: Collaborative management toachieve treatment guidelines: impact on depression in primarycare. JAMA 1995; 273:1026–1031

63. Mechanic D: Approaches for coordinating primary and specialtycare for persons with mental illness. Gen Hosp Psychiatry 1997;19:395–402

64. Borson S, Curtis JR, Schmaling KB: Disorders of mood and cog-nition in patients with chronic obstructive pulmonary disease:what we know and what we need to do. National Institutes ofMental Health Transcripts, NIH Research Initiative, 2000

65. Goldberg HI, Wagner EH, Fihn SD, et al: A randomized controlledtrial of CQI teams and academic detailing: can they alter com-pliance with guidelines? Jt Comm J Qual Improv 1998; 24:130–142

66. Lin EH, Katon WJ, Simon GE, et al: Achieving guidelines for thetreatment of depression in primary care: is physician educationenough? Med Care 1997; 35:831–842

67. Davis D: Does CME work? an analysis of the effect of educationalactivities on physician performance or health care outcomes. IntJ Psychiatry Med 1998; 28:21–39

68. Avorn J, Soumerai SB, Everitt DE, et al: A randomized trial of aprogram to reduce the use of psychoactive drugs in nursinghomes. N Engl J Med 1992; 327:168–173

69. Soumerai SB: Principles and uses of academic detailing to im-prove the management of psychiatric disorders. Int J PsychiatryMed 1998; 28:81–96

70. Lehman AF, Steinwachs DM: Translating research into practice:the schizophrenia Patient Outcomes Research Team (PORT)treatment recommendations. Schizophr Bull 1998; 24:1–10

71. Unutzer J, Katon W, Sullivan M, et al: Treating depressed olderadults in primary care: narrowing the gap between efficacy andeffectiveness. Milbank Q 1999; 77:225–256, 174

72. Reynolds CF, Frank E, Perel JM, et al: Nortriptyline and interper-sonal psychotherapy as maintenance therapies for recurrent ma-jor depression: a randomized controlled trial in patients olderthan 59 years. JAMA 1999; 281:39–45

73. Teri L, Logsdon RG, Uomoto J, et al: Behavioral treatment of de-pression in dementia patients: a controlled clinical trial. J Ger-ontol B Psychol Sci Soc Sci 1997; 52:159–166

74. Estes CL: Mental health services for the elderly: key policy ele-ments, in Emerging Issues in Mental Health and Aging. Edited byGatz M. Washington, DC, American Psychological Association,1995, pp 303–328

75. Cooper-Patrick L, Gallo JJ, Powe NR, et al: Mental health serviceutilization by African Americans and whites: The Baltimore Epi-demiologic Catchment Area Follow-Up. Med Care 1999; 37:1034–1045

76. Gutterman EM, Markowitz JS, Lewis B, et al: Cost of Alzheimer’sdisease and related dementia in managed-Medicare. J Am GeriatrSoc 1999; 47:1065–1071

77. Medicare Payment Advisory Commission Report to Congress: Se-lected Medicare Issues: Chapter 5: Managed care for frail Medi-care beneficiaries: payment method and program standards,1999, pp 79–104

78. Soumerai SB, Ross-Degnan D: Inadequate prescription-drug cov-erage for Medicare enrollees: a call to action. N Engl J Med 1999;340:722–728

79. Wells KB, Astrachan BM, Tischler GL, et al: Issues and approachesin evaluating managed mental health care. Milbank Q 1995;73:57–75

80. Bartels SJ, Colenda CC: Mental health services for Alzheimer’sdisease: current trends in reimbursement and public policy, andthe future under managed care. Am J Geriatr Psychiatry 1998;6:S85–100

81. Newcomer R, Harrington C, Preston S: Satisfaction in the social/health maintenance organization: a comparison of members, di-senrollees, and those in fee-for-service, in HMOs and the Elderly.Edited by Luft H. Ann Arbor, MI, Administration Press, 1994, pp111–139

82. Cano C, Hennessy KD, Warren JL, et al: Medicare Part A utiliza-tion and expenditures for psychiatric services: 1995. Health CareFinance Review 1997; 18:177–193

83. Rosenbach ML, Ammering CJ: Trends in Medicare Part B mentalhealth utilization and expenditures: 1987–92. Health Care Fi-nance Review 1997; 18:19–42

84. Larson MJ, Farrelly MC, Hodgkin D, et al: Payments and use ofservices for mental health, alcohol, and other drug abuse dis-orders: estimates from Medicare, Medicaid, and private healthplans, in Mental Health United States, 1998. Edited by Mander-scheid RW, Henderson MJ. Washington, DC, Dept. of Healthand Human Services Publication: (SMA) 99-3285, 1998, pp 124–141

85. Newcomer R, Clay T, Luxenberg JS, et al: Misclassification and



selection bias when identifying Alzheimer’s disease solely fromMedicare claims records. J Am Geriatr Soc 1999; 47:215–219

86. Snowden M, McCormick W, Russo J, et al: Validity and respon-siveness of the Minimum Data Set. J Am Geriatr Soc 1999;47:1000–1004

87. Malmgren R: Epidemiology of aging, in The American PsychiatricPress Textbook of Geriatric Neuropsychiatry. Edited by CoffeyCE, Cummings JL. Washington, DC, American Psychiatric Press,1999, pp 17–33

Geriatric mental health services research: Strategic Plan for an Aging Population: Report of the...

Documents

Transcript of Geriatric mental health services research: Strategic Plan for an Aging Population: Report of the...