Goals of Treating Clinical Depression in Older Adults: Better Life, Longer Life
Within this context, the goal of treatment for depression in later life is both better life and longer life (Table 71-1). The clinician should seek to achieve symptomatic remission in order to promote functional recovery and to reduce the risks of a relapsing, recurrent illness course. Treatment should aim to attenuate, to delay, or even to prevent the downstream consequences of depression in older adults (eg, worsening of coexisting medical conditions, complicated grief, dementia, and death). A good treatment plan can also reduce the considerable caregiver burden that attends depression in older family members. Depression is indeed a “contagious” illness, often associated with demoralization, stress, and depression in family caregivers. Finally, the goals of treating depression in old age include facilitating the appropriate use of general medical and social casework services.
TABLE 71-1GOALS OF TREATMENT: BETTER LIFE, LONGER LIFE |Favorite Table|Download (.pdf) TABLE 71-1 GOALS OF TREATMENT: BETTER LIFE, LONGER LIFE
Achieve symptomatic remission
Promote functional recovery
Prevent relapse and recurrence
Attenuate, delay, prevent downstream consequences
Reduce caregiver burden
Facilitate appropriate use of general medical and social casework services
Remission of symptoms is the first goal of treating major depression in older adults because response without full symptomatic remission continues to be associated with disabling symptoms, higher rates of relapse and recurrence, impaired psychosocial functioning, and higher rates of often inappropriate health care utilization. Getting well, however, is not enough; it is staying well that counts. There is a strong rationale for long-term or maintenance treatment of major depressive disorder in older adults. Longer-term treatment to prevent relapse and recurrence may attenuate or prevent at least some, if not many, of the negative health outcomes of major depression in old age, including cognitive and functional impairment. Treatment may promote brain health and may reduce the risk for dementia. Finally appropriate treatment of depression in older adults may prolong life and also reduce the risk for completed suicide.
A recent meta-analysis of response rates to antidepressant pharmacotherapy versus placebo in 10 randomized clinical trials showed an overall drug response rate of 44.4% among drug-randomized patients versus 34.7% among placebo-randomized patients. The odds ratio for response to medication was 1.4 with a 95% confidence interval of 1.24 to 1.57. Thus drug treatment was significantly better than pill placebo but only modestly so, with a number needed to treat (NNT) of 13.
There are clinically actionable predictors of remission to antidepressant pharmacotherapy in depressed older adults. These predictors can be modeled in decision trees and then used to guide clinical decision making, such as how long to persist with a given medication versus switching to another course of action (Figure 71-3). It is useful clinically to measure pretreatment illness severity and change in symptoms during the first 2 weeks of treatment, in order to predict the likelihood of remission of late life major depression treated with medication. Higher baseline depression symptom severity and smaller symptom improvement during the first 2 weeks of treatment are useful predictors of likely nonresponse and nonremission, thereby enabling the clinician to discuss alternative treatment approaches sooner in the course of intervention.
Hierarchy of predictors of treatment remission with an aggressive treatment approach (ie, a threshold of 30% for sensitivity). BSI, Brief Symptom Inventory; MADRS, Montgomery-Asberg Depression Rating Scale.
There is, however, still a paucity of evidence to guide clinical decision making in patients who are not responding well to first-line antidepressant pharmacotherapy. Incomplete or nonresponse is an issue in approximately 50% of older adults with clinical depression. While decision trees can inform the decision to stay the course or to switch to something else, the evidence base supporting second-line treatment is still sparse. Clinical practice typically involves adding a second antidepressant in patients who are responding partially to the first antidepressant. The Food and Drug Administration (FDA) has approved the use of the atypical antipsychotic agent aripiprazole as an augmentation strategy in patients with difficult-to-treat depression. In patients who are showing little evidence of response to the first agent, clinical practice typically involves switching to another class of antidepressant medication, for example, from an SSRI (eg, sertraline) to an SNRI (eg, venlafaxine or duloxetine). Cumulative response rates to stepped-care algorithms of this sort usually approach 80% if patients are willing to remain in treatment. Family support is extremely important in this regard, to support patients in staying the course long enough to derive benefit. The most common error in general medical practice involves inadequate exposure to antidepressant medication—either too low a dose or too little time at maximal dose to achieve full benefit. For the 10% to 20% of patients whose illness is resistant to pharmacotherapy or to the combination of pharmacotherapy and depression-specific psychotherapy, electroconvulsive therapy (ECT) remains a viable option and one that is typically very helpful.
It is also important for the clinician to be mindful of the hazards, as well as the benefits, of acute or short-term antidepressant pharmacotherapy in older adults. This is particularly the case with respect to the sedative and anticholinergic properties of the older tricyclic antidepressants (TCAs), such as amitriptyline, imipramine, and nortriptyline. The use of these agents should probably be reserved to mental health specialists, and they should generally be considered third-line treatment. One advantage of these agents is the possibility to monitor blood levels for both adherence and for estimating the likelihood of efficacy. Thus there is a predictable relationship between TCA blood levels and likelihood of treatment response—a relationship that has not been documented in the case of newer SSRI and SNRI agents. The latter are, however, far safer in overdose than the TCAs. The use of antidepressants of any class is associated with falls and fractures in older adults, though it is difficult to know whether this is a direct consequence of the medication per se or, as seems more likely, the result of depressive illness itself. A small percentage of frail older adults, especially after the age of 75, are vulnerable to hyponatremia as a result of syndrome of inappropriate antidiuretic hormone secretion (SIADH), especially those with pretreatment serum sodium levels of 135 or lower. Also, bradycardia is a risk in patients with pretreatment heart rates of less than 60 and in the presence of β-blockers. Finally, per an FDA Medwatch published in August, 2010, there is a dose-dependent increased risk of arrhythmia associated with the use of citalopram in adults aged 60 and older, including risk for torsade de pointes.
Drug-drug interactions also need to be considered, especially with monoamine oxidase inhibitors and with medications metabolized by the cytochrome P450 (CYP450) family of isozymes. SSRIs and SNRIs are metabolized by the CYP450 complex in the liver (especially 2D6, 2C19, and 2C9). The patient’s metabolizer status (eg, poor or ultrarapid) and the coadministration of other drugs that are metabolized by this complex may influence antidepressant response and the emergence of adverse side effects. CYP polymorphism testing may be helpful in patients who do not show antidepressant response with the usual doses, those with severe side effects at low doses, and those under polypharmacy. The emergence of suicidal thoughts and attempts, usually in the first 2 months of treatment, is an important concern in adolescents and young adults, while in older adults antidepressants are deemed by the FDA to protect against suicide. The coprescription of blood center anticoagulants and antiplatelet agents may pose some risk for patients who were taking SSRI medications. Table 71-2 lists antidepressants recommended for use in older patients, with starting and maximal doses. Usually, short-term or acute treatment lasts on the order of 12 to 16 weeks, but may last 24 to 28 weeks if second- or third-line agents are needed to achieve wellness. The important principles of acute pharmacotherapy in older adults encompass the following “clinical pearls”: (1) start low (with respect to dose), go slow, but go all the way; (2) educate both patients and family caregivers about the rationale for treatment and importance of adherence to achieve full symptomatic remission; (3) if a second augmentation agent is required for the achievement of remission, it is also very likely to be necessary to sustain remission; and (4) the dose that brings about remission is the dose that sustain remission (“the dose that gets you well is the dose that keeps you well”; in other words, dosage should not be decreased once remission is obtained; doing so puts the patient at risk for early relapse.)
TABLE 71-2COMMONLY USED ANTIDEPRESSANT MEDICATIONS FOR LATE-LIFE DEPRESSION |Favorite Table|Download (.pdf) TABLE 71-2 COMMONLY USED ANTIDEPRESSANT MEDICATIONS FOR LATE-LIFE DEPRESSION
|MEDICATION ||STARTING DOSE ||COMMON THERAPEUTIC DOSES ||HALF-LIFE (hour) ||COMMON SIDE EFFECTS ||COMMENTS |
|SSRI (selective serotonin reuptake inhibitors) ||Nausea, dyspepsia, anorexia, tremors, anxiety, insomnia, sexual dysfunction, jitteriness, hyponatremia ||Risk of serotonin syndrome if combined with certain drugs. |
|Fluoxetine ||10 mg ||10–40 mg once daily ||70–80 || ||Very long acting. |
|Sertraline ||25 mg ||50–200 mg once daily ||25–30 ||Loose stools, diarrhea || |
|Citalopram ||10 mg ||20–40 mg once daily ||40–50 || ||Note: FDA Medwatch regarding arrhythmias. |
|Escitalopram ||10 mg ||10–30 mg once daily ||40–50 || || |
|Paroxetine ||10 mg ||20–40 mg once daily ||10–20 ||Dry mouth, drowsiness, fatigue, weight gain ||More anticholinergic side effects. High risk of discontinuation syndrome if drug stopped abruptly. |
|SNRIs (serotonin-norepinephrine reuptake inhibitors) ||Nausea, drowsiness, fatigue, weight gain, hyponatremia, diastolic hypertension at higher doses ||Risk of serotonin syndrome if combined with certain drugs. High risk of discontinuation syndrome if medication stopped abruptly. |
|Venlafaxine XR ||37.5 mg ||75–300 g once daily ||5–9 || || |
|Duloxetine ||20 mg ||20–120 mg once daily ||8–17 || || |
|Other newer antidepressants || |
|Mirtazapine ||15 mg ||15–45 mg at bedtime ||20–40 ||Sedation, increased appetite/weight gain ||No sexual side effects. |
|Bupropion SR ||100 mg ||100–150 mg twice daily ||15 || ||No sexual side effects. Contraindicated in patients with seizures. Not recommended for patients with comorbid anxiety. |
|TCAs (tricyclic antidepressants) ||Sedation, weight gain, dry mouth, urinary retention, constipation, blurry vision, orthostatic hypotension, impairment of cardiac conduction ||High risk in overdose: 10 days of typical daily dose may result in a fatal cardiac arrhythmia. Get baseline ECG and follow-up ECG at steady state or if new cardiac symptoms occur. |
|Nortriptyline ||10 mg ||50–125 mg every night ||18–56 ||Fatigue ||Therapeutic blood levels range from 50 to 150 ng/mL. |
|Desipramine ||25 mg ||100–200 mg once daily ||12–28 ||Insomnia, agitation || |
The adverse effects of long-term antidepressant pharmacotherapy in older adults are relatively few but should be reviewed with patients and family caregivers as part of shared decision making. Some epidemiologic studies suggest higher risk for osteoporosis, especially in women, but it is not clear whether this is the product of antidepressant medication or is related to the underlying depressive illness and inactivity. The use of venlafaxine over 12 weeks has been associated with increased markers of bone resorption and reduced bone formation in older adults. With tricyclic antidepressants, patients must be monitored for orthostatic hypotension and cardiac arrhythmia as well as worsening of cognitive performance, typically due to anticholinergicity. Finally there is an increased risk for falls with patients on antidepressant medication. Tolerability is an important issue for the long-term use of nortriptyline because of side effects such as daytime somnolence, dry mouth, constipation, urinary retention, and all anticholinergic side effects. Given that SSRIs are safer in overdose than the older TCAs, the use of TCAs should be reserved for difficult-to-treat cases and is probably best conducted by a mental health specialist experienced in their use. It is appropriate to monitor blood levels of TCAs as a gauge of adherence and of adequate exposure, given that there are threshold blood levels required to maximize the probability of satisfactory response (eg, 80–120 ng/mL in nortriptyline-treated patients).
Most of the available studies have not included the oldest old, that is, those 85 and older. This is the most rapidly growing segment of the population and there is no evidence-based information to inform maintenance treatment in this group. Many of these patients have coexisting depression and dementia; the weight of the evidence suggests that antidepressant pharmacotherapy is no more effective than pill placebo in alleviating the suffering of such patients, but that supportive care can be effective in alleviating their distress. Finally, we lack data on biomarkers and pharmacogenetic information that can help to individualize treatment with respect to the best drug, dose regimen, or treatment length of antidepressant pharmacotherapy in individual patients.
Because of the complexity of medical decision making required to manage older adults with bipolar depression and those with psychotic major depression, and because of the substantial risk for suicide, the treatment of such patients is left most appropriately to psychiatrists, especially those with added qualifications in the care of older adults with complex medical, neurologic, and psychosocial comorbidities.
Continuing and maintenance treatment of depression
Continuing and maintenance treatment of depression in older adults is important for the purpose of preventing relapse and recurrence, as well as other “downstream” consequences of depression. A systematic review and meta-analysis of double-blind, randomized controlled trials with antidepressant medication showed a NNT of 3.4 for active pharmacotherapy versus placebo (a considerably more favorable NNT than the 13 documented for acute, short-term, placebo-controlled trials). Placebo-controlled studies conducted by the author’s group have shown the benefits of long-term (3-year) nortriptyline and monthly interpersonal psychotherapy in older adults with long-standing histories of recurrent major depression (Figure 71-4). In a second trial involving the use of the antidepressant paroxetine, the relative risk of recurrence among patients receiving placebo was 2.4 times greater than among patients randomly assigned to paroxetine (Figure 71-5). In this study patients with a greater number and more severe concomitant medical illnesses (as indicated by a score of 10 or more on the Cumulative Illness Rating Scale for Geriatrics) had higher rates of recurrent depression and did not fare as well during 2-year maintenance treatment with paroxetine, as compared with patients having fewer and less severe concomitant medical illnesses (Figure 71-6). A third study from the author’s group addressed maintenance treatment of depression in old age with a specific focus on coexisting cognitive impairment. While cognitive impairment is a core feature of major depression in older adults, there has been relatively little information concerning the value of adding cholinesterase inhibitors (ChEIs) such as donepezil to maintenance antidepressant pharmacotherapy. Since depression is a risk factor for both mild cognitive impairment (MCI) and for progression of MCI to dementia, it is clinically important to assess the place of ChEI pharmacotherapy in depression’s long-term (maintenance) management. The study found that relative to placebo plus antidepressant, the combination of donepezil (5–10 mg/day) plus antidepressant temporarily improved global cognition at 1 year, but effect sizes were small. Of 57 participants with MCI, 10% on donepezil and 33% on placebo converted from MCI to dementia (primarily of the Alzheimer type) over the course of 2 years of maintenance pharmacotherapy, suggesting a protective effect for the cholinesterase inhibitor when added to antidepressant medication. By contrast, of 73 cognitively normal subjects, 16% on donepezil experienced cognitive decline and 22% on placebo showed cognitive decline, suggesting no protective effect for patients initially with normal cognition. Although donepezil showed some benefit in enhancing performance on cognitive instrumental activities of daily living (IADLs), its use was also associated with a greater incidence of recurrent depressive episodes (which were managed relatively easily by dosage adjustment of antidepressant medications such as escitalopram, venlafaxine, or duloxetine, usually to higher doses).
Recurrence Rates of Major Depressive Episodes. Survival function of 4 treatment groups (log-rank statistic = 34.31; df = 3; P=.001). On pairwise analysis, each of the 3 active treatment groups was significantly better than placebo. IPT indicates interpersonal psychotherapy. (Reproduced with permission from Reynolds CF 3rd, Frank E, Perel JM, et al. Nortriptyline and interpersonal psychotherapy as maintenance therapies for recurrent major depression: a randomized controlled trial in patients older than 59 years. JAMA. 1999;281(1):39–45. Copyright © 1999 American Medical Association. All rights reserved.)
Time from Randomization to Recurrence. The relative risk of recurrence among patients receiving placebo was 2.4 times that among patients receiving paroxetine (p = 0.02; chi-square statistic = 9.77 with 2 df). No effect of maintenance psychotherapy on recurrence was detected. Kaplan–Meier survival analysis with log-rank chi-square statistics was used to test for overall differences in recurrence rates among the groups. P values were based on the log-rank chi-square test. (reprinted with permission of the Massachusetts Medical Society). (Reynolds CF 3rd, Dew MA, Pollock BG, et al. Maintenance treatment of major depression in old age. N Engl J Med. 2006;354(11):1130–1138.)
Effect of the Number and Severity of Concomitant Medical Illnesses on the Efficacy of Maintenance Therapy with Paroxetine. Patients with a greater number of and more severe concomitant medical illnesses, as indicated by scores of 10 or more on the Cumulative Illness Rating Scale for Geriatrics (CIRS-G), had higher rates of recurrent depression and did not fare as well during treatment with paroxetine as those with fewer and less severe concomitant medical illnesses. Although both paroxetine use and the score on the CIRS-G affected risk (main or direct effect, p = 0.004), paroxetine was more effective in preventing recurrence in patients with fewer and less severe concomitant medical illnesses (interaction effect, p = 0.03). Kaplan–Meier survival analysis with log-rank chi-square statistics was used to test for overall differences in recurrence rates among the groups. P values were based on the log-rank chi-square test. (reprinted with permission of the Massachusetts Medical Society). (Reynolds CF 3rd, Dew MA, Pollock BG, et al. Maintenance treatment of major depression in old age. N Engl J Med. 2006;354(11):1130–1138.)
Long-Term Effect of Evidence-Based Depression Care Management on Mortality in Older Adults
The primary question here is whether an evidence-based depression care management program reduces mortality for depressed older adults (nonbipolar, nonpsychotic) in primary care. The summary answer is that a depression care manager working with primary care physicians to provide evidence- and algorithm-based care for depression can indeed save lives. Prospective studies have consistently shown a relation between depression and increased mortality in older adults, but until recently no randomized trials have reported whether treatment can actually prolong life. Using data from Prevention of Suicide in Primary Care Elderly Collaborative Trial (PROSPECT), investigators recently observed that after 8 years of follow-up, relative mortality risk was reduced by 24% among participants who had received evidence-based care, as compared with patients who had received usual care. PROSPECT involved 20 primary care practices in New York City, Philadelphia, and Pittsburgh, enrolled in a cluster randomized trial. Practices were randomly assigned to either the intervention arm of PROSPECT or to enhanced usual care. PROSPECT’s intervention consisted of guideline-based treatment involving both antidepressants and, in about one-third of participants, the use of interpersonal psychotherapy (Figure 71-7). A depression care manager, typically a social worker or mental health nurse, was placed in primary care practices to monitor symptoms, side effects, and treatment adherence. The depression care manager also educated family caregivers and assisted in medication management by the primary care physician through a process of consultation with primary care and with attending geriatric psychiatrists. Participants were followed for up to 2 years through acute, continuation, and maintenance phases of treatment. After 1 year patients with major depression in the intervention condition were more likely to have experienced remission of major depression as measured by at least a 50% improvement in symptoms and reductions in suicidal ideation. The benefits on depression persisted for at least 24 months. On 8-year follow-up, the hazard ratio for mortality relative to nondepressed control patients in the same practices was 1.9 in usual care practices, but only 1.09 in intervention practices as compared with nondepressed controls in those practices (Figure 71-8). Examination of specific causes of death indicated that reductions in mortality on the order of 24% at 8-year follow-up reflected reduced number of deaths due to cancer among patients in PROSPECT’s intervention practices (Figure 71-9).
Survival probability among person with no depression (red line) or major depression (green line) in practices randomized to usual care or to the intervention. (Gallo JJ, Morales KH, Bogner HR, et al. Long-term effect of depression care management on mortality in older adults: follow-up of cluster randomized clinical trial in primary care. BMJ. 2013;346:f2570.)
Adjusted hazard ratios for specific causes of death comparing major depression to no depression within intervention (green) or usual care (red). (Gallo JJ, Morales KH, Bogner HR, et al. Long-term effect of depression care management on mortality in older adults: follow-up of cluster randomized clinical trial in primary care. BMJ. 2013;346:f2570.)
Collaborative care, algorithm-based models
Over the past 15 years, the most public health–relevant advance in the treatment of depression in older adults has been the development of evidence-based, collaborative care interventions in the setting of primary care. The development of these models was stimulated by the fact that more than half of mental health treatment for depression in later life is delivered in primary care settings. Depression like other chronic diseases with a relapsing and recurring course responds favorably to collaborative care treatment models. Such models include a structured management plan, together with a multiprofessional approach to patient care including case managers such as nurses, social workers, or psychologists. One study, conducted in Goa, India, found the use of lay health counselors, working under the supervision of primary care physicians and mental health specialists, to be effective for the treatment of common mental disorders, such as depression and anxiety, in mixed-aged adults, including older adults. Some of the most influential models of collaborative care in the primary care management of late-life depression include Improving Mood-Promoting Access to Collaborative Treatment (IMPACT), PROSPECT, and MANAS (from the Konkani language meaning “promoting mental health”), a trial that utilized lay health counselors in the implementation of a simple, behaviorally oriented, stepped care algorithm. Collaborative care interventions such as IMPACT and PROSPECT have been shown to be effective at treating depression and bringing about response and remission. Of these studies, PROSPECT has also been shown to reduce mortality over 8-year follow-up among those who had been in the intervention arm of the study, who were 24% less likely to have died than were participants in practices that had been randomly assigned to usual care. Collaborative care models have been successfully implemented into nonresearch practice settings. A meta-analysis reviewing 37 trials of depression collaborative care found that antidepressant adherence rates were twice those of usual care and favorable treatment outcomes persisted for up to 5 years. Other studies have applied collaborative care approaches to the treatment of depression in the context of chronic medical diseases such as diabetes, cancer, stroke, and postcoronary artery bypass grafting surgery.
Collaborative care interventions for depression in old age have better response rates than usual care, even when usual care has been enhanced by regular feedback to primary care physicians. The advantages of collaborative care interventions over usual care generally have modest but clinically meaningful effect sizes of around 0.3 or 0.4. This finding underscores the need for consultation with or referral to mental health specialists for patients who have not responded satisfactorily to first-line treatment.
Managing depression in older age is also strongly supported by the evidence of efficacy for psychological interventions. A recent meta-analysis of psychotherapy of depression in old age identified 44 studies comparing psychotherapies to control groups to other therapies or to pharmacotherapy. The overall effect size indicating the difference between psychotherapy and control groups was 0.64, corresponding to a NNT of 3. These effects were maintained at 6 months or longer after randomization. Certain types of psychotherapy were found to be particularly effective. These included cognitive behavioral therapy (CBT), life review therapy, and problem-solving therapy (PST). Comparison between different types of psychotherapy suggested that learning-based approaches such as CBT and PST may be more effective than nondirective counseling. Learning-based approaches encourage the development of active coping and self-management skills, the use of behavioral techniques to improve sleep, and engagement in pleasant activities. Another systematic review of studies that used medication, psychotherapy, or both found that older adults with clinical depression derive significantly greater benefit from combined therapy compared to pharmacotherapy alone or psychotherapy alone. Table 71-3 shows the component steps for PST in primary care (PS-PC).
TABLE 71-3SEVEN STEPS OF PROBLEM-SOLVING THERAPY (PST) |Favorite Table|Download (.pdf) TABLE 71-3 SEVEN STEPS OF PROBLEM-SOLVING THERAPY (PST)
Clarify and define the problem.
Establish an objective, achievable goal.
Brainstorming—develop solution alternatives.
Follow decision guidelines—pros and cons.
Choose the preferred solution.
Action plan: implement the solution.
Evaluate the outcome.
Restart if needed.
Psychosocial interventions in older adults with major depression should involve interdisciplinary collaboration beyond mental health. These may include primary care, specialty care, rehabilitation, and homecare. Also, psychosocial interventions should involve family members and caregivers whenever possible, to increase retention and adherence and to lower family caregiving burden. The clinician should plan for at least 6 weekly sessions in order to ensure adequate exposure to an evidence-based treatment such as PST for depression. Effective first-line treatments include behavioral activation, particularly encouraging patients and family caregivers to engage in pleasurable activities. Behavioral activation is a core part of both PST and CBT with or without antidepressant medications. For older adults with anxiety disorders, current evidence indicates first-line treatment is the use of a SSRI. For those who resist taking medications, first-line behavioral interventions include relaxation training together with motivational interviewing to encourage a medication trial. In older adults with co-occurring depression and anxiety, the depression should be treated first.