Evidence-Based Practices in Geriatric Mental Health Care

Home , Geriatric psychiatry

Evidence-Based Practices Focusing on Evidence- in Geriatric Mental Health Care Based Practices Stephen J. Bartels, M.D. Aricca R. Dums, B.A. Thomas E. Oxman, M.D. Lon S. Schneider, M.D. Patricia A. Areán, Ph.D. George S. Alexopoulos, M.D. Dilip V. Jeste, M.D.

The past decade has seen dramatic growth in research on treatments for reatment of mental disorders the psychiatric problems of older adults. An emerging evidence base among older Americans has supports the efficacy of geriatric mental health interventions. The au- Tbecome a major public health thors provide an overview of the evidence base for clinical practice. need. The number of people over the They identified three sources of evidence—evidence-based reviews, age of 65 with psychiatric disorders meta-analyses, and expert consensus statements—on established and will more than double by the year emerging interventions for the most common disorders of late life, 2030, from 7 million in 2000 to 15 which include depression, dementia, substance abuse, schizophrenia, million (1). The past decade has seen and anxiety. The most extensive research support was found for the ef- dramatic growth in research on the fectiveness of pharmacological and psychosocial interventions for geri- causes and treatments of the psychiatric major depression and for dementia. Less is known about the ef- atric problems of older adults. fectiveness of treatments for the other disorders, although emerging ev- In this article we provide an idence is promising for selected interventions. Empirical support was overview of empirically validated also found for the effectiveness of community-based, multidisciplinary, treatments as reflected in systematic geriatric psychiatry treatment teams. The authors discuss barriers to reviews of the literature on geriatric implementing evidence-based practices in the mental health service de- mental health interventions. Three livery system for older adults. They describe approaches to overcoming types of evaluations of the literature these barriers that are based on the findings of research on practice on major geriatric mental health dis- change and dissemination. Successful approaches to implementing orders are summarized: systematic ev- change in the practices of providers emphasize moving beyond tradi- idence-based-practice reviews, meta- tional models of continuing medical education to include educational analytic studies, and expert consensus techniques that actively involve the learner, as well as systems change statements. Next we summarize ma- interventions such as integrated care management, implementation jor barriers to the dissemination and toolkits, automated reminders, and decision support technologies. The implementation of these practices. anticipated growth in the population of older persons with mental dis- Finally, we describe possible strate- orders underscores the need for a strategy to facilitate the systematic gies for disseminating and imple- and effective implementation of evidence-based practices in geriatric menting evidence-based practices in mental health care. (Psychiatric Services 53:1419–1431, 2002) geriatric mental health care.

An impending public health crisis Dr. Bartels and Dr. Oxman are affiliated with the department of psychiatry at Dartmouth At least one in five people over the Medical School in Hanover, New Hampshire. They and Ms. Dums are affiliated with the age of 65 suffers from a mental disor- New Hampshire–Dartmouth Psychiatric Research Center in Lebanon, New Hampshire. der (1). By 2030 the number of per- Dr. Schneider is with the department of psychiatry and behavioral sciences at the Keck sons with psychiatric disorders in this School of Medicine at the University of Southern California, Los Angeles. Dr. Areán is older group will equal or exceed the with the department of psychiatry at the University of California, San Francisco. Dr. Alexopoulos is with the department of psychiatry at the Weill Medical College of Cornell number with such disorders in University in White Plains, New York. Dr. Jeste is affiliated with the department of psy- younger age groups (age 18 to 29 or chiatry at the University of California, San Diego, and the San Diego Veterans Affairs age 30 to 44) (1). Despite the growing Medical Center. Send correspondence to Dr. Bartels at the New Hampshire–Dartmouth requirement for mental health servic- Psychiatric Research Center, 2 Whipple Place, Suite 202, Lebanon, New Hampshire es for older persons, there is substan- 03766 (e-mail, [email protected]). tial unmet need. The 1999 Surgeon

PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1419 General’s report on mental health (2), lished evidence-based reviews and of antidepressant studies that includ- the Administration on Aging’s 2001 meta-analyses is not intended to be an ed all age groups found that antide- report (3), and an expert consensus exhaustive summary of the research pressants offer only a 20 percent (2 statement (1) underscore the need to literature but rather a starting point points) greater reduction in scores on plan for the provision of services for that defines geriatric mental health the Hamilton Rating Scale for De- the growing number of elderly per- treatments with proven effectiveness. pression compared with placebo. This sons with major mental disorders. English-language review articles analysis suggests that placebo med- Older adults with mental disorders that examined the effectiveness of ication combined with visits by the are more likely than younger adults to geriatric mental health services were prescribing physician may account for receive inappropriate or inadequate identified for the most common psy- 80 percent of the effect of antidepres- treatments (4). Bridging the gap be- chiatric problems among older adults: sants (23). In addition, the compara- tween research and clinical services depression, dementia, alcohol abuse, tive efficacy and tolerability of differ- has been identified as one of the most schizophrenia, and anxiety disorders ent classes of antidepressants remain important priorities in health care (10) through searches of MEDLINE, controversial. Meta-analyses have not (5,6). Among the greatest challenges PsycINFO, and the Cochrane Li- shown significant differences be- is the “expertise gap” that affects cli- brary. Disorders for which we were tween the selective serotonin reup- nicians practicing in routine settings. unable to identify evidence-based re- take inhibitors (SSRIs) and the older This gap is the result of inadequate views, meta-analyses, or consensus tricyclic agents in terms of efficacy or training in geriatric care and a failure statements specifically targeting older treatment dropout from adverse ef- to incorporate contemporary re- adults, such as bipolar disorder, were fects. In contrast, expert consensus search findings and evidence-based not considered. statements recommend SSRIs as practices into usual care. Searches were conducted of articles first-line agents for geriatric depres- published through the year 2001, in- sion and suggest avoiding tertiary Evidence-based practices cluding but not limited to the terms amine antidepressants, such as Many of the underlying principles of evidence-based review, meta-analysis, amitriptyline, imipramine, and dox- evidence-based practice reflect consensus statement, consensus re- epin, because of the serious side ef- Cochrane’s (7) assertion three view, and review article. The evi- fects, including cardiovascular and decades ago that our limited health dence-based reviews included were anticholinergic side effects, associat- care resources should be applied to those that systematically categorized ed with their use (18–22). providing interventions that have studies and applied strict criteria for Although the meta-analyses did not proven effectiveness based on well- rating the level of evidence. The find a difference in tolerability be- designed evaluation trials, with em- meta-analyses included were those tween SSRIs and tricyclics on the ba- phasis on randomized controlled tri- that described and applied standard- sis of rates of discontinuation due to als. In this respect, evidence-based ized meta-analytic statistical proce- side effects, clinically significant dif- practice draws heavily on the use of dures. Expert consensus statements ferences may still be present. For ex- external evidence to support clinical were included that described a sys- ample, common reasons for discon- judgment (8). Criteria for evidence- tematic method of obtaining, evaluat- tinuing SSRIs include sleep distur- based practices define different levels ing, and summarizing expert consen- bance, gastrointestinal distress, anxi- of empirical support based on the sus opinion on effective treatments. ety, headaches, and weight loss, quality of the data (8,9). The specific This approach identified eight evi- whereas common complications of criteria vary, but the underlying prin- dence-based reviews, 26 meta-analytic tricyclic agents include more worri- ciples for identifying effective treat- studies, and 12 expert consensus state- some side effects, such as postural hy- ments are the same: support must be ments, which are summarized in the potension and arrhythmia (14). derived from well-designed con- five tables. The first two categories Table 2 summarizes the efficacy of trolled trials, and findings must be were used to determine the evidence psychosocial treatments for geriatric replicated by different investigators base defining effective treatments and depression (11,14,16,18,20–22,24– with sufficiently large samples from services, and the third was included to 30). In general, cognitive therapy, be- which results can be generalized (8,9). provide a synopsis of effective treat- havioral therapy, and cognitive-be- In the hierarchy of evidence-based ments and best practices from the per- havioral therapy have the greatest reviews of the literature, the highest spective of researchers and clinicians. empirical support for effectiveness in level is occupied by systematic re- the treatment of geriatric depression. views that evaluate the level of evi- Geriatric depression A variety of other psychosocial inter- dence with strict criteria and by ag- As shown in Table 1, there is general ventions are likely to be efficacious gregate meta-analyses of all relevant agreement on the effectiveness of an- among older adults, including prob- randomized controlled trials (8). The tidepressants for geriatric depression lem-solving therapy, interpersonal following section provides an (11–22). In general, more than half of therapy, brief psychodynamic thera- overview of the evidence base for older adults treated with antidepres- py, and reminiscence therapy. More- geriatric mental health interventions sants experience at least a 50 percent over, it is likely that a combination of derived from this standard of empiri- reduction in depressive symptoms pharmacological and psychosocial in- cal evidence. This overview of pub- (12). However, a recent meta-analysis terventions is more effective than ei-

1420 PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 Table 1 Pharmacological treatments for geriatric depression

Evidence source and reference Comments

Evidence-based reviews Thorpe et al., 2001 (11) 18 studies rated using standard guideline development criteria to determine preferred treatments. First-line treatments include bupropion, citalopram, fluvoxamine, mirtazapine, mo- clobemide, nefaxodone, paroxetine, sertraline, and venlafaxine. Electroconvulsive therapy (ECT) is effective. Meta-analyses Wilson et al., 2001 (12) 17 randomized controlled trials (RCTs) reviewed of community patients and inpatients over age 55. Tricyclics are effective. Selective serotonin reuptake inhibitors (SSRIs) and monoamine oxidase inhibitors (MAOIs) are likely to be effective. Discontinuation rates are similar across agents and placebo.

Anderson, 2000 (13) 11 RCTs support efficacy and tolerability for SSRIs and tricyclics among depressed adults over age 65. No significant difference in efficacy or tolerability between SSRIs and tricyclics.

Gerson et al., 1999 (14) 41 RCTs reviewed for treatment of major or unipolar depression for patients over age 55. Tricyclics, SSRIs, and “other” antidepressants are superior to placebo. Comparable efficacy and tolerability between tricyclics and SSRIs, but patients taking “other” antidepressants had lower dropout rates due to side effects.

Mulrow et al., 1999 (15) 27 RCTs reviewed for treatment of major depression among outpatients over age 60. SSRIs, newer SSRIs, and tricyclics are superior to placebo. Efficacy and discontinuation rates do not differ between classes.

McCusker et al., 1998 (16) 26 controlled studies reviewed for treatment of depression among adults aged 55 and over in community, outpatient, or nursing home settings. Heterocyclics and SSRIs are equally effective.

Mittmann et al., 1997 (17) 49 RCTs evaluated through 1996 for treatment of moderate or severe major or unipolar depression among patients over age 60. SSRIs, newer SSRIs, tricyclics, and MAOIs have similar efficacy, safety, and tolerability. Expert consensus statements Alexopoulos et al., 2001 (18) SSRIs preferred for all depression types; especially favorable results for citalopram, sertraline, and paroxetine. SSRIs or venlafaxine plus psychotherapy preferred for major depression. SSRIs plus psychotherapy preferred for mild depression or dysthymia. SSRIs or venlafaxine plus an atypical antipsychotic preferred for psychotic major depression. ECT is effective and a first-line treatment.

Mulsant et al., 2001 (19) SSRIs recommended as first-line antidepressants. Patients unable to tolerate or unrespon- sive to antidepressants can be switched to another agent or be treated with interpersonal psychotherapy.

American Society of Health- Similar efficacy for different classes of antidepressants. Selection based on side effect pro- System Pharmacists, 1998 (20) file, prior treatment response, type of depression, severity of symptoms, and concurrent drug therapy.

Lebowitz et al., 1997 (21) SSRIs and tricyclics have comparable efficacy. However, SSRIs may be preferred because they are easier to use, require less dosage adjustment, and have more favorable side effect profiles.

National Institutes of Health, Most antidepressants are equally effective. Amitriptyline and imipramine should be avoid- 1992 (22) ed. Newer antidepressants favored due to decreased anticholinergic and cardiovascular side effects. Treatment should be continued at a sufficient dose for six to 12 weeks.

ther intervention alone in preventing episodes in which there is a clearly cyclics and SSRIs) and psychothera- recurrence of major depression, al- identified psychosocial stressor (18). peutic interventions (cognitive-be- though replication of these findings is Finally, a meta-analysis that com- havioral, behavioral, and psychody- warranted (2,18). Expert consensus pared the rates of response to phar- namic therapies), although firm con- findings recommend the combined macological and psychological treat- clusions are not possible given the use of antidepressants and psy- ments of depression among patients small number of studies (14). chotherapy in the treatment of late- over the age of 55 found similar ef- Evidence-based reviews of inter- life depression, especially for fectiveness for antidepressants (tri- ventions for geriatric depression pri-

PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1421 Table 2 Psychosocial treatments for geriatric depression

Evidence source and reference Comments

Evidence-based reviews Laidlaw, 2001 (24) 6 meta-analyses and 10 outcome studies evaluated cognitive therapy for older adults with depression. It is effective for geriatric depression.

Thorpe et al., 2001 (11) 4 studies evaluated using standardized procedures to evaluate efficacy. Cognitive-behavioral therapy and interpersonal therapy for mild to moderate depression have the most support among psychotherapies.

Gatz et al., 1998 (25) 21 studies evaluated using evidence-based criteria to determine efficacy. Cognitive therapy, behavior therapy, cognitive-behavioral therapy, brief psychodynamic therapy, life review, reminiscence therapy are “likely to be effective.” Meta-analyses Pinquart and Soerensen, 122 controlled psychotherapeutic studies compared; each had an untreated control group with 2001 (26) depression (mean age of over 55). Cognitive-behavioral therapy, brief psychodynamic therapy, and supported psychotherapy are effective.

Gerson et al., 1999 (14) 4 randomized controlled trials (RCTs) evaluated, comparing treatment response and tolerability for people over age 55. Cognitive-behavioral therapy, brief psychodynamic therapy, and drug treatment have similar efficacy and tolerability.

Cuijpers, 1998 (27) 14 studies evaluated of the effectiveness of outpatient psychotherapy for adults over age 55, including 12 RCTs. Comparable efficacy found for cognitive-behavioral therapy, problem-solving therapy, behavior therapy, supportive therapy, reminiscence therapy, and brief psychodynamic therapy.

McCusker et al., 1998 (16) 14 controlled studies evaluated for treatment of depression in adults age over age 55 in community, outpatient, and nursing home settings. Cognitive therapy and behavior therapy better than no treatment but not better than a similar amount of nontherapeutic contact or attention given to a control group.

Engels and Verney, 1997 (28) 17 studies assessed for treatment of depression; age range, 52 to 81 years; mean age, 69 years. Treatment is more effective than placebo or no treatment. Behavior therapy and cognitive therapy are equally effective, and more effective than cognitive-behavioral therapy and brief psychodynamic psychotherapy. Individual therapy is more effective than group therapy.

Koder et al., 1996 (29) 7 studies evaluated of cognitive therapy for older people; mean age range, 65 to 70 years. It is more effective than wait-list control group and may be more effective than behavior therapy or brief psychodynamic psychotherapy.

Scogin and McElreath, 1994 (30) 17 studies of the efficacy of psychosocial treatments for depressed patients; mean age, 62 to 85 years. Comparable efficacy of cognitive therapy, behavior therapy, interpersonal therapy, and supportive therapy. Expert consensus statements Alexopoulos et al., 2001 (18) Preferred psychotherapies include cognitive-behavioral therapy, supportive psychotherapy, problem-solving therapy, and interpersonal therapy. Psychoeducation and family counseling are also supported.

American Society of Health- Cognitive therapy, behavior therapy, and interpersonal therapy are effective as primary inter- System Pharmacists, 1998 (20) ventions for older adults with mild to moderate depression or can be used in combination with pharmacotherapy.

Lebowitz et al., 1997 (21) Cognitive-behavioral therapy, behavior therapy, problem-solving therapy, and interpersonal therapy are effective alone and in combination with drug treatments.

National Institutes of Health, Psychosocial treatments are indicated in patients who do not tolerate or accept biological 1992 (22) treatments. Cognitive-behavioral therapy, behavior therapy, interpersonal therapy, and brief psychodynamic psychotherapy are moderately effective.

marily address major depression, with number of studies addressing the controlled studies of SSRIs in the little attention to the treatment of as- treatment of minor depression among treatment of older adults with minor sociated conditions such as minor de- older persons is limited. For example, depression suggest only modest ben- pression or suicidal behaviors. The the results of randomized placebo- efits (31). In addition, little is known

1422 PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 Table 3 Pharmacological and psychosocial treatment for cognitive symptoms of dementia

Evidence source and reference Comments

Evidence-based reviews Brodaty et al., 2001 (33) 7 meta-analyses or systematic reviews and 21 studies evaluated using U.S. Food and Drug Administration standards. Cholinesterase inhibitors supported for people with mild to moderate dementia. Ginkgo biloba has a small effect on cognitive performance, but evidence is weaker than for cholinesterase inhibitors. Doody et al., 2001 (34) 82 studies evaluated using standardized criteria to assign levels of evidence. Cholinesterase inhibitors have modest benefit for patients with Alzheimer’s disease; vitamin E likely delays clinical worsening; selegiline, other antioxidants, and anti-inflammatories require further study. Estrogen should not be prescribed to treat Alzheimer’s disease. Kasl-Godley and Gatz, 22 studies evaluated to determine empirical support for reality orientation therapy and 2000 (35) memory training. Memory training may optimize remaining ability. Reality orientation is useful for interpersonal but not cognitive functioning. Gatz et al., 1998 (25) 18 studies evaluated using evidence-based criteria. Memory and cognitive retraining programs may be effective in slowing decay of skills. Reminiscence is not effective. Reality orientation may improve orientation but does not generalize to other settings. American Psychiatric Association, 1997 (36) 8 randomized controlled trials (RCTs), 5 for tacrine and 3 for donepezil, support modest improvement of cognition with cholinesterase inhibitors. Possible delay of poor outcomes suggested by 1 RCT of vitamin E and 7 RCTS of selegiline. No indication for ergot mesylate in treatment of Alzheimer’s disease based on 7 RCTs. National Institute for Clinical Excellence, 2001 (37) 5 RCTs for donepezil, 5 for rivastigmine, and 3 for galantamine and 3 systematic reviews for donepezil, 3 for rivastigmine, and 1 for galantamine. All agents significantly improved cognitive functioning. Meta-analyses Birks and Flicker, 2001 (38) 15 RCTs evaluated for selegiline (a selective monoamine oxidase inhibitor at low doses) administered to patients with dementia for more than 1 day. Improvement in several memory tests, yet not enough evidence to recommend use in routine practice. Birks et al., 2001 (39) 8 RCTs of donepezil (a cholinesterase inhibitor) evaluated for patients with mild to moderate Alzheimer's disease treated for 12, 24, and 52 weeks. Modest improvements in cognitive functioning and global clinical ratings compared with placebo. Fioravanti and Flicker, 11 RCTs of nicergoline (an ergot derivative) assessed. Potential benefit found for cognitive 2001 (40) and behavioral symptoms in vascular dementia, but is associated with an increased risk of adverse effects. No definitive studies of use in Alzheimer’s disease. Fioravanti and Yanagi, 12 RCTs of cytidinediphosphocholine (a phosphatidylcholine precursor) assessed. Some 2001 (41) evidence suggests positive short-term effects on memory, behavior, and global impression. Olin and Schneider, 7 RCTs evaluated of galantamine (a cholinesterase inhibitor) for mild to moderate 2001 (42) Alzheimer’s disease administered for 3 to 6 months. Galantamine improves global ratings, cognition, activities of daily living, and behavior with effect and tolerability comparable to that of other cholinesterase inhibitors. Spector et al., 2001 (43) 6 RCTs of reality orientation assessed. Some evidence that reality orientation benefits cognitive and behavioral symptoms of dementia. Birks et al., 2000 (44) 7 RCTs of rivastigmine (a cholinesterase inhibitor) evaluated for patients with Alzheimer’s disease treated for more than 2 weeks. Rivastigmine is more effective than placebo in improving cognitive function and activities of daily living. Higgins and Flicker, 2000 (45) 12 RCTs of lecithin (a dietary source of choline) evaluated. Evidence does not support the use of lecithin in the treatment of dementia or cognitive impairment. Qizilbash et al., 1998 (46) 12 RCTs of tacrine (a cholinesterase inhibitor) assessed. It reduces deterioration in cognitive performance over the first 3 months of treatment and may result in global improvement. Oken et al., 1998 (47) 4 RCTs of gingko biloba for patients with Alzheimer’s disease. Small effect on cognitive function over 4 to 6 months of treatment. No significant adverse effects. Expert consensus statements Patterson et al., 1999 (48) 12 studies reviewed supporting an evidence-based consensus statement. Donepezil improves cognitive functioning in mild to moderate dementia. Insufficient evidence to recommend vitamin E or ginkgo biloba for treatment or prevention of Alzheimer’s disease. Small et al., 1997 (49) Cholinesterase inhibitors slow cognitive decline. Evidence for other agents is inconclusive. Reality orientation and memory retraining may be beneficial, but the associated risks of frustration and depression may outweigh the small benefits.

PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1423 Table 4 Pharmacological and psychosocial treatment for behavioral symptoms of dementia Evidence source and reference Comments

Evidence-based reviews Doody et al., 2001 (34) 94 studies of pharmacological and psychosocial treatment evaluated. Antipsychotics are effective for agitation or psychosis when environmental approaches fail; antidepressants are effective in depression with dementia. Behavior modification and skills training can also be effective.

Kasl-Godley and Gatz, 22 studies of psychosocial and behavioral interventions evaluated. Reminiscence and life re- 2000 (35) view result in small improvements in interpersonal behavior. Support groups and cognitive therapy or behavior therapy help build coping skills and reduce distress. Behavioral approaches are helpful with early-stage dementia.

Gatz et al., 1998 (25) 12 studies of psychosocial and behavioral interventions evaluated using DSM criteria. Rein- forcement, extinction, and stimulus control are effective, although they have a limited scope. Environmental approaches, such as milieu therapy, a token economy, and environmental modifications, are effective.

American Psychiatric 7 randomized controlled trials (RCTs) of pharmacological interventions reviewed. Modest Association, 1997 (36) improvement of agitation and psychosis in dementia with conventional antipsychotics. 7 RCTs of benzodiazepines show improvement of agitation compared with placebo but not better than antipsychotics. Insufficient data to assess atypical antipsychotics or anticonvulsants. 5 RCTs of antidepressant treatment of depression in dementia suggest benefit, although limited by small samples and by selection criteria. Meta-analyses Kirchner et al., 2001 (59) 12 RCTs evaluated for thioridazine (a conventional neuroleptic). No evidence to support its use in the treatment of dementia. Only positive effect was reduction in anxiety.

Lonergan et al., 2001 (60) 5 RCTs of haloperidol (a conventional neuroleptic) assessed for agitation in dementia. Evi- dence supports its use in the control of aggression. No evidence for improvement in other forms of agitation, and there are frequent side effects.

Olin et al., 2001 (61) 19 RCTs evaluated for hydergine (an ergoloid mesylate). Significant treatment effects when assessed by either global ratings or comprehensive rating scales. Because of uncertainty in diagnostic criteria, the efficacy of hydergine for dementia is not clear.

Lanctot et al., 1998 (62) 16 RCTs (1966 to 1997) of conventional neuroleptics. No difference in efficacy between different agents. Conventional neuroleptics have modest efficacy compared with placebo.

Schneider et al., 1990 (63) 7 RCTs of conventional neuroleptics (1960 to 1982). They are modestly more effective than placebo, but the effect size was small (r=.18). No single agent is better than another. Expert consensus statements Herrmann, 2001 (64) Nonpharmacological approaches favored as first-line treatment for behavioral symptoms of dementia, although high-quality research is limited. Atypical antipsychotics, antidepressants, and anticonvulsants are modestly effective in reducing behavioral symptoms. Benzodi- azepines may be used if necessary. Pharmacotherapy should be monitored for effectiveness and side effects.

Patterson et al., 1999 (48) 24 studies reviewed supporting an evidence-based consensus statement. Environmental and behavioral modifications should be first-line treatments for behavioral problems. If medications are required, low doses of antipsychotics, a selective serotonin reuptake inhibitor (SSRI), or trazodone should be considered.

Alexopoulos et al., 1998 (65) Combined medication and environmental interventions favored as first-line treatment for agitation in dementia. Mild agitation treatment: structured routines, reassurance, and social- ization; severe agitation treatment: supervision and environmental safety. Both should include education and support for family and caregivers. Preferred medication varies with presenting conditions: For psychosis, risperidone or a conventional, high-potency antipsychotic; for depression, an antidepressant alone (sertraline or paroxetine); for aggression and anger, divalproex, risperidone, a conventional, high potency antipsychotic, an SSRI, trazodone, or buspirone.

Small et al., 1997 (49) SSRIs favored as first-line treatments for depression in dementia. Tricyclics are effective but have greater side effects. Antipsychotics are modestly effective for behavioral problems and psychotic symptoms. More studies are needed to establish efficacy of other agents. Psy- chotherapy may decrease behavioral problems and improve mood.

1424 PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 Table 5 Pharmacological and psychosocial treatment for geriatric alcohol abuse Evidence source and reference

Evidence-based reviews Gatz et al., 1998 (25) 3 studies evaluated using evidence-based criteria to determine treatment efficacy. Reminis- cence, age segregation, and a supportive climate are promising but require further study. Expert consensus statements Center for Substance Abuse Brief interventions, motivational counseling, and family interventions recommended. Treat- Treatment, 1998 (72) ment principles: age specific; supportive group treatment; focus on coping with depression, loneliness, and loss; rebuilding social support network; pace and content appropriate for older persons; clinicians interested and experienced in older adult populations; linkage with medical and aging services, case management, and referral sources.

Council on Scientific Affairs, Detoxification should occur in a hospital setting and medications should be carefully moni- 1996 (73) tored. Age-specific groups and programs that emphasize social relationships and positive as- pects of a patient’s life have better outcomes for older adults.

about the efficacy of interventions in ness of cholinesterase inhibitors com- pessimistic view of treatments for de- preventing suicidal behaviors among pared with placebo in modestly reduc- mentia, there is an emerging evidence older adults, even though the rate of ing the rate of decline or enhancing base supporting modest effectiveness suicide among older adults is greater cognitive functioning over the course of cholinesterase inhibitors in tem- than in any other age group (32). An of six to 12 months. In addition, evi- porarily decreasing cognitive decline evidence-based review of the litera- dence is emerging that cholinesterase and enhancing cognitive functioning ture suggests that the only supported inhibitors may be effective in delaying for mild to moderate Alzheimer’s de- preventive intervention for late-life nursing home placement (50) and that mentia over six to 12 months. suicide is the identification and effec- they may improve cognitive function- tive treatment of depression (32). ing in severe Alzheimer’s dementia Behavioral symptoms In summary, there is a well-sub- (51) and in some non-Alzheimer’s de- Thirty to 40 percent of persons with stantiated evidence base supporting mentias (52). Selegiline may also be Alzheimer’s dementia experience be- the efficacy of antidepressants and used, although it has a less favorable havioral symptoms, including agita- cognitive, behavioral, and cognitive- risk-benefit ratio and less supporting tion, psychosis, and depression, at behavioral therapy in the acute and evidence (34,38). In the one placebo- some point during the disease (49). short-term treatment of geriatric ma- controlled trial of vitamin E, no signifi- Limited research consisting of indi- jor depression. However, caution is cant differences were found in cogni- vidual randomized placebo-controlled indicated in interpreting the results of tive signs and symptoms, although vita- studies of conventional antipsychotics individual studies that report the su- min E minimally slowed progression to (54–56) and novel antipsychotics periority of one treatment over anoth- institutionalization (53). The effective- (56–58) supports the modest effec- er—for example, SSRIs over tricyclic ness of other antioxidants, anti-inflam- tiveness of these agents for the treat- agents—because of the potential matories, estrogen replacement, gink- ment of agitation and dementia com- sources of bias, which include indus- go biloba extracts, and other agents is pared with placebo. However, aggre- try sponsorship of clinical trials, sam- not supported by evidence. However, gate analyses of multiple trials of an- ple selection, and study design. research on the pharmacotherapy of tipsychotics are less conclusive. Alzheimer’s disease is a rapidly advanc- As shown in Table 4 (25,34– Dementia ing field, and a variety of multicenter 36,48,49, 59–65), evidence-based re- Evidence of treatment effectiveness trials that hold promise for expanding views of pharmacological treatments for dementia can be separated into the array of available evidence-based generally have found that antipsychot- studies of cognitive symptoms, such treatments are under way. ic agents are effective in the treatment as problems with memory, language, In general, psychosocial treatments of behavioral symptoms; however, and abstraction, and studies of behav- for the cognitive symptoms of demen- meta-analyses of studies of single ioral symptoms, such as agitation, tia are not effective (25,34,35,49). A agents or classes of antipsychotics have psychosis, and depression. review of the empirical evidence for shown no effect or modest improve- cognitive retraining programs and re- ment. In contrast, consensus state- Cognitive symptoms ality orientation suggests that these in- ments widely support the use of an- As shown in Table 3 (25,33–49), for terventions may temporarily improve tipsychotics and favor the use of novel patients with mild to moderate de- cognitive, behavioral, and functional antipsychotics over conventional mentia associated with Alzheimer’s skills, but compelling evidence of sus- agents (48,49). In addition, there is ac- disease, evidence-based reviews and tained benefit is lacking (25,43). cumulating evidence that antidepres- meta-analyses agree on the effective- In summary, despite a traditionally sants and anticonvulsants are effective

PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1425 in reducing agitation and other behav- Alcohol abuse younger patients did not find evi- ioral symptoms of dementia (64). Consensus statements (72,73) and dence of significant differences in ef- A limited body of literature suggests general reviews (74,75) provide little ficacy (91,92). However, atypical that the use of cholinesterase in- endorsement for the effectiveness of agents have been shown to be safer hibitors can result in changes in be- pharmacological interventions for for older persons in terms of motor havior and functioning that are de- geriatric alcohol abuse. In contrast, side effects, especially tardive dyski- tected by both physicians and care- psychosocial interventions are likely nesia (93). givers, although these findings are to be effective for older persons with Data on psychosocial interventions based on subanalyses of trials that did alcohol use disorders (Table 5). for older adults with schizophrenia are not enroll patients with dementia Promising treatment components in- lacking. The literature on the effec- specifically for behavioral problems clude separate treatment groups for tiveness of psychosocial treatments (34). In addition, a considerably small- older persons, supportive and non- for geriatric schizophrenia is limited er literature base has examined treat- confrontational treatment approach- to a single controlled pilot study sug- ments for depression in dementia. es, and group or individual cognitive- gesting the potential benefits of a The effectiveness of tricyclic antide- behavioral therapy (25). In particular, combination of cognitive-behavioral pressants for depression in dementia there is compelling evidence that therapy and skills training (94). A con- is not supported (66). However, a re- brief cognitive-behavioral interven- sensus statement supports residential cent review suggests that SSRIs may tions are effective in treating late-life alternatives rather than long-term have some benefit (34). alcohol abuse (76). hospitalization and the provision of Behavioral and environmental mod- In summary, age-specific, noncon- social and vocational skills training, ifications are also effective in enhanc- frontational, brief motivational, and community support programs, and ing functioning and reducing problem cognitive-behavioral therapies show psychoeducational programs for fami- behaviors associated with dementia. promise as interventions for alcohol ly members (77). However, the lack of Interventions include light exercise or abuse in geriatric populations. data supporting these recommenda- music (34,35,67), behavioral or social tions is noted, which underscores rec- reinforcement, and environmental Schizophrenia ommendations for studies addressing modifications, such as access to an out- We found no evidence-based reviews this research gap (1). door area, simulated home environ- or meta-analyses of treatment for In summary, the efficacy of antipsy- ments, and reduced-stimulation units schizophrenia among older persons. chotic treatment of schizophrenia for agitated residents (25,67). Psychoe- Nonetheless, a consensus statement among older persons is supported by ducational training and support groups on late-life schizophrenia (77) and individual studies and general reviews; for caregivers have been shown to de- general reviews of the treatment of however, no evidence-based reviews lay placement in nursing homes and to psychosis in the elderly population or meta-analyses have been published. decrease caregiver stress (34). (78,79) endorsed the view that an- In summary, empirical evidence sup- tipsychotic medications are effective. Anxiety disorders ports the value of psychosocial inter- For example, reviews have com- Anxiety is one of the most common ventions in addressing behavioral pared the relative merits and potential mental health problems affecting old- symptoms of dementia, but there is complications of conventional an- er adults (2). However, there is a less agreement on the effectiveness of tipsychotic agents (80,81) and novel paucity of research on the effective- antipsychotic, anticonvulsant, and anti- antipsychotics (80–84) for the treat- ness of available treatments. General depressant agents. However, aggregate ment of psychosis among older per- reviews of the literature provide a lim- analyses of the literature should be in- sons. Clinical reviews have reported ited perspective on the effectiveness terpreted with caution because of the that older persons are more suscepti- of treatments for geriatric anxiety dis- substantial heterogeneity in diagnostic ble to adverse effects of conventional orders (2,95–98). These reviews re- criteria, the inclusion of patients with antipsychotics, including parkinsonian port that benzodiazepines are the different types of dementia, the vari- side effects and tardive dyskinesia most frequently prescribed antianxi- ability in specification of the interven- (80–82). Recent research on the use ety medication among older persons tions, and the difficulty of rigorously as- of novel antipsychotics among older and recommend consideration of sessing outcomes in this population. adults is largely limited to open-label pharmacological alternatives. Howev- Finally, it is imperative that clinical uncontrolled studies (85–88) and a er, few double-blind placebo-con- assessment of all behavioral and cog- small number of controlled trials trolled trials have been conducted nitive symptoms includes the differ- (89,90). Overall, the reports and re- with this population (96). ential diagnosis of delirium. The in- views suggest that atypical antipsy- Despite preliminary results sug- creased risk of delirium among older chotics should be considered as first- gesting possible benefits of cognitive- persons (68) and the poor prognosis line agents in the treatment of schizo- behavioral therapy in the treatment of (69) warrant a careful and systematic phrenia among older persons. Recent geriatric anxiety disorders, conclusive assessment of the wide spectrum of systematic reviews comparing the ef- findings are not available (25,96,99). possible etiologies and the appropri- fectiveness and cost-effectiveness of Other promising but inadequately re- ate treatment of the cause and associ- conventional agents and atypical searched psychotherapy treatments ated symptoms (70,71). agents other than clozapine among include cognitive-behavioral group

1426 PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 therapy, cognitive restructuring, indi- and social service professionals who For example, academic detailing, vidual behavioral therapy, and sup- have training and expertise in geri- which consists of brief one-on-one portive group psychotherapy (2). atric mental health care (1,3). The educational sessions coupled with In summary, the limited empirical different priorities, capacities, and provider-specific feedback on treat- evidence confirms the efficacy of levels of expertise between primary ment practices, is effective in influ- treatment with conventional antianxi- care, long-term care, and specialty encing the practice behavior of pri- ety agents, while acknowledging the mental health providers in the areas mary care physicians (109). Changes potential problems associated with of aging and mental health care fur- resulting from this novel educational benzodiazepines. Cognitive-behav- ther complicate implementation of intervention include short-term im- ioral therapy has the greatest support evidence-based treatments (3,102). provement in rates of detection of the among psychosocial interventions. In summary, there is a substantial target disorder (110) and a decrease shortfall in the provision of psychi- in prescriptions for medications that Models of service delivery atric interventions in usual-care set- are not indicated (109). In addition to research on treatments tings. Nearly half of older adults with Other effective interventions in- for specific disorders, a limited body a recognized mental disorder have clude changing the process of care of literature has examined the effec- unmet needs for services (103). within a physician’s office. For exam- tiveness of various models of service ple, interventions to change the sys- delivery. A review of the evidence Implementation research tem have been found to result in sig- base found the greatest support for An evolving practice research litera- nificant improvements in quality of community-based, multidisciplinary, ture describes methods that may ef- care and patient outcomes (107). geriatric mental health treatment fectively improve the implementation Such interventions include combina- teams (100). Promising although in- and use of evidence-based practices tions of physician and patient educa- conclusive data were found on the ef- by mental health providers who serve tion, care management, and im- fectiveness of hospital-based geriatric older adults in usual care settings. proved coordination among mental psychiatry consultation-liaison servic- Primary care. Most older persons health and primary care providers. es. In contrast, no randomized con- who receive mental health care are Simple but effective interventions for trolled studies have examined the ef- treated by primary care physicians facilitating the process of care also in- fectiveness of geropsychiatric inpa- (102,103). Yet the many demands of clude tools that monitor patients’ tient units or day hospital programs. primary care present substantial chal- progress, such as severity measures Finally, the effectiveness of geriatric lenges to such care (2). Older persons (111) and systems for scheduling rou- consultation services to nursing with psychiatric illnesses are more tine follow-up visits (112). Care man- homes is inconclusive. One review likely to receive inappropriate phar- agement can also help improve treat- (101) found a randomized controlled macological treatment and less likely ment adherence and facilitate moni- trial that showed no significant differ- to be treated with psychotherapeutic toring of treatment response. In this ences in clinical outcomes between interventions than younger primary model, the geriatric psychiatrist or patients who received psychiatry con- care patients (104). other specialist supervises care man- sultation services and those who re- Considerable attention has been agers and provides limited consulta- ceived usual care. focused on educational efforts to im- tion to the primary care or general In summary, empirical evidence prove screening for and treatment of psychiatrist providers. supports the effectiveness of commu- depression by primary care providers, Finally, another approach is inte- nity-based, multidisciplinary geriatric yet the failures of these traditional ap- gration of services through collabora- mental health treatment teams. proaches as a means of improving tion between providers of specialty physicians’ practices are well docu- mental health care and primary care Implementing evidence-based mented. For example, providing in a common setting. A mental health practices practice guidelines to clinicians with- clinician who provides collaborative Challenges out additional incentives or interven- care is situated in the primary care Despite evidence supporting the effi- tions aimed at changing practices is practice setting and coordinates as- cacy of a variety of interventions for ineffective in changing their behavior sessment and treatment services with geriatric mental disorders, the imple- (105–107). Although physician edu- the medical provider (113). mentation of these interventions in cation is necessary, it alone is not ef- Long-term care. Observational usual care settings is limited. Reasons fective in enhancing guideline-con- studies suggest that mental health con- for this limited implementation in- cordant care. Grand rounds presenta- sultation services in nursing homes clude organizational barriers, bias tions and physician conferences, the may be associated with better out- and ageism among providers, inade- mainstay of conventional continuing comes for residents (101). However, quate and discriminatory financing of medical education, are generally inef- few randomized controlled studies of mental health services for older per- fective by themselves (108). Educa- these programs have been conducted. sons, and lack of collaboration and co- tional interventions that actively in- Training in assessment and manage- ordination between providers (2,3). volve the learner and use multiple ment of behavioral problems has been These barriers are further complicat- techniques are most effective in shown to reduce turnover of clinical ed by national shortages of medical changing physicians’ behavior (106). staff (114) and improve the knowledge

PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1427 and performance of nursing staff by consumers and their families, who include small samples and lack of pow- (115,116). Educational outreach inter- will ultimately decide to follow or re- er, heterogeneity of samples, lack of in- ventions are also effective in changing ject recommended treatments. Multi- terchangeable instruments, lack of ex- the clinical practice behavior of pre- media educational materials that are tractable data, different definitions of scribing physicians when education sensitive to the preferences and needs outcomes, differences in the quality of and feedback is provided on an indi- of consumers and their family are im- research and the duration of the stud- vidual basis (117). In addition to evi- portant in supporting the acceptance ies, and reliance on statistical rather dence-based interventions, a series of and use of evidence-based treatments. than clinical significance (63,122). guidelines specific to the treatment of In summary, a successful strategy Furthermore, evidence-based re- major mental health disorders have for implementing evidence-based views and meta-analyses are largely been developed to assist nursing home mental health interventions is ground- dependent on data from randomized professionals in caring for older adults ed in a systems approach, combined controlled trials that compare a single with mental illness (118,119). with the development and dissemina- well-defined intervention with a Specialty services in the com- tion of easy-to-use implementation placebo or other control. Thus they munity. There are few data on im- kits and well-described procedures for are less suited to inform more com- proving the adherence of community- changing practices (121). plex decisions, such as choosing the based specialty mental health next step after a series of failed inter- providers to empirically based geri- Conclusions ventions for a treatment-refractory atric mental health practices. General This overview of research defining ev- condition or making the most effec- mental health clinicians lack training idence-based practices in geriatric tive use of the many different possible in basic assessment and treatment of mental health care suggests that there combinations of agents. The large the mental disorders of aging. System is a need to address the profound gap number of potential combinations and change interventions that support cli- between research findings on effec- sequences of treatments and the large nicians in the use of assessment and tive treatments and the current avail- number of different clinical condi- treatment planning toolkits have been ability of such treatments for older tions and comorbid physical condi- shown to improve clinicians’ adher- persons with mental disorders. tions make it virtually impossible to ence to standardized geriatric assess- However, several caveats are indi- support all clinical decisions with data ment practices (120). However, data cated in considering such an effort. from randomized controlled trials (8). are lacking on interventions aimed at First, identification of evidence-based One approach to addressing gaps improving the use of evidence-based practices should be considered as a left by standardized evidence-based treatments in community settings. starting point for improving the quali- reviews and meta-analyses is the use ty of care. In essence, evidence-based of expert consensus guidelines. Re- Strategies practices define “the floor” in quality cently published guidelines on the The gap between research findings and should not be confused with best, pharmacotherapy of depression on empirically supported treatments optimal, or promising practices. Sec- among older patients are an example and clinical practice suggests the ond, there is a misperception that only of treatment recommendations based need for an organized strategy to fa- randomized controlled trials, meta- on an aggregate analysis of independ- cilitate the implementation of geri- analyses, or systematic reviews can ent ratings by experts on the appro- atric evidence-based practices in usu- constitute the evidence base. Evi- priateness of various treatment op- al care settings. Key elements of an dence-based practice is based on care- tions (18). In addition, the guidelines approach to implementing evidence- ful and appropriate use of the findings for major psychiatric disorders devel- based practices emphasize the in- of the best relevant studies, accompa- oped by the American Psychiatric As- volvement of stakeholder groups, in- nied by an appreciation of the limits of sociation (APA) (36,70, 123–127) pro- cluding administrators, clinicians, the existing data. In some instances, vide treatment recommendations consumers, and families (121). Imple- the best studies include randomized that are assigned one of three levels mentation guides for mental health controlled trials, whereas in other sit- of confidence based on clinical con- authorities and administrators are de- uations, nonrandomized outcome sensus. With the exception of the signed to address the reorganization studies or case reports may constitute guidelines on dementia, the APA of practice environments, procedures, the evidence base. guidelines are not age specific, sug- incentives, and reimbursements to in- An additional and important consid- gesting that future initiatives may be corporate empirically supported treat- eration involves inherent limitations undertaken to develop clinical guide- ment practices. In contrast, materials in the methodology used to identify lines specific to older adults. for clinicians are designed to accom- evidence-based practices, which may In general, guidelines and treat- modate different levels of expertise result in the overly conservative exclu- ment algorithms can provide the cli- and to provide decision support tech- sion of informative studies or may nician with a practical and compre- nologies and treatment guides that fa- group studies together without ade- hensive summary of recommenda- cilitate the use of evidence-based quate attention to important differ- tions. However, caution is warranted. treatments in routine practice. Final- ences between studies. For example, Guidelines should be evaluated on ly, the adoption of empirically sup- common problems affecting meta- the basis of their level of support from ported treatments depends on buy-in analyses and evidence-based reviews systematic reviews of the evidence,

1428 PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 because the consensus of experts may Evidence-Based Medicine Working Group, 23. Kirsch I, Moore TJ, Scoboria A, et al: The American Medical Association, AMA Press, emperor’s new drugs: an analysis of antide- inadvertently incorporate the bias of 2002 pressant medication submitted to the US specialties and disciplines and may Food and Drug Administration. Prevention 9. Sackett DL, Rosenberg WM, Gray JA, et al: and Treatment, 5, 2002. Available at misrepresent treatment effectiveness Evidence-based medicine: what it is and www.journals.apa.org/prevention/vol- or side effects (128,129). what it isn’t. British Medical Journal 312: ume5/pre0050023a.html 71–72, 1996 Finally, despite advances in clinical 24. Laidlaw K: An empirical review of cognitive expertise and research, a substantial 10. Robins LN, Regier DA: Psychiatric Disor- therapy for late life depression: does re- body of literature chronicles the fail- ders in America: The Epidemiologic Catch- search evidence suggest adaptations are ment Area Study. New York, Free Press, necessary for cognitive therapy with older ure of conventional educational ap- 1991 adults? Clinical Psychology and Psy- proaches and the limited impact of 11. Thorpe L, Whitney DK, Kutcher SP, et al: chotherapy 8:1–14, 2001 disseminating treatment guidelines. Clinical guidelines for the treatment of de- 25. Gatz M, Fiske A, Fox LS, et al: Empirically Meaningful changes in the quality of pressive disorders: VI. special populations. validated psychological treatments for old- Canadian Journal of Psychiatry 46:63S– er adults. Journal of Mental Health and Ag- care require innovative technologies 76S, 2001 that use research findings focused on ing 4:9–46, 1998 12. Wilson K, Mottram P, Sivanranthan A, et al: organizational and provider change. 26. Pinquart M, Soerensen S: How effective Antidepressant versus placebo for de- are psychotherapeutic and other psychoso- This challenge is compounded in a pressed elderly (Cochrane Review) in cial interventions with older adults? A system of geriatric mental health care Cochrane Library. Oxford, England, Up- meta-analysis. Journal of Mental Health date Software, 2001 that includes different organizations and Aging 7:207–243, 2001 and sources of financing and treat- 13. Anderson IM: Selective serotonin reuptake 27. Cuijpers P: Psychological outreach pro- inhibitors versus tricyclic antidepressants: a grammes for the depressed elderly: a meta- ment by providers from different dis- meta-analysis of efficacy and tolerability. analysis of effects and dropouts. Interna- Journal of Affective Disorders 58:19–36, ciplines. Despite these challenges, tional Journal of Geriatric Psychiatry 13: 2000 there is a clear and urgent demo- 41–48, 1998 14. Gerson S, Belin TR, Kaufman A, et al: graphic imperative to address the 28. Engels GI, Verney M: Efficacy of nonmed- Pharmacological and psychological treat- emerging public health problem of ical treatments of depression in elders: a ments for depressed older patients: a meta- the mental disorders of aging. It is quantitative analysis. Journal of Clinical analysis and overview of recent findings. Geropsychology 3:17–35, 1997 time for geriatric psychiatry to take Harvard Review of Psychiatry 7:1–28, 1999 up the mantle of evidence-based 29. Koder DA, Brodaty H, Anstey KJ: Cogni- 15. Mulrow CD, Williams JW, Trivedi M, et al: tive therapy for depression in the elderly. practices and translate research find- Treatment of Depression: Newer Pharma- International Journal of Geriatric Psychia- ings into the mainstream of clinical cotherapies. Evidence Report, Technology try 11:97–107, 1996 o Assessment 7. AHCPR pub no 99-E014. treatment for older Americans. Rockville, Md, Agency for Health Care Pol- 30. Scogin F, McElreath L: Efficacy of psy- icy and Research, 1999 chosocial treatments for geriatric depres- References sion: a quantitative review. Journal of Con- 16. McCusker J, Cole M, Keller E, et al: Effec- sulting and Clinical Psychology 62:69–74, 1. Jeste DV, Alexopoulos GS, Bartels SJ, et al: tiveness of treatments of depression in old- 1994 Consensus statement on the upcoming cri- er ambulatory patients. Archives of Internal sis in geriatric mental health: research Medicine 158:705–712, 1998 31. Oxman TE, Sengupta AS: Treatment of mi- agenda for the next two decades. Archives nor depression. American Journal of Geri- of General Psychiatry 56:848–853, 1999 17. Mittmann N, Herrmann N, Einarson TR, atric Psychiatry 10:256–264, 2002 et al: The efficacy, safety, and tolerability of 2. Mental Health: A Report of the Surgeon antidepressants in late life depression: a 32. Pearson JL, Brown GK: Suicide prevention General. Rockville, Md, US Department of meta-analysis. Journal of Affective Disor- in late life: directions for science and prac- Health and Human Services, 1999 ders 46:191–217, 1997 tice. Clinical Psychology Review 20:685– 705, 2000 3. Older Adults and Mental Health: Issues and 18. Alexopoulos GS, Katz IR, Reynolds CF, et Opportunities. Rockville, Md, Administra- al: The Expert Consensus Guideline Series: 33. Brodaty H, Ames D, Boundy KL, et al: tion on Aging, US Department of Health pharmacotherapy of depressive disorders in Pharmacological treatment of cognitive and Human Services, 2001 older patients. Postgraduate Medicine 110 deficits in Alzheimer’s disease. Medical (Oct special report):1–86, 2001 Journal of Australia 175:324–329, 2001 4. Bartels SJ: Quality, costs, and effectiveness of services for older adults with mental dis- 19. Mulsant BH, Alexopoulos GS, Reynolds CF 34. Doody RS, Stevens JC, Beck C, et al: Prac- orders: a selective overview of recent ad- III, et al: Pharmacological treatment of de- tice parameter: management of dementia vances in geriatric mental health services pression in older primary care patients: the (an evidence-based review): report of the research. Current Opinion in Psychiatry 15: PROSPECT algorithm. International Jour- quality standards subcommittee of the 411–416, 2002 nal of Geriatric Psychiatry 16:585–592, 2001 American Academy of Neurology. Neurolo- gy 56:1154–1166, 2001 5. Institute of Medicine: Crossing the Quality 20. American Society of Health-System Phar- Chasm: A New Health System for the 21st macists: Therapeutic position statement on 35. Kasl-Godley J, Gatz M: Psychosocial inter- Century. Washington, DC, National Acade- the recognition and treatment of depres- ventions for individuals with dementia: an my Press, 2001 sion in older adults. American Journal of integration of theory, therapy, and a clinical Health-System Pharmacy 55:2514–2518, understanding of dementia. Clinical Psy- 6. Bridging Science and Service. Rockville, 1998 chology Review 20:755–782, 2000 Md, National Institute of Mental Health, 1999 21. Lebowitz BD, Pearson JL, Schneider LS, 36. American Psychiatric Association: Practice et al: Diagnosis and treatment of depres- guideline for the treatment of patients with 7. Cochrane AL: Effectiveness and Efficien- sion in late life: consensus statement up- Alzheimer’s disease and other dementias of cy: Random Reflections on Health Ser- date. JAMA 278:1186–1190, 1997 late life. American Journal of Psychiatry vices. London, Nuffield Provincial Hospi- 154(5 suppl):1–39, 1997 tals Trust, 1972 22. NIH consensus development conference statement: diagnosis and treatment of de- 37. Guidance on the Use of Donepezil, Ri- 8. Guyatt G, Rennie D: Users’ Guide to the pression in late life. JAMA 268:1018–1024, vastigmine, and Galantamine for the Treat- Medical Literature: A Manual for Evi- 1992 ment of Alzheimer’s Disease. Technology dence-Based Clinical Practice. Chicago, Appraisal Guidance no 19. National Insti-

PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1429 tute for Clinical Excellence, London, 2001 bined with cerebrovascular disease: a ran- Psychiatry 9:361–381, 2001 domised trial. Lancet 359:1283–1290, 2002 38. Birks J, Flicker L: Selegiline for Alzheimer’s 68. Elie M, Cole MG, Primeau FJ, et al: Delir- disease (Cochrane Review) in Cochrane Li- 53. Tabet N, Birks J, Grimley Evans J, et al: Vi- ium risk factors in elderly hospitalized pa- brary. Oxford, England, Update Software, tamin E for Alzheimer’s disease (Cochrane tients. Journal of General Internal Medi- 2001 Review) in Cochrane Library. Oxford, Eng- cine 13:204–12, 1998 land, Update Software, 2001 39. Birks JS, Melzer D, Beppu H: Donepezil 69. Cole MG, Primeau FJ: Prognosis of deliri- for mild and moderate Alzheimer’s disease 54. Devanand DP, Marder K, Michaels KS, et um in elderly hospital patients. Canadian (Cochrane Review) in Cochrane Library. al: A randomized, placebo-controlled dose- Medical Association Journal 149:41–46, Oxford, England, Update Software, 2001 comparison trial of haloperidol for psy- 1993 chosis and disruptive behaviors in 40. Fioravanti M, Flicker L: Efficacy of nicer- Alzheimer’s disease. American Journal of 70. American Psychiatric Association: Practice goline in dementia and other age associated Psychiatry 155:1512–1520, 1998 guideline for the treatment of patients with forms of cognitive impairment (Cochrane delirium. American Journal of Psychiatry Review) in Cochrane Library. Oxford, Eng- 55. Teri L, Logsdon RG, Peskind E, et al: 156(5 suppl):1–20, 1999 land, Update Software, 2001 Treatment of agitation in Alzheimer’s disease: a randomized, placebo-controlled 71. Cole MG, Primeau FJ, Elie LM: Delirium: 41. Fioravanti M, Yanagi M: Cytidinediphos- clinical trial. Neurology 55:1271–1278, prevention, treatment, and outcome stud- phocholine (CDP choline) for cognitive 2000 ies. Journal of Geriatric Psychiatry and and behavioral disturbances associated with Neurology 11:126–137, 1998 chronic cerebral disorders in the elderly 56. De Deyn P, Rabheru K, Rasmussen A, et al: 72. Substance Abuse Among Older Adults. (Cochrane Review) in Cochrane Library. A randomized trial of risperidone, placebo, Treatment Improvement Protocol, 26. Oxford, England, Update Software, 2001 and haloperidol for behavioral symptoms of dementia. Neurology 53:946–955, 1999 Rockville, Md, Center for Substance Abuse 42. Olin J, Schneider L: Galantamine for Treatment, 1998 Alzheimer’s disease (Cochrane Review) in 57. Katz IR, Jeste DV, Mintzer JE, et al: Com- parison of risperidone and placebo for psy- 73. Council on Scientific Affairs of the Ameri- Cochrane Library. Oxford, England, Up- can Medical Association: Alcoholism in the date Software, 2001 chosis and behavioral disturbances associated with dementia: a randomized, double- elderly. JAMA 275:797–801, 1996 43. Spector A, Orrell M, Davies S, et al: Reali- blind trial. Journal of Clinical Psychiatry 60: 74. Schonfeld L, Dupree LW: Treatment ap- ty orientation for dementia (Cochrane Re- 107–115, 1999 proaches for older problem drinkers. Inter- view) in Cochrane Library. Oxford, Eng- national Journal of the Addictions land, Update Software, 2001 58. Bhana N, Spencer CM: Risperidone: a review of its use in the management of the 30:1819–1842, 1995 44. Birks J, Grimley EJ, Iakovidou V, et al: Ri- behavioural and psychological symptoms of 75. Fingerhood M: Substance abuse in older vastigmine for Alzheimer’s disease dementia. Drugs and Aging 16:451–471, people. Journal of the American Geriatrics (Cochrane Review) in Cochrane Library. 2000 Society 48:985–995, 2000 Oxford, England, Update Software, 2000 59. Kirchner V, Kelley CA, Harvey RJ: Thiori- 76. Blow FC, Barry KL: Older patients with at- 45. Higgins JPT, Flicker L: Lecithin for de- dazine for dementia (Cochrane Review) in risk and problem drinking patterns: new mentia and cognitive impairment Cochrane Library. Oxford, England, Up- developments in brief interventions. Jour- (Cochrane Review) in Cochrane Library. date Software, 2001 nal of Geriatric Psychiatry and Neurology Oxford, England, Update Software, 2000 60. Lonergan E, Luxenberg J, Colford J: 13:115–123, 2000 46. Qizilbash N, Whitehead A, Higgins J, et al: Haloperidol for agitation in dementia 77. Cohen CI, Cohen GD, Blank K, et al: Cholinesterase inhibition for Alzheimer (Cochrane Review) in Cochrane Library. Schizophrenia and older adults: an disease: a meta-analysis of the tacrine trials. Oxford, England, Update Software, 2001 overview: directions for research and poli- JAMA 280:1777–1782, 1998 61. Olin J, Schneider L, Novit A, et al: Hy- cy. American Journal of Geriatric Psychia- 47. Oken BS, Storzbach DM, Kaye JA: The ef- dergine for dementia (Cochrane Review) in try 8:19–28, 2000 ficacy of ginkgo biloba on cognitive func- Cochrane Library. Oxford, England, Up- 78. Soares JC, Gershon S: Therapeutic targets tion in Alzheimer disease. Archives of Neu- date Software, 2001 in late-life psychoses: review of concepts rology 55:1409–1415, 1998 62. Lanctot KL, Best TS, Mittmann N, et al: and critical issues. Schizophrenia Research 48. Patterson CJ, Gauthier S, Bergman H, et al: Efficacy and safety of neuroleptics in be- 27:227–239, 1997 The recognition, assessment, and manage- havioral disorders associated with demen- 79. Lake JT, Rahman AH, Grossberg GT: Di- ment of dementing disorders: conclusions tia. Journal of Clinical Psychiatry 59: agnosis and treatment of psychotic symp- from the Canadian Consensus Conference 550–561, 1998 toms in elderly patients. Drugs and Aging on Dementia. Canadian Medical Associa- 11:170–177, 1997 tion Journal 160(12 suppl):S1–15, 1999 63. Schneider LS, Pollock VE, Lyness SA: A meta-analysis of controlled trials of neu- 80. Maixner SM, Mellow AM, Tandon R: The 49. Small GW, Rabins PV, Barry PP, et al: Di- roleptic treatment in dementia. Journal of efficacy, safety, and tolerability of antipsy- agnosis and treatment of Alzheimer disease the American Geriatrics Society 38: chotics in the elderly. Journal of Clinical and related disorders: consensus statement 553–563, 1990 Psychiatry 60:29–41, 1999 of the American Association for Geriatric Psychiatry, the Alzheimer’s Association, and 64. Herrmann N: Recommendations for the 81. Finkel SI: Antipsychotics: old and new. the American Geriatrics Society. JAMA management of behavioral and psychologi- Clinics in Geriatric Medicine 14:87–100, 278:1363–1371, 1997 cal symptoms of dementia. Canadian Jour- 1998 nal of Neurological Sciences 28(suppl 50. Lopez OL, Becker JT, Wisniewski S, et al: 1):S96–S107, 2001 82. Chan YC, Pariser SF, Neufeld G: Atypical Cholinesterase inhibitor treatment alters antipsychotics in older adults. Pharma- the natural history of Alzheimer’s disease. 65. Alexopoulos GS, Silver JM, Kahn DA, et al: cotherapy 19:811–822, 1999 Journal of Neurology and Neurosurgical The expert consensus guideline series: Psychiatry 72:310–314, 2002 treatment of agitation in older persons with 83. Kumar V: Use of atypical antipsychotic dementia. Postgraduate Medicine 104(Apr agents in geriatric patients: a review. Inter- 51. Feldman H, Gauthier S, Hecker J, et al: A special report):1–88, 1998 national Journal of Geriatric Psychophar- 24-week, randomized, double-blind study macology 1:15–23, 1997 of donepezil in moderate to severe 66. Mayeux R, Sano M: Treatment of Alzheimer’s disease. Neurology 57:613–620, Alzheimer’s disease. New England Journal 84. Sweet RA, Pollock BG: New atypical an- 2001 of Medicine 341:1670–1679, 1999 tipsychotics: experience and utility in the elderly. Drugs and Aging 12:115–127, 1998 52. Erkinjuntti T, Kurz A, Gauthier S, et al: Ef- 67. Cohen-Mansfield J: Nonpharmacologic in- ficacy of galantamine in probable vascular terventions for inappropriate behaviors in 85. Davidson M, Harvey PD, Vervarcke J, et al: dementia and Alzheimer’s disease com- dementia. American Journal of Geriatric A long-term, multicenter, open-label study

1430 PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 of risperidone in elderly patients with psy- of Geriatric Psychiatry 15:687–703, 2000 279, 1994 chosis. International Journal of Geriatric Psychiatry 15:506–514, 2000 101. Bartels SJ, Moak GS, Dums AR: Models 116. Smyer MA, Brannon D, Cohn MD: Im- of mental health services in nursing proving nursing home care through train- 86. Tariot PN, Salzman C, Yeung PP, et al: homes. Psychiatric Services 53:1390– ing and job redesign. Gerontologist 33: Long-term use of quetiapine in elderly pa- 1396, 2002 327–333, 1992 tients with psychotic disorders. Clinical Therapeutics 22:1068–1084, 2000 102. Burns BJ, Taube CA: Mental health serv- 117. Thomson O’Brien MA, Oxman AD, Davis ices in general medical care and nursing DA, et al: Educational outreach visits: ef- 87. McManus D, Arvanitis L, Kowalcyk B: homes, in Mental Health Policy for Older fects on professional practice and health Quetiapine, a novel antipsychotic: experi- Americans: Protecting Minds at Risk. care outcomes (Cochrane Review) in ence in elderly patients with psychotic dis- Edited by Fogel B, Furino A, Gottlieb G. Cochrane Library. Oxford, England, Up- orders. Journal of Clinical Psychiatry Washington, DC, American Psychiatric date Software, 2001 60:292–298, 1999 Press, 1990 118. Dementia. Columbia, Md, American 88. Madhusoodanan S, Brenner R, Suresh P, et 103. George LK, Blazer DG, Winfield-Laird I, Medical Directors Association, 1998 al: Efficacy and tolerability of olanzapine in et al: Psychiatric disorders and mental elderly patients with psychotic disorders: a health service use in later life, in Epi- 119. Pharmacotherapy Companion to Depres- prospective study. Annals of Clinical Psy- demiology and Aging. Edited by Brody sion. Columbia, Md, American Medical chiatry 12:11–18, 2000 JA, Maddox GL. New York, Springer, Directors Association, 1998 1988 120. Bartels SJ, Miles KM, Dums AR: Improv- 89. Verma S, Orengo CA, Kunik ME, et al: Tol- ing the quality of care for older adults erability and effectiveness of atypical an- 104. Bartels SJ, Horn S, Sharkey P, et al: Treat- ment of depression in older primary care with mental disorders: the Outcomes- tipsychotics in male geriatric inpatients. In- patients in health maintenance organiza- Based Treatment Planning System of the ternational Journal of Geriatric Psychiatry tions. International Journal of Psychiatry New Hampshire–Dartmouth Psychiatric 16:223–227, 2001 in Medicine 27:215–231, 1997 Research Center. The Home Care Re- 90. Howanitz E, Pardo M, Smelson DA, et al: search Initiative, Spring 2002, pp 1–6 105. Grimshaw JM, Russell IT: Effect of clini- The efficacy and safety of clozapine versus cal guidelines on medical practice: a sys- 121. Torrey WC, Drake RE, Dixon L, et al: chlorpromazine in geriatric schizophrenia. tematic review of rigorous evaluations. Implementing evidence-based practices Journal of Clinical Psychiatry 60:41–44, Lancet 342:1317–1322, 1993 for persons with severe mental illnesses. 1999 Psychiatric Services 52:45–50, 2001 106. Oxman TE: Effective educational tech- 91. Leucht S, Pitschel-Walz G, Abraham D, et niques for primary care providers: appli- 122. Schneider LS: Starving to Death in a Sea al: Efficacy and extrapyramidal side-effects cation to the management of psychiatric of Data: Perspectives on Translating Geri- of the new antipsychotics olanzapine, que- disorders. International Journal of Psychi- atric Psychiatry Research. Presented at tiapine, risperidone, and sertindole com- atry in Medicine 28:3–9, 1998 the annual meeting of the American Asso- pared to conventional antipsychotics and ciation for Geriatric Psychiatry, Orlando, 107. Callahan CM: Quality improvement re- placebo: a meta-analysis of randomized Fla, February 24–27, 2002 controlled trials. Schizophrenia Research search on late life depression in primary 35:51–68, 1999 care. Medical Care 39:772–784, 2001 123. American Psychiatric Association: Prac- tice guideline for the treatment of pa- 108. Davis DA, Thomson MA, Oxman AD, et 92. Geddes J, Freemantle N, Harrison P, et al: tients with substance use disorders: alco- al: Changing physician performance: a Atypical antipsychotics in the treatment of hol, cocaine, opioids. American Journal of systematic review of the effect of continu- schizophrenia: systematic overview and Psychiatry 152(11 suppl):1–80, 1995 meta-regression analysis. British Medical ing medical education strategies. JAMA Journal 321:1371–1376, 2000 274:700–705, 1995 124. American Psychiatric Association: Prac- tice guideline for the treatment of pa- 109. Soumerai SB: Principles and uses of aca- 93. Jeste DV, Okamoto A, Napolitano J, et al: tients with schizophrenia. American Jour- demic detailing to improve the manage- Low incidence of persistent tardive dyski- nal of Psychiatry 154(4 suppl):1–63, 1997 nesia in elderly patients with dementia ment of psychiatric disorders. Interna- treated with risperidone. American Jour- tional Journal of Psychiatry in Medicine 125. American Psychiatric Association: Prac- nal of Psychiatry 157:1150–1155, 2000 28:81–96, 1998 tice guideline for the treatment of patients with panic disorder. American Jour- 110. Pond C, Mant A, Kehoe L, et al: General 94. Granholm E, McQuaid JR, McClure FS, nal of Psychiatry 155(5 suppl):1–34, 1998 et al: A randomized controlled pilot study practitioner diagnosis of depression and of cognitive behavioral social skills training dementia in the elderly: can academic de- 126. American Psychiatric Association: Prac- for older patients with schizophrenia. tailing make a difference? Family Practice tice guideline for the treatment of pa- Schizophrenia Research 53:167–169, 2002 11:141–147, 1994 tients with major depressive disorder (revision). American Journal of Psychiatry 111. Brody D, Dietrich AJ, deGruy F: The De- 95. Krasucki C, Howard R, Mann A: Anxiety 157(4 suppl):1–45, 2000 and its treatment in the elderly. Interna- pression in Primary Care Tool Kit. Inter- tional Psychogeriatrics 11:25–45, 1999 national Journal of Psychiatry in Medicine 127. American Psychiatric Association: Prac- 30:99–110, 2000 tice guideline for the treatment of patients 96. Stanley MA, Beck JG: Anxiety disorders. with bipolar disorder (revision). American 112. Lin EH, Von Korff M, Katon W, et al: The Clinical Psychology Review 20:731–754, Journal of Psychiatry 159(4 suppl):1–50, role of the primary care physician in pa- 2000 2002 tients’ adherence to antidepressant thera- 97. Sheikh JI, Cassidy EL: Treatment of anxi- py. Medical Care 33:67–74, 1995 128. Shaneyfelt TM, Mayo-Smith MF, Roth- ety disorders in the elderly: issues and wangl J: Are guidelines following guide- 113. Katon W, Von Korff M, Lin E, et al: Col- strategies. Journal of Anxiety Disorders lines? The methodological quality of clin- laborative management to achieve treat- 14:173–190, 2000 ical practice guidelines in the peer-re- ment guidelines: impact on depression in viewed medical literature. JAMA 281: 98. Dada F, Sethi S, Grossberg GT: General- primary care. JAMA 273:1026–1031, 1995 ized anxiety disorder in the elderly. Psy- 1900–1905, 1999 114. Sbordone RJ, Sterman LT: The psycholo- chiatric Clinics of North America 24: 129. Grilli R, Margrini N, Penna A, et al: Prac- gist as a consultant in a nursing home: ef- 155–164, 2001 tice guidelines developed by specialty so- fect on staff morale and turnover. Profes- cieties: the need for a critical appraisal. 99. Stanley MA, Novy DM: Cognitive-behav- sional Psychology: Research and Practice Lancet 355:103–106, 2000 ior therapy for generalized anxiety in late 14:240–250, 1983 life: an evaluative overview. Journal of Anxiety Disorders 14:191–207, 2000 115. Smith M, Mitchell S, Buckwalter KC, et al: Geropsychiatric nursing consultation: a 100. Draper B: The effectiveness of old age valuable resource in rural long term care. psychiatry services. International Journal Archives of Psychiatric Nursing 8:272– PSYCHIATRIC SERVICES o http://psychservices.psychiatryonline.org o November 2002 Vol. 53 No. 11 1431