Treatment of anxiety
Although pharmacological interventions are most frequently employed, no randomized clinical trials have evaluated the use of medication for treating anxiety disorders in persons with dementia. Thus, all recommendations for drug therapy must be cautiously interpreted.
Drug options. Among various options, SSRIs may be useful. One small, prospective, controlled study showed citalopram(Drug information on citalopram) to be superior to placebo for elderly but cognitively intact patients with GAD.24 The high comorbidity of depression with anxiety in dementia patients may further justify a trial of an SSRI.
Benzodiazepines may have an unacceptably high risk of cognitive adverse effects and falls in patients with dementia.
Psychosocial interventions. Psychosocial treatments for anxiety should be considered viable, safe alternatives to medications in patients with dementia although, again, research is limited. Outcome data on cognitive-behavioral therapy (CBT) for anxiety in dementia have shown promising results in several case studies.25 Successful CBT in these patients relies on strategies to circumvent cognitive limitations in learning and applying new coping tools. These include simplifying skill training, repetition, and recruiting collaterals (eg, caregivers) to act as coaches.25
Other non-drug interventions that show promising results in case series or small pilot studies include milieu therapy, addressing patients’ specific environmental needs, and caregiver psychoeducation.26,27
Psychotic symptoms of delusions and hallucinations have been shown to be present in 18% and 14%, respectively, of patients with dementia in a community-based cohort.6 Considerably higher estimates are often quoted in clinical samples, especially in patients with Lewy body dementia.
The psychosis of patients with dementia is typically characterized by persecutory and misidentification delusions.5 The latter include phenomena such as Capgras syndrome (belief that a close relative or friend has been replaced by an impostor) and “phantom boarder syndrome” (belief that strangers are living in the home), that may be associated with agnosia.5 In patients with hallucinations, visual hallucinations are more than twice as common as auditory hallucinations.5 Psychosis in AD is frequently comorbid with other cognitive symptoms (global deficits, anosognosia), affective symptoms (depression, elevated mood), and behavioral symptoms (agitation and overt aggression).28
The presence of psychotic symptoms is typically elicited from a history (from the patient and caregiver) and a mental status examination. The Behavioral Pathology in Alzheimer’s Disease (BEHAVE-AD) scale can aid in assessing behavioral and psychotic symptoms.29 Another instrument is the Dementia Psychosis Scale, rated with the caregiver to quantify the severity and types of delusions.30
Treatment of psychosis
Behavioral and environmental interventions, including redirection and reassurance, are recommended as first-line treatment for psychotic symptoms that do not cause significant danger or distress. Such interventions can also be important adjuncts to psychopharmacological therapy.3,31
Behavioral measures. A large randomized controlled study stressing nonpharmacological management in older adults with AD demonstrated the effectiveness of collaborative care in reducing behavioral symptoms, reflected by significant improvements in neuropsychiatric inventory scores.32 The study’s research protocol included caregiver instructions and handouts to address hallucinations and delusions; these entailed using reassuring touch, establishing a daily routine and adequate light, and avoiding arguments and changes in surroundings.
Pharmacotherapy. When behavioral measures are insufficient, the APA practice guidelines support use of antipsychotic medication for treating psychosis in patients with dementia.3 However, one must exercise a great deal of caution when starting these medications. Increased mortality, particularly from cerebrovascular events, has now been well documented in a large meta-analysis of randomized clinical trials.33 In June 2008, the FDA mandated a black-box warning about increased risk of death in elderly patients with dementia who are taking conventional or atypical antipsychotics.
The multicenter, double-blind, placebo-controlled Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE), which assessed the efficacy of atypical antipsychotics in outpatients with AD, further suggest that the risk of adverse effects may outweigh the benefits. The discontinuation rate because of lack of efficacy was lower with olanzapine and risperidone(Drug information on risperidone) than with placebo.34
Once the decision to initiate pharmacotherapy is made, the choice of medication should be individualized. For instance, patients with Parkinson disease or dementia with Lewy bodies are highly sensitive to extrapyramidal adverse effects. Quetiapine(Drug information on quetiapine) is preferable to higher-potency antipsychotics. In general, among antipsychotics, atypical agents are better tolerated than conventional neuroleptics.31 Still, the therapeutic window can be quite narrow; and it is necessary to closely monitor patients for adverse effects, such as parkinsonism, sedation, anticholinergic effects, delirium, postural hypotension, and increased risk of falls.3
Alternatives to antipsychotic medications include SSRIs. There are data that, at least with AD, serotonergic deficits may contribute to depression and also to psychosis and aggression.35 One short-term, randomized, double-blind, placebo-controlled study demonstrated that citalopram was more effective in diminishing psychosis and agitation than either placebo or perphenazine(Drug information on perphenazine).35
As always, start with the lowest dose and titrate slowly. The goal of pharmacological treatment in this setting should be to reduce behavioral disturbances rather than to eliminate psychosis altogether.31
Among individuals with dementia in the community, 27% exhibit agitation/aggression.6 The prevalence increases as dementia progresses (13% in mild dementia; 24% in moderate dementia; and 29% in severe dementia).6 In persons with dementia, agitation may manifest in a variety of behavioral disturbances, from intermittent psychomotor hyperactivity and disinhibition to physical aggression and combativeness.3 These symptoms often have multiple causes; they may reflect underlying pathophysiology of dementia (eg, serotonergic deficiency) or arise as a result of an inability to communicate needs (eg, hunger) or physical discomfort (eg, pain, constipation).35,36 Agitation may stem from psychosocial stressors, such as a change in living situation, caregiver, or environment.3
Patients who exhibit agitation warrant a medical evaluation to rule out occult medical problems, medication side effects, and delirium—all potential culprits.3
Instruments to quantify agitation/aggression and monitor target behaviors in patients with dementia in-clude BEHAVE-AD and the Cohen-Mansfield Agitation Inventory. The latter is a clinician- or caregiver-rated questionnaire that categorizes disruptive behaviors (either verbal or physical) on a spectrum from nonaggressive to aggressive.37
Therapy for agitation/aggression
The effectiveness and safety concerns of medication use, especially antipsychotics, argue for greater emphasis on nonpharmacological interventions in treating behavioral disturbances.
Behavioral approaches. A systematic “ABC” approach to implementing a behavioral plan helps individualize treatment and monitor improvement. This entails identifying specific Antecedents of target problem Behavior and their Consequences and devising specific strategies to address these.3 For example, repetitive screaming (depending on the antecedent identified) can be treated by fulfilling unmet needs (eg, pain, toileting), providing increased socialization, or reducing overstimulation in the environment.38
In addition to individualized behavior plans, some persons with dementia also benefit from activity or sensory-oriented psychosocial treatments. In a recent review of 5 randomized clinical trials and 14 observational studies of activity interventions, significant decreases in problem behaviors or aggression/agitation were seen with sensory-stimulating and sensory-calming activities, physical activity-based in-terventions (including exercise and walking programs), and recreational activities.36 A review of 3 randomized clinical trials and 21 observational studies of sensory therapy interventions, many of which involved music interventions, also largely favored intervention groups on measures of agitation.36
Pharmacotherapy. The consequences of untreated agitation/aggression can be dire; agitation may lead to excess disability, threats to personal health and the safety of others, increased caregiver burden, and institutionalization.38 Thus, despite the well-established risks of pharmacotherapy in this population, when behavioral measures alone are insufficient, the APA guidelines support use of antipsychotic medication for treating agitation, as with psychosis.3 The atypical antipsychotics risperidone and olanzapine(Drug information on olanzapine) have the best evidence for efficacy.39
Antidepressants (SSRIs and trazodone) have not been well studied for symptoms other than depression, although their relative safety profile may warrant a therapeutic trial, especially for nonpsychotic patients with mild agitation.3 Results from a small, randomized, clinical trial with trazodone showed promising results for decreasing problematic behaviors in patients with frontotemporal dementia.40
There is also evidence of modest, but statistically significant, efficacy of cholinesterase inhibitors.39 There is limited evidence of efficacy for anticonvulsants, lithium(Drug information on lithium), and alpha-blockers. All these agents can cause significant adverse effects and thus are not recommended, except for patients who have not responded to other treatments.3
The use of physical restraints should be restricted to behavioral emergencies if the patient is combative and puts self or others at imminent risk.3 Their use beyond these circumstances may increase risk of falls and contribute to cognitive decline. In fact, restraint reduction has been shown to decrease serious injuries in nursing home residents.41-43
Psychiatric comorbidity in persons with dementia reflects phenomenology and diagnostic treatment challenges that are distinct from those in elderly, cognitively intact individuals with psychiatric illness. To date, large systematic reviews of available pharmacological treatments highlight their lack of efficacy and increased adverse effects.39
Systematic reviews of nonpharmacological interventions fault many studies for lack of methodological rigor; most are single-case designs.44 Given the rising incidence of dementia, the ubiquitous nature of associated neuropsychiatric disturbances, limits of current pharmacological treatments, and modest effect of pharmacological and nonpharmacological interventions, more randomized controlled studies are needed to establish consensus treatment guidelines and improve care for this underserved patient population.
Acknowledgment—This study was supported in part by the Houston VA HSR&D Center of Excellence (Houston Center for Quality of Care and Utilization Studies, HFP90-020).