Primary Psychiatry. 2010;16(9 suppl):1-12
Release Date: September 3, 2010
Expiration Date: September 3, 2011
Provided by the American Association for Geriatric Psychiatry (AAGP)
This activity is supported by an unrestricted educational grant from Lilly USA, LLC.
The goal of this activity is to discuss best practices in the assessment, diagnosis, and ongoing management of depression in older adults.
This activity is intended for psychiatrists who treat older patients with depression.
Statement of Need
Major depressive disorder (MDD) is a serious medical illness affecting approximately 15 million adults in the United States, including 1% to 5% of older community-dwelling persons and over 13% of older adults requiring home healthcare.1 Unlike the normal experience of sadness, depression is persistent and can significantly interfere with thought processes, physical health, mood, activity, and behavior; it is the leading cause of disability in the United States.2,3
There are numerous clinical challenges to the successful diagnosis and treatment of older adults with MDD. The course of MDD in older adults is distinct from that in younger patients, for example, which may confound clinicians. Further, age-related changes in drug metabolism necessitate careful selection and adjustment of pharmacologic therapy, and older adults are often resistant to initiating either psychological or medical therapies. The presence of multiple medical comorbidities—ranging from cardiovascular disease to diabetes and cancer—may complicate treatment as well.4 Hence, healthcare providers need to improve their understanding of key issues in the diagnosis and treatment of older adults with depression, and tailor therapies accordingly.
Depression in older adults is often overlooked. A recent observational study of visits by elderly patients to their primary care physicians revealed that discussions about mental health averaged 2 minutes in 22% of all office visits despite the fact that 50% of patients met the criteria for depression.5 The study noted wide variations in physician effort to provide care and few referrals to mental-health specialists—even for severely depressed and suicidal patients.5 Available evidence suggests that late-life depression in particular is often undetected or undertreated.6 Reasons for underdiagnosis in this population include emphasis on somatic rather than cognitive and mood complaints, reluctance to stigmatize patients with psychiatric diagnosis, mild or nonspecific symptoms of depression, fear of medication side effects, the incorrect notion that reactive depressions are not pathological, and time limitations.7 The proposed activity addresses both pharmacologic and nonpharmacologic strategies for individualizing assessment, diagnosis, treatment, and ongoing care in this patient population to meet the clinical challenges and to assist in the clinical decision making.
At the conclusion of this program, participants will be better prepared to:
• Conduct initial and ongoing assessment of older patients with depression, identify comorbidities, review current and prior medical history, and set treatment goals
• Diagnose major depressive disorder in older adults, differentiating it from dysthymia, bipolar disorder, and dementia
• Design and implement an appropriate multimodal treatment plan for an older patient with depression
• Select and appropriately dose and monitor an antidepressant for an older patient with depression, taking into account the patient’s other medical disorders, preexisting pharmacologic regimen, and cognitive symptoms
• Implement psychosocial interventions shown to improve treatment outcomes in older patients with depression
1. Hybels CF, Blazer DG. Clin Geriatr Med. 2003;19:663-696, v;
2. NAMI. Major depression fact sheet. http://www.nami.org/Template.cfm?Section=Depression&Template=/ContentManagement/ContentDisplay.cfm&ContentID=88956. Accessed August 26, 2010;
3. Omnicare. Geriatric Pharmaceutical Care Guideline: Depression. Covington, KY: Omnicare, Inc.; 2009;
4. Prabhakaran P, Butler R. J Affect Disord. 2002;70:319-322;
5. Tai-Seale M, et al. J Am Geriatr Soc. 2007;55:1903-1911;
6. Unutzer J, et al. J Am Geriatr Soc. 2003;51:1903-1911;
7. Piven M. Detection of Depression in the Cognitively Intact Older Adult. Iowa city, IA: University of Iowa Gerontological Nursing Interventions Research Center, Research Dissemination Core; 2005.
This activity is provided by the American Association for Geriatric Psychiatry, facilitated by Asante Communications, LLC, and supported by an educational grant from Lilly USA, LLC. Activity content was developed independently by the faculty. The opinions or views expressed in this CME activity are those of the presenters and do not necessarily reflect the opinions or recommendations of the sponsor or commercial supporter.
Method of Participation
There are no fees for participating in and receiving credit for this activity. Participants should complete the activity and mail the multiple-choice posttest answer section and activity evaluation to Janice Allen, American Association for Geriatric Psychiatry, 7910 Woodmont Avenue, Suite 1050, Bethesda, MD 20814, or fax to (301) 654-4137. To access these forms, please click on the following URL, https://www.aagponline.org/Uploads/CME/CE-MDD-101_postest_R1.pdf. A score of at least 80% is required to successfully complete this activity. Certificates will be mailed approximately 6 to 8 weeks after receipt of mailed/faxed submissions and verification of a passing grade. Credit is available through September 2, 2011.
Statement of Disclosure and Independence
The opinions and recommendations expressed by faculty and other experts whose input is included in this program are their own. This enduring material is produced for educational purposes only. Use of the American Association for Geriatric Psychiatry or Asante Communications, LLC, name implies review of educational format design and approach. Please review the complete prescribing information for specific drugs or combinations of drugs, including indications, contraindications, warnings, and adverse events before administering pharmacologic therapy to patients.
Faculty Disclosure Statement
CME-certified activities must be unbiased and based upon scientific evidence. To help participants make judgments about the presence of bias, faculty are required to disclose to the activity participant any real or apparent conflict(s) of interest that may have a direct bearing on the subject matter of the educational activity. Any relationships faculty members may have with commercial entities have been disclosed and reviewed, and any potential conflicts have been resolved.
The content contained within this monograph, in part, has been based upon transcripts from a live symposium held on Monday, March 8, 2010, in Savannah, Georgia. This live dinner symposium was held in conjunction with the American Association for Geriatric Psychiatry’s 2010 Annual Meeting. The below disclosures represent those received from the faculty who participated in the live meeting and faculty/author of the Enduring Monograph.
Faculty Disclosures and Relevant Financial Relationships—AAGP Live Symposium, Monday, March 8, 2010
Yeates Conwell, MD
Dr Conwell has no actual or potential conflict of interest in relation to this educational activity or presentation.
Mark D. Miller, MD
Dr Miller has no actual or potential conflict of interest in relation to this educational activity or presentation.
Steven P. Roose, MD
Forest Laboratories (research support).
Faculty/Author Disclosures and Relevant Financial Relationships—Enduring Monograph
Marc E. Agronin, MD
Dr Agronin has no actual or potential conflict of interest in relation to this educational activity or presentation.
Steven P. Roose, MD
Forest Laboratories (research support).
The American Association for Geriatric Psychiatry is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.
The American Association for Geriatric Psychiatry designates this educational activity for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should only claim credit commensurate with the extent of their participation in the activity.
Copyright © 2010 American Association for Geriatric Psychiatry (AAGP) and Asante Communications, LLC. All rights reserved. No part of this monograph may be used or reproduced in any manner whatsoever without written permission except in the case of brief quotations embodied in articles or reviews. The AAGP owns the copyright for this material or has obtained permission to distribute this material from all authors.
Major Depressive Disorder (MDD) in Older Adults: A Misunderstood Condition
Myths about late-life depression abound, especially the belief that this illness is a normal consequence of aging and is therefore untreatable. Some would imagine that the increasing burden of personal and social losses, physical frailty, medical illnesses, and anxiety over approaching mortality predisposes older adults to becoming depressed and staying that way. In his 1995 Atlantic Monthly article on depression in late life, psychotherapist Stanley Jacobson wrote, “Oldness itself is reason to be sad if you dwell on it.”1 Jacobson further commented on the prevalence of depression among nursing-home residents as an almost inevitable consequence of being “immersed in infirmity and impending death.” This particularly nihilistic belief has troubling implications, suggesting that treatment must address the untreatable, namely the aging process itself; thus, short of discovering the fountain of life, treatment of depression in older adults is futile.
The myth that major depressive disorder (MDD) in older adults is less responsive to treatment than in younger cohorts may reflect the stigma associated with aging and mental illness, or the seemingly increased variability in treatment response with increasing age.2, 3 In fact, MDD in older adults is a serious disorder that is neither an inevitable nor intractable consequence of age. Further, patients with late-life depression are no less likely to respond to treatment than middle-aged patients with depression.4
MDD in Older Adults: Epidemiology
While the prevalence of MDD in the community-dwelling elderly is estimated at less than 5%,5 broader epidemiologic surveys within older populations who present with significant medical and psychiatric comorbidities provide a more complete view. In outpatient clinics, for instance, MDD has been observed in 5% to 12% of older medically ill individuals.6 While depression has been diagnosed in 44% of elderly patients admitted to hospitals, the prevalence may be even higher in nursing homes.5 These and other elderly patients may meet many, though not all, diagnostic criteria for MDD; nevertheless, they remain depressed and functionally impaired.5 If diagnosed with minor or subsyndromal depression, the other subtypes of late-life depression, the consequences are measurable and often indistinct from patients with MDD. A recent study by Grabovich and colleagues, for example, found that older primary care patients with subsyndromal depression had poorer one-year outcomes across multiple psychiatric symptoms and functional domains compared with nondepressed elders. Importantly, the findings were not always significantly different from those with actual MDD.6
MDD in older adults is a serious disorder that is neither an inevitable nor intractable consequence of age.
Burden of MDD and its Underdiagnosis/Undertreatment in Older Adults
Although late-life depression is an important public health concern because of its high morbidity and mortality, it is often underdiagnosed and undertreated.7 Reflecting on his own episode of depression, novelist William Styron described the pain associated with depression as “. . . quite unimaginable to those who have not suffered it, and it kills in many instances because its anguish can no longer be borne.”8 According to Styron, the sufferer’s inability to imagine future relief differentiates depression from other forms of suffering. “It is hopelessness even more than pain that crushes the soul,” he elaborated. Most older individuals who have suffered from depression demonstrate symptoms in line with Styron’s description—that is, they experience a disorder that causes “. . . functions of the body and nearly all of those of instinct and intellect to slowly disconnect . . .” Depression disrupts all aspects of day-to-day life, significantly impairing alertness, concentration, appetite, and sleep. Attributing the devastating impact of these changes to age is of little clinical value and may in fact unintentionally delay timely diagnosis and treatment.
The loss of a loved one is one of the most traumatic events in a person’s life, a misfortune especially affecting the elderly population. By age 65, more than 50% of women and 10% of men in the United States have been widowed at least once.9 Mirroring some of the misperceptions associated with MDD in the elderly, many clinicians do not consider bereavement-related depression to be as serious as MDD.10 Recent studies, however, are prompting these clinicians to question their long-held assumptions that have led to the exclusion of bereavement-related sadness as a diagnosis for MDD. In their population-based evaluation, Kendler and colleagues, for instance, reported more similarities than differences in patients with bereavement-related depression versus stressful life event associated depression.11 Severity and duration of depressive episodes in bereaved patients were comparable to those observed in depressed patients who had recently experienced divorce/separation, illness, or job loss. The two subpopulations shared similar rates of suicidal ideation, number of prior depressive episodes, and levels of risk for future depressive episodes. Interestingly, patients with bereavement-associated depression were less likely to seek treatment than were those with depression related to other stressful life events. Taken together, current data suggest that bereavement-associated depression may not be significantly different from MDD, both in its severity and in the degree to which it is undertreated.
Comprehensive and Continual Assessment
According to the Diagnostic and Statistical Manual of Mental Disorders IV, Text Revision (DSM-IV-TR), the diagnosis of MDD is considered for mood disorders that do not demonstrate evidence of manic, hypomanic, or mixed episodes12 and cannot be better explained by schizoaffective disorder, schizophrenia, schizophreniform disorder, delusional disorder, or psychotic disorder not otherwise specified. A number of scales, such as the Patient Health Questionnaire 9-symptom checklist (PHQ-9)13 (Figure 1), the Geriatric Depression Scale (GDS) short form,14, 15 the Hamilton Depression Rating Scale (HAM-D or also called HDRS),16 the Zung Self-Rating Depression Scale (SDS),17 the Montgomery-Asberg Depression Rating Scale (MADRS),18 and the Cornell Scale for Depression in Dementia (CSDD),19 are available to assess the severity of a patient’s depression. While a comprehensive discussion of all these depression-rating tools is outside the scope of this review, we will focus on two commonly used scales—PHQ-9 and GDS—that have demonstrated validity and reliability in the geriatric population.14, 15, 20 The 9 items included in the PHQ-9 are based on the DSM-IV-TR diagnostic criteria for MDD13 and are listed in Figure 1. The GDS14, 15 short form is a 15-item scale with a yes/no format as illustrated in Figure 2. It was designed to address specific problems such as fatigue and poor concentration in frail elderly patients.
Figure 1 (click to enlarge)
Figure 2 (click to enlarge)
As part of a comprehensive assessment, several unique features of depression in late life should be noted. Instead of a depressed mood, symptoms of irritability, anger, or anxiety may predominate. There may be unexplained agitation, resistance to care, isolative behaviors, or an excessive focus on pain or somatic complaints. Acute cognitive impairment may also be observed.12, 21 A history of progressive cognitive decline indicates dementia rather than MDD.12 Older individuals in rehabilitation may demonstrate slow or absent progress with physical therapy as a result of depression. Failure to thrive may occur in more severely depressed individuals and is characterized by complete loss of appetite or feeding behaviors.22 The potential for suicidal ideation must always be assessed carefully, especially in elderly men who have a suicide rate over 7 times that of similarly aged women.23 Important risk factors for suicide in late life include previous suicide attempts, family history of suicide, depression, substance abuse, widowhood, physical illness, pain, and disability.24, 25 Suicide risk may also be expressed passively or indirectly, such as with a patient who refuses life-sustaining tests, treatments, or nutrition.
Differential Diagnosis of MDD in Older Adults
Differential diagnosis necessarily involves casting a wide net, covering age-related psychological, social, and physical factors. The clinician must document and evaluate all medical conditions and medications (prescribed and over-the-counter) for their potential role in causing or exacerbating depressive symptoms. For example, Alzheimer’s disease, chronic pain, sleep disorders, congestive heart failure, hypothyroidism, Parkinson’s disease, and stroke are associated with a significantly increased risk of depression. However, determining causality is challenging, since common symptoms such as fatigue, insomnia, and confusion can confound differentiation of depression from medical illness.26 Substance use disorders, particularly alcohol abuse, may also play a key role in depression.27 And medications such as beta-blockers, corticosteroids, and narcotics increase the risk of depression as well.
Severe depression may be related to pseudodementia (originally called reversible dementia), a syndrome in which depressive symptomatology produces cognitive impairment, causing significant disability and placing patients at risk of subsequent development of an actual dementia.28 MDD and dysthymic disorder, another late-life depression subtype, are differentiated based on symptom pattern and severity. In MDD, symptoms are present nearly every day for most of the day; whereas in dysthymic disorder, symptoms are less frequent and less severe, but the history may go back 2 or more years.12
Comorbidities in Older Adults With MDD
Common comorbidities in elderly depressed patients include several major medical conditions such as stroke, cancer, coronary heart disease, sleep disorders, HIV/AIDS, diabetes, Alzheimer’s disease, and Parkinson’s disease.5, 29 The relationship between depression and physical illness is complex and reciprocal; when depression coexists with physical illness, each exacerbates the other, and symptoms of depression and the physical ailment are both more severe.29 Depression in late life is an independent predictor of hospital and outpatient services.5 Importantly, major depression is also a leading cause of disability in adults and is a risk factor for physical disability and death in elderly patients with medical illness.5
Laboratory testing can uncover hormonal or nutritional imbalances that may mimic or exacerbate depression, as well as those that may arise as a result of depression or antidepressant treatment. Reduced levels of thyroid-stimulating hormone (TSH) in men is associated with depression, warranting routine monitoring.30 Interestingly, this association was not observed in women, presumably because of a greater use of thyroid medications and antidepressants in female patients.30 Depression may also be caused by deficiency in vitamin B12 or folate, or by excess levels of homocysteine.31 Vitamin B12 supplementation should be considered if levels fall below 400 pmol/L. Folate deficiency, detected under a general nutritional assessment, should be treated as necessary. Vitamin B12 and folate are necessary for maintaining optimal homocysteine levels. A complete blood count with white cell differential is also indicated. In particular, iron level/mean corpuscular volume is important, as microcytosis caused by iron deficiency has been identified as a causative factor in depression, especially when folate deficiency is present.32 Cognitive decline, a common comorbidity in depression, is associated with hypercalcemia among patients older than 75 years in some studies.33 While this result still needs to be confirmed by further research, it is advisable to test calcium levels in elderly patients with depression or decreased cognition, especially in those who are taking calcium supplementation for bone loss.
Cerebral Abnormalities in Older Adults
Age-related cerebral abnormalities are common in older adults and predict a poor treatment response. Vascular damage caused by multiple cerebral infarcts, either from large cortical strokes or smaller subcortical damage, are associated with degenerative changes in the brain that may lead to vascular depression.34, 35 In many respects this depression subtype resembles a subcortical dementia characterized by depression, apathy, cognitive decline, executive dysfunction, and psychomotor retardation.34, 35 Vascular depression is associated with poorer outcomes and increased risk of relapse.5 Importantly, the degenerative changes in vascular depression may explain the decreased efficacy of antidepressants seen in some trials with elderly patients.3, 34
The clinical construct of vascular depression was initially concept-ualized when magnetic resonance imaging studies demonstrated higher incidence of cerebral white matter hyperintensities in patients with late-onset compared with early-onset depression. Of note, the cerebral changes were identified in mood-relevant neural circuits.36-38 Ischemia appears to play a causal if poorly understood role. As observed in one study, white matter hyperintensities were exclusively ischemic in older adults with depression; whereas only approximately a third were ischemic in those without depression.38
Patients with vascular depression may have impaired executive function. Geriatric patients with executive dysfunction exhibit deficits in planning, organizing, abstracting, and initiating appropriate actions.39 In their 16-week follow-up of older patients who achieved remission from major depression following nortriptyline treatment, Alexopoulos and coworkers reported that impaired executive function (as measured by deficits in initiation and perseveration) predicted relapse (Figure 3), while memory impairments and age did not.39 These findings are consistent with studies showing that patients with abnormal measures of prefrontal dysfunction such as low initiation/perseveration scores, psychomotor retardation (impaired thought or movement), and longer P300 latency* are more likely to have a poor or delayed response to antidepressants.39, 40 Clinicians may need to monitor patients who are depressed and who have executive dysfunction more closely than patients without this impairment. Rehabilitative techniques can reduce the impact of executive dysfunction on daily life, but whether these techniques decrease the susceptibility to depression relapse remains an open question.39 Investigational treatment modalities for this patient subpopulation include repetitive transcranial magnetic stimulation and vestibular rehabilitation treatment (with or without cholinergic stimulation).41, 42
* The P300 wave is generated when performing tasks involving sustained attention, and its production depends on the integrity of the prefrontal system. Long latency of the P300 wave is a measure of prefrontal dysfunction and executive impairment.
Vascular damage caused by multiple cerebral infarcts, either from large cortical strokes or smaller subcortical damage, are associated with degenerative changes in the brain that may lead to vascular depression.
Figure 3 (click to enlarge)
Treatment: General Considerations
A treatment plan should include a supportive multidisciplinary team to identify and address specific causative or exacerbating factors.43 For example, the patient’s primary care physician can address medical symptoms while a social worker can address major psychosocial issues. Family members and other caregivers are critical to understanding the background and course of the patient and can assist with medication adherence, transportation to appointments, and social support. An interdisciplinary collaborative ensemble working with older adults will provide comprehensive insights into their depression and thus help identify the most salient target symptoms.43 Collaborative care, in which a multidisciplinary team of healthcare professionals and nonmedical specialists work together to bring best-practices depression treatments in primary care, has well-established efficacy in improving outcomes in geriatric depression.44
Clearly, the physician must work with the patient to formulate individualized treatment strategies and help the patient take an active role in managing his or her depression.43, 45
Strategies to Accomplish Treatment Goals
Establishing achievable treatment goals in elderly patients with MDD is essential. General considerations for identifying goals include supporting and encouraging the patient throughout the current episode45; preventing recurrence, once depression has subsided5; and achieving remission.45, 46 Specific treatment plans vary, however, according to patient-specific needs, attitudes, and perceptions. Thus, to understand where best to begin treatment, clinicians should obtain a complete picture of the patient’s views on depression.45 For example, the majority of patients seeking help are focused on seeking a cure from the outset, and therefore value fast, definitive diagnosis and action; but approximately 1 in 4 patients are more focused on getting by from day to day, and their immediate needs may include maintenance medication, a professional caregiver with whom to engage in dialogue, and practical help with everyday issues.45 Treatment goals can naturally evolve as a function of progress. Accordingly, patients may prefer short- and long-term goals.45
Treatment success, while often achievable in select patient populations, is of course not preordained. In addition to vascular depression, factors such as symptom severity, poor social support, and comorbid dysthymia (reported in as many as 1 out of 3 older patients with MDD) predict a poor outcome or treatment response.5, 47-49 Therapeutic management is predicated on open communication and tailoring treatment consistent with initial, comprehensive and ongoing monitoring.
Good communication is critical to treatment success, particularly with patients who have never seen a mental health clinician before. Effective communication with patients can often be achieved through a 4-step process:
1. Explain the clinician’s role and how treatment works.
2. Assess and discuss expectations, since many older individuals are unfamiliar with psychiatric treatment.
3. Establish realistic treatment plans sensitive to any limitations due to cognitive impairment or physical disability.
4. Be aware that some individuals may be hesitant about treatment because of a social stigma associated with depression or with psychiatric treatment.50-52
Extent and severity of social stigma may be more prevalent in older patients who have not kept pace with changing social norms.53, 54 And although the burden of stigma may vary by culture,55-58 it has been observed in a substantial majority of depressed patients across all sociocultural backgrounds.50 Stigma has both an external component, involving fear of mistreatment or discrimination from others, and an internal component, involving personal feelings of guilt or shame surrounding the condition.50-52 Depressed patients are more likely to feel internal stigma,50 which may be a greater detriment than external stigma particularly regarding initial treatment.51 Finally, the emotional burden that stigma imposes on depressed patients is as great as that imposed by addiction or psychosis.50
In addition to social stigma, elderly patients may also hesitate to undergo pharmacotherapy because of bad past experiences, fear of dependence or addiction, concerns that antidepressants will affect their ability to experience natural sadness, and/or a reluctance to view their depression as a psychiatric illness.59 It is important to address these concerns to improve treatment response.59
Treatment of MDD in Older Adults: Nonpharmacologic Modalities
The role of psychotherapy in older adults with MDD has evolved considerably over the past two decades.5 For patients who are highly motivated, articulate, and not suffering from significant cognitive impairment, psychodynamic psychotherapy can be used to focus on uncovering unconscious conflicts as well as conscious grief and loss that may lie at the root of the depression.60 All forms of psychotherapy used to treat depression in younger adults are also applicable in late life, although adaptations are commonly necessary for older patients, addressing such challenges as cognitive impairment, hearing or visual loss, physical disability, chronic pain, lack of transportation, and limited income.61 There is some empirical evidence supporting the efficacy of several forms of psychotherapy in older depressed individuals both with and without pharmacotherapy,62 including cognitive-behavioral therapy (CBT),63 problem-solving therapy,64 interpersonal therapy,65 reminiscence therapy,66 and group therapy.67, 68 CBT aims to identify and change distorted thought patterns as well as to develop more relaxing and adaptive behaviors to deal with depressive symptoms and other problems that are driving the depression.69 In problem-solving therapy, the older patient and therapist work as partners to select a problem that is associated with the depression, brainstorm potential solutions, and then choose and implement one solution until successful.60 Interpersonal psychotherapy, on the other hand, focuses on identifying and addressing interpersonal conflicts that may be causing and perpetuating the depression.60 Reminiscence therapy (RT; also called “life review”) was developed specifically for older individuals.66, 70 In RT the therapist and patient explore past experiences in order to seek a greater understanding of one’s current situation as well as to identify and bolster longstanding strengths.70 Group therapy may also be extremely useful for older individuals who either resist or are not responding to individual therapy, or who need help with socialization skills. Listening to older peers helps to normalize late-life struggles, and facilitates “venting” otherwise difficult to express feelings and provides opportunities to support others.68
All forms of psychotherapy used to treat depression in younger adults are also applicable in late life, although adaptations are commonly necessary for older patients, addressing such challenges as cognitive impairment, hearing or visual loss, physical disability, chronic pain, lack of transportation, and limited income.
Treatment of MDD in Older Adults: Pharmacotherapy
Pharmacotherapy is the mainstay of treatment for the vast majority of older adults with MDD.5 Overall, the current evidence base supports the efficacy of a range of antidepressant strategies for older patients. For this special patient population, it is key to start with lower doses and to select medications tailored to patient-specific comorbid medical or psychiatric illness, drug-drug interactions, or vulnerabilities (such as risk of falls).5 Safety and tolerability across antidepressant classes appear to be similar, as demonstrated by similar withdrawal rates in clinical trials.5
While a thorough review of all randomized, double-blind, and placebo controlled or active comparator studies is beyond the scope of this monograph, important principles have emerged from select studies in older patients, especially regarding variability in antidepressant response.3, 5 In their recent meta-analysis of 10 antidepressant studies in adults 60 years or older, Nelson and coworkers reported that response rates increased from an estimated 38% at 6 to 8 weeks (versus 31%, placebo; P = 0.01) to 55% (versus 41%, placebo; P < 0.001) at 10 to 12 weeks (Figure 4).3 Remission rates were generally similar across all antidepressant categories, including serotonin-norepinephrine reuptake inhibitors (SNRIs), selective serotonin reuptake inhibitors (SSRIs), and dopamine-norepinephrine reuptake inhibitors (DNRIs)3. A review by Beyer concluded that all classes of antidepressants, including the SSRIs, tricyclic antidepressants (TCAs), and monoamine oxidase inhibitors (MAOIs), have similar efficacies.5 In a placebo-controlled study by Halikas, 6 weeks of treatment with a tetracyclic antidepressant, resulted in significant improvement in response rates.71 Still, these findings do not absolutely support one agent over another, since the confidence intervals for the odds ratios overlapped and there are many potential variables that would support or contraindicate certain antidepressants for individual patients. And in their recent study, Sackheim and coworkers found no significant difference in rates of response, remission, and time to remission between young and old cohorts of depressed individuals treated with antidepressants.4 These data are in sharp contrast to the widely accepted assumption, discussed earlier, that older patients tend to respond slower and with less robustness to antidepressants compared with younger patients.
Figure 4 (click to enlarge)
In clinical trials, variability in antidepressant response may be at least in part related to the study site. Roose and coworkers conducted an 8-week multicenter study of citalopram versus placebo in patients of very advanced age (≥75 years).72 They reported a remarkable degree of heterogeneity by study site: while most locations showed a similar response rate for drug versus placebo, the contrast between treatment centers was very high, with responses ranging from 18% (citalopram) and 16% (placebo) to 82% (citalopram) and 80% (placebo). In their study, severity of depression modulated response to placebo; patients with severe depression (HAM-D score >24) exhibited a significantly diminished response to placebo but not to active therapy (P = 0.008).61 A similar but nonsignificant trend was observed toward reduced placebo response in patients whose depression began at 60 years or older. These results are generally supported by Sneed and coworkers, who in an independent report suggested that antidepressant response may depend on whether a study medication is being compared to an active control or placebo, with active comparator trials generally eliciting greater overall response.73
Importance of Frequent Monitoring
Poor antidepressant adherence predicts poor clinical outcome. A secondary analysis of a prospective antidepressant study found that self-reported nonadherence and psychosocial factors associated with adherence (eg, perceived barriers to taking medication, poor social support, greater disability) were linked to higher depression scores at 12 months.74 Because self-report tends to overestimate actual adherence,74 it may be helpful to ask the patient to report on specific instances of missed doses.74 It is also advisable to flesh out the patient’s own report through other sources when available. Family or other caregivers provide a useful resource for adherence information. Assess the level of support from family and friends, and if support is in doubt, look for ways to improve it.
Inter-patient variability in treatment response also warrants continual monitoring and adjustments. Treatment should be personalized, with attention to individual variables such as past treatments, possible drug interactions, comorbidities, predictors of adherence, cognitive dysfunction, and frailty, as well as to patient concerns about pharmacologic therapy.5 Some studies have shown that partial response (or lack thereof) at 4 to 6 weeks is usually predictive of long-term outcome in both older and younger patients.5, 75 However, additional studies have demonstrated that outcomes observed in short trials do not reflect those achieved in longer trials.3 Mean response rates in 6- and 8-week trials ranged from 35% to 46.5%, compared with response rates of 45.7% to 68.9% seen in 10- to 12-week trials.3 Data from a pooled analysis (Figure 5) indicate that after about 4 to 6 weeks of treatment, patients who show a poor response should have their medication changed, while treatment regimens in partial responders should not be altered.49, 76 Relapse is more common in older patients, so successful treatment with an antidepressant should be continued for 1 to 3 years after remission, and longer in high risk patients, up to lifelong antidepressant treatment if needed.5, 77
Figure 5 (click to enlarge)
Longitudinal monitoring should also address cognitive decline, a particularly important concern in depressed elders.21 Psychotropic medications are likely to demonstrate anticholinergic properties, which are associated with low cognitive performance.78 Again, variable responses to antidepressant monotherapy is commonly observed in this patient population. Doraiswamy and coworkers, for example, compared the cognitive effects of two selective SSRIs (fluoxetine and sertraline) and a tricyclic antidepressant (TCA, nortriptyline) in 444 older patients with major depression and found improvement at 12 weeks in cognitive function among responders to all three agents.79 Among these, sertraline was associated with the greatest improvement from baseline, followed by nortriptyline and then by fluoxetine. In another study, after 4 weeks of open-label escitalopram, Savaskan and coworkers observed significant improvement in cognitive function in elderly depressed patients (n = 18) who, prior to the study, had markedly reduced cognitive function compared with healthy controls (n = 22).80 Further, in older patients with recurrent depression, Raskin and coworkers compared the efficacy of SNRI duloxetine (n = 204) and placebo (n = 104).81 The investigators observed significantly greater improvements in cognition over 8 weeks’ treatment in patients receiving duloxetine versus placebo. However, more recently Culang and coworkers assessed cognitive functioning in depressed patients 75 years or older who were randomized to the SSRI citalopram (n = 84) or placebo (n = 90).82 While citalopram responders (defined as 50% reduction from baseline HAM-D score) showed improvement in some aspects of cognitive function, the improvement was no greater than that seen in placebo responders. Citalopram nonresponders demonstrated a decline in verbal learning, which Culang et al attributed to the anticholinergic effects of SSRIs. The investigators concluded that antidepressant nonresponders should not be maintained on their medication.
Take Comorbidities Into Account
Disorders coexisting with depression should be directly addressed when selecting antidepressant therapy. Of note, Caughey and coworkers demonstrated that 87% of their study cohort had at least one comorbid condition that could cause a potential treatment conflict with antidepressant therapy.83 Importantly, cardiovascular disease is common in elderly patients with depression, and may limit treatment choices; in particular, TCAs increase heart rate and may not be appropriate for select patients, especially for those with ischemic heart disease.84 In their 6-week trial of older adults with MDD and comorbid ischemic heart disease, Roose and coworkers reported that both paroxetine (n = 41) and nortriptyline (n = 40) proved effective as per the HAM-D score.84 Adverse cardiac events (specifically, angina, tachycardia, or proarrhythmias) were significantly more common in the nortriptyline group (18%, 7/40) when compared with those treated with paroxetine (2%, 1/41). Wise and coworkers evaluated the efficacy of duloxetine (n = 207) versus placebo (n = 104) in elderly patients with MDD and coexisting medical disorders.85 Vascular disease (cardiovascular, cerebrovascular, or peripheral vascular disease), diabetes, or arthritis were present in 33% (68/207), 17% (35/207), and 57% (117/207) of patients in the duloxetine group, respectively. Duloxetine was associated with improved cognitive function versus placebo (P = 0.013) and was well tolerated, independent of the presence or absence of comorbidity. Sheikh and coworkers reported on the efficacy, safety, and tolerability of sertraline versus placebo in patients 60 years or older with MDD (n = 728).86 Comorbidity (vascular disease, diabetes, or arthritis) was present in 61% (442/728) of patients in the study. Sertraline treatment was associated with reduced depressive symptoms versus placebo and was well tolerated in patients with and without comorbidity.
Implications of Polypharmacy
The risk of drug-drug interactions increases as a function of age, with comorbid illness, and with polypharmacy.87 Of note, Nobili and coworkers analyzed a prescription database and showed that patients taking 6 or more medications daily on a chronic basis had approximately 2 times higher risk of drug-drug interactions than those taking 3 to 5 medications and 6 times higher risk than those taking 0 to 2 medications.88 This has particular implications for clinicians considering adding or switching antidepressants for patients who have demonstrated partial response to the maximum dose of an antidepressant.49
In elderly patients, MDD is prevalent, underdiagnosed, and undertreated. Encouraging results from recent studies have found treatment response to be similar to that seen in younger patients, except when depression in the elderly patient is associated with vascular damage. MDD adversely affects all aspects of life and may be overlooked because of cognitive dysfunction, fatigue, and other age-related symptoms, or mistaken for normal responses to end-of-life issues such as infirmity and mortality. Individualized and successful treatment requires realistic treatment goals. Among them, alleviating the current episode and achieving long-term remission are perhaps the overarching goals that require thoughtful discussion with the patient. Treatment response is influenced by a number of biological, clinical, and psychosocial factors. It is essential to begin by treating causative or exacerbating factors and counseling patients who have unrealistic expectations, or who stigmatize depression and/or psychiatric treatment. Suicidal ideation is not uncommon in older adults with depression, and the risk of suicide should be carefully stratified and managed accordingly. Appropriate laboratory work should be evaluated, particularly for factors that reflect age-related organ and/or cognitive function. All forms of psychotherapy and pharmacotherapy used in younger adults are also applicable in those with late-life depression, but both treatment types are adapted as necessary for older patients. Most antidepressant classes have roughly equivalent efficacy in this population. Polypharmacy is common in older patients, as are the risks or consequences of drug-drug interactions. Predictors of poor outcome include symptom severity, comorbid dysthymia, vascular damage, and poor social support. Diminished adherence in patients with impaired cognition may also lead to poorer outcomes. For optimal treatment, physicians and their multidisciplinary team should approach late-life MDD as a condition with a unique clinical profile distinct from that observed in younger patients. It is critical to develop a treatment plan in collaboration with the patient. When starting pharmacotherapy, prescribe low doses and go slow, monitoring adherence and side effects closely. If there is no response by 4 to 6 weeks and the medication is well tolerated, reassess the dose and consider an increase, with the plan to switch to another antidepressant if there is no sign of response within another week or two.49 If there is a partial response at 4 to 6 weeks, the trial should be extended to 12 weeks, during which time the decision can be made to either increase the dose, augment with another agent, or switch to a new agent in the same or better yet in a different antidepressant family.49 Patients with residual depressive symptoms despite long-term antidepressant therapy are at significant risk of relapse and may benefit from addition of such atypical antipsychotics as aripiprazole and quetiapine. These and other augmentation strategies are an active area of clinical research, as reviewed extensively elsewhere. Always remain hopeful and persistent, bearing in mind that MDD is not an inevitable consequence of old age, and through treatment, social support, and close monitoring it can be properly treated and, in many cases, placed into remission.
1. Jacobson S. Atlantic Monthly. April 1995:46-51.
2. NIH Consensus Development Panel on Depression in Late Life. JAMA. 1992;268:1018-1024.
3. Nelson JC, et al. Am J Geriatr Psychiatry. 2008;16:558-567.
4. Sackeim HA, et al. Biol Psychiatry. 2006;59:483-492.
5. Beyer JL. Ann Clin Psychiatry. 2007;19:221-238.
6. Grabovich A, et al. Am J Geriatr Psychiatry. 2010;18:227-235.
7. Small GW. J Clin Psychiatry. 2009;70:e47.
8. Styron W. Vanity Fair. 1989(December):212-286.
9. LaRue A, Dessonville C, Jarvik LF. Aging and mental disorders. In: Birren JE, Schaie KW, eds. Handbook of the Psychology of Aging. 2nd ed. New York: Van Nostrand Reinhold; 1985:664-702.
10. Wakefield JC, et al. Arch Gen Psychiatry. 2007;64:433-440.
11. Kendler KS, et al. Am J Psychiatry. 2008;165:1449-1455.
12. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Fourth Edition, Text Revision, DSM-IV-TR™. Washington, DC: American Psychiatric Association; 2000.
13. Kroenke K, et al. J Gen Intern Med. 2001;16:606-613.
14. Herrmann N, et al. Int J Geriatr Psychiatry. 1996;11:457-460.
15. Sheikh J, Yesavage J. Geriatric Depression Scale: Recent evidence and development of a shorter version. New York: Routledge; 1986.
16. Hamilton M. J Neurol Neurosurg Psychiatry. 1960;23:56-62.
17. Zung WW. Arch Gen Psychiatry. 1965;12:63-70.
18. Montgomery SA, Asberg M. Br J Psychiatry. 1979;134:382-389.
19. Alexopoulos GS, et al. Biol Psychiatry. 1988;23:271-284.
20. Spitzer RL, et al. JAMA. 1999;282:1737-1744.
21. Dillon C, et al. Neuropsychiatric Dis Treat. 2009;5:517-526.
22. Robertson RG, Montagnini M. Am Fam Physician. 2004;70:343-350.
23. American Association of Suicidology. Elderly suicide fact sheet. http://www.suicidology.org/c/document_library/get_file?folderId=232&name=DLFE-242.pdf. Accessed July 22, 2010.
24. Conwell Y, et al. Biol Psychiatry. 2002;52:193-204.
25. Duberstein PR, Heisel MJ, Conwell Y. Suicide in older adults. In: Agronin ME, Maletta GJ, eds. Principles and Practice of Geriatric Psychiatry. Philadelphia, PA: Lippincott Williams & Wilkins; 2006:393-405.
26. St John PD, et al. Int J Geriatr Psychiatry. 2009;24:369-375.
27. Borson S, Thompson DJ. Major depression and related disorders in late life. In: Agronin ME, Maletta GJ, eds. Principles and Practice of Geriatric Psychiatry. Philadelphia, PA: Lippincott Williams & Wilkins; 2006:349-368.
28. National Institute of Mental Health. Depression. Bethesda, MD: National Institute of Mental Health, US Department of Health and Human Services; 2008. NIH publication 08-3561. http://www.nimh.nih.gov/health/publications/depression/nimhdepression.pdf. Accessed July 22, 2010.
29. Alexopoulos GS, et al. Am J Psychiatry. 1993;150:1693-1699.
30. Kritz-Silverstein D, et al. J Nutr Health Aging. 2009;13:317-321.
31. Kim JM, et al. Br J Psychiatry. 2008;192:268-274.
32. Alves de Rezende CH, et al. J Nutr Health Aging. 2009;13:617-621.
33. Schram MT, et al. J Am Geriatr Soc. 2007;55:1786-1792.
34. Alexopoulos GS, et al. Arch Gen Psychiatry. 1997;54:915-922.
35. Sneed JR, et al. Biol Psychiatry. 2008;64:491-497.
36. Thomas AJ, et al. Arch Gen Psychiatry. 2002;59:785-792.
37. Taylor WD, et al. Arch Gen Psychiatry. 2003;60:1090-1096.
38. Figiel GS, et al. Neurobiol Aging. 1991;26:245-247.
39. Alexopoulos GS, et al. Arch Gen Psychiatry. 2000;57:285-290.
40. Kalayam B, Alexopoulos GS. Arch Gen Psychiatry. 1999;56:713-718.
41. Jorge RE, et al. Arch Gen Psychiatry. 2008;65:268-276.
42. Monzani D, et al. Acta Otorhinolaryngol Ital. 2010;30:11-19.
43. Unützer J, et al. JAMA. 2002;288:2836-2845.
44. Johnston O, et al. Br J Gen Pract. 2007;57:872-879.
45. Rabheru K. Can J Psychiatry. 2004;49(supp 1):41S-50S.
46. Devanand DP, et al. Am J Psychiatry. 1994;151:1592-1599.
47. Gilbody S, et al. Arch Intern Med. 2006;166:2314-2321.
48. Hybels CF, et al. Am J Geriatr Psychiatry. 2005;13:713-721.
49. Driscoll HC, et al. Drugs Aging. 2007;24:801-814.
50. Dinos S, et al. Br J Psychiatry. 2004;184:176-181.
51. Schomerus G, et al. Eur Arch Psychiatry Clin Neurosci. 2009;259:298-306.
52. Kantner JW, et al. J Nerv Ment Dis. 2008;196:663-670.
53. Segal DL, et al. Aging Ment Health. 2005;9:363-367.
54. Sirey JA, et al. Am J Psychiatry. 2001;158:479-481.
55. Conner KO, et al. J Gerontol Soc Work. 2009;52:695-712.
56. Georg Hsu LK, et al. Psychiatry. 2008;71:210-218.
57. Menke R, Flynn H. J Nerv Ment Dis. 2009;197:407-411.
58. Fogel J, Ford DE. Can J Psychiatry. 2005;50:470-478.
59. Givens JL, et al. J Gen Intern Med. 2006;21:146-151.
60. Frazer CJ, et al. Med J Aust. 2005;182:627-632.
61. Wyman MF, Gum A, Areán PA. Psychotherapy with older adults. In: Agronin ME, Maletta GJ, eds. Principles and Practice of Geriatric Psychiatry. Philadelphia, PA: Lippincott Williams & Wilkins; 2006.
62. Areán PA, Cook BL. Biol Psychiatry. 2002;52:293-303.
63. Thompson LW, et al. Am J Geriatr Psychiatry. 2001;9:225-240.
64. Alexopoulos GS, et al. Am J Geriatr Psychiatry. 2003;11:46-52.
65. Reynolds CF III, et al. JAMA. 1999;281:39-45.
66. Hsieh HF, Wang JJ. Int J Nurs Stud. 2003;40:335-345.
67. Leszcz M, Kobos JC. J Clin Psychol. 2008;64:1238-1260.
68. Agronin M. Curr Psychiatry Rep. 2009;11:27-32.
69. Gallagher-Thompson D, Thompson LW. Applying cognitive-behavioral therapy to the psychological problems of later life. In: Zarit SH, Knight BG, eds. A Guide to Psychotherapy and Aging. Washington, DC: American Psychological Association; 1996:61-82.
70. Korte J, et al. BMC Public Health. 2009;9:250.
71. Halikas JA. Hum Psychopharmacol. 1995;10(suppl):S125-S133.
72. Roose SP, et al, for the Old-Old Depression Study Group. Am J Psychiatry. 2004;161:2050-2059.
73. Sneed JR, et al. Am J Geriatr Psychiatry. 2008;16:65-73.
74. Bosworth HB, et al. Int J Geriatr Psychiatry. 2008;23:129-134.
75. Sackeim HA, et al. J Clin Psychopharmacol. 2005;25(suppl 1):S34-S37.
76. Mulsant BH, et al. J Clin Psychopharmacol. 2006;26:113-120.
77. Rajji TK, et al. Drugs Aging. 2008;25:841-853.
78. Chew ML, et al. J Am Geriatr Soc. 2008;56:1333-1341.
79. Doraiswamy PM, et al. J Gerontol A Biol Sci Med Sci. 2003;58:M1137-M1144.
80. Savaskan E, et al. The Int J Neuropsychopharmacol. 2008;11:381-388.
81. Raskin J, et al. Am J Psychiatry. 2007;164:900-909.
82. Culang ME, et al. Am J Geriatr Pscyhiatry. 2009;17:881-888.
83. Caughey GE, et al. Age Ageing. 2010;39:488-494.
84. Roose SP, et al. JAMA. 1998;279:287-291.
85. Wise TN, et al. Int J Clin Pract. 2007;61:1283-1293.
86. Sheikh JI, et al. J Am Geriatr Soc. 2004;52:86-92.
87. Pollock B, et al. Clin Pharmacol Ther. 2009;85:89-93.
88. Nobili A, et al. J Clin Pharm Ther. 2009;34:377-386.