Show Summary Details
Page of



Helen Lavretsky

, Martha Sajatovic

, and Charles F. Reynolds

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 20 January 2020

FOLLOWING THIS comprehensive review of recent advances in research in late-life mood disorders, we can say that we have come a long way in understanding risk factors and pathophysiology in late-life mood disorders. Various psychosocial and neurobiological risk factors for developing mood disorders in later life are relatively well understood. Neuroimaging studies have helped us to gain a better understanding of the anatomy, functional circuitry, and biochemistry of mood disorders in older people. New developments in information technology have helped to improve the assessment and management of late-life mood disorders. However, more needs to be done in developing personalized pharmacological and psychosocial treatments for late-life depression and bipolar disorders. Promoting understanding of genetic and epigenetic contribution to late-life mood disorders can lead to the development of selective and indicated preventive approaches that increase resilience in later life. Integration of biomarkers in the development of interventions remains a challenge. Understanding the contribution of cognitive impairment and biological aging-related physiological changes is still in the early stage. Global mental health initiatives are likely to be more prominently present in prevention research. Next we outline the current state of art in research and clinical care, and the immediate and distant goals for future research.


Late-life mood disorders continue to be significantly underdiagnosed in primary care, which is the health care setting used most often by elderly persons. Depression is missed in approximately half of all elderly persons with mood disorder. Given the risks for deleterious effects of depression on health, the failure to diagnose late-life mood disorders remains a serious public health problem and suggests that geriatric psychiatrists need to be more proactive in working with primary health care systems.

There are a variety of barriers in primary care to the recognition and diagnosis of late-life depression. Primary care providers may miss the diagnosis of late-life depression because of insufficient training, a focus on medical conditions that mask the presence of depression, or overlapping symptoms of dementia. Time constraints, inadequate reimbursement, and lack of support staff also hinder the recognition of late-life mood disorders in primary care settings. Patient barriers include stigma about a psychiatric diagnosis. Better education and training of primary care providers and general public can improve appropriate and timely diagnosis of mood disorders and aid in suicide prevention.

Diagnosis of late-life mood disorders can be difficult, in part, because patients often present with somatic symptoms, overlapping symptoms of medical illness, or coexisiting mild cognitive impairment. Depression should be suspected and investigated in elderly persons with anhedonia, apathy, hopelessness, anxiety, psychomotor retardation, and unexplained fatigue and weight loss. Among older adults, mood and anxiety disorders are associated with physical illness and cognitive impairment, and result in reduced quality of life and increased functional disability and suicide risk. Dimensional diagnosis can be very useful in approaching geriatric depression given significant medical, neurological, and psychiatric comorbidity. Of particular concern, a significant risk associated with late-life depression is that of excess mortality. Depression has been shown to predict less successful outcomes of treatment for other physical conditions, particularly cardiovascular and cerebrovascular disease, and to increase death rates among the elderly. Depression should be considered in the differential diagnosis of hospitalized elderly persons with myocardial infarction, congestive heart failure, stroke, hip fracture, cancer, or alcohol/substance abuse who exhibit delayed recovery, poor compliance with rehabilitation programs, or treatment refusal. Depression should also be investigated in patients who are apathetic, withdrawn, or agitated, or who exhibit increased dependency, functional decline, or delayed rehabilitation, and in patients receiving long-term care.

The DSM-5/ICD-10 diagnostic criteria for major depression and bipolar disorder are explicit and may improve diagnosis of late-life depression and include such common features as comorbid anxiety (e.g., in the category of mixed Anxiety-Depression) or comorbid mild-cognitive impairment. However, the language used to describe other milder forms of mood disorder remains imprecise and consists of poorly defined terms, including clinical depression, depressive symptoms, subsyndromal or subthreshold depression, minor depression, and depressed mood. This continuous ambiguity about clinical significance of subclinical depression might still interfere with the ability to make substantive changes in clinical practice. Current diagnostic tools do not adequately address an individual’s potential for suicide. Clinicians would benefit from the development of assessment tools that assist in identifying patients with acute and chronic suicidal risk. The effect of possible correlates of suicide (e.g., hopelessness and anxiety) and protective factors (e.g., resilience, ethnicity, and religious beliefs) needs further study. However, future research encouraged by the National Institute of Mental Health “R-DoC” research criteria for dimensional diagnosis of depression or anhedonia may improve the nature of diagnosis in research and clinical practice to include subclinical mood symptoms.

Several subtypes of late-life depression have been identified that may benefit from further study and subsequent development of age-specific diagnostic criteria. Vascular depression may be associated with a later age at onset, and patients may exhibit symptoms of motor retardation, anhedonia, apathy, poor insight, functional disability, and impaired executive function. The features, treatment response, and illness course of late-onset depression, which is defined as a first episode of depression occurring after age 60 years, may differ from those of depression in younger adults. Late-onset depression can be characterized by a greater degree of apathy, cognitive dysfunction, and temporal lobe abnormalities that are similar to changes associated with dementia. Understanding brain aging contribution to the pathophysiology of late-life mood disorders will open new opportunities for the development of new treatment approaches.

The diagnosis of late-life bipolar disorder is an especially neglected and understudied area. Patients with late-onset bipolar disorder may be more likely to have medical comorbidity, particularly cerebrovascular illness, and psychotic features. The differential diagnosis of late-life bipolar disorder must include assessment of comorbid conditions or medications that can cause mania. Once these factors are excluded, a detailed cognitive assessment may inform the potential for recovery and long-term outcome. A better understanding of the risk of precipitating mania in depressed elderly individuals treated with antidepressant agents is also needed. Stigma can delay treatment initiation and a favorable outcome in patients who screen positive for depression.


While a range of effective interventions are available for mood and anxiety disorders, there remains a considerable need for developing innovative approaches that overcome the shortcomings of these interventions when applied to geriatric mental health care. Current psychopharmacological approaches to the treatment of late-life mood disorder have been shown to be effective for alleviating symptoms in about two thirds of those receiving treatment Medications are effective in treating late-life depression and combination therapies (medications and psychotherapy) are effective as well. However, a large proportion of patients still do not fully recover and can remain disabled and at risk for other negative outcomes. Moreover, older adults’ response to treatment often occurs somewhat more slowly and/or proves to be less sustained over time than that of younger adults. Although such treatment shortcomings are not unique to older adults, they may pose more acute problems for geriatric patients.

Advances in delineating the pathophysiology of mood and anxiety disorders have opened pathways for developing novel interventions that, for example, might target different or additional neurotransmitter systems to those addressed by current antidepressant medications, or that might utilize varied techniques (e.g., pharmacological, psychosocial, cognitive, device-driven) to modulate the activity of brain circuits that have been identified as hyper- or hypoactive in depression and anxiety. This proliferation of studies has contributed to increased understanding of the public health impact of late-life mood disorders, including data needed to guide treatment and inform basic research. Treatment is effective across settings, including primary care, rehabilitation, and long-term care. Controlled studies have demonstrated the efficacy of medication therapy, psychotherapy, and electroconvulsive therapy. Compared with younger adults, older persons often exhibit higher and more variable drug concentrations and a greater sensitivity to adverse effects.

The gold standard of treatment remains the achievement of full remission and a return to wellness. It is now known that antidepressant drug therapy, interpersonal psychotherapy, and the combination of medication use and psychotherapeutic intervention are as efficacious in preventing recurrent major depression in elderly patients as in younger adults. Treatment of psychotic depression and bipolar disorder in late life remains especially understudied, although two major collaborative studies have been recently been completed. There is a critical need to both better understand how pharmacotherapies widely used in younger patients with mood disorders should be optimally used in older patients as well as research with novel agents that may be particularly helpful for older adults. Partial response to the first-line treatment is common and occurs in 30%–50%. Some elderly patients with mood disorders do not respond to treatment, and more needs to be known about factors associated with treatment response and nonresponse. Developing biomarkers of response or nonresponse can accelerate development of individualized and biomarker-driven treatment approaches.

Underlying the positive evidence from treatment studies is the need to address nonadherence. Patients who do not follow treatment advice do so for many different reasons, including stigma associated with antidepressants, inadequate education and support from providers, adverse effects or perceived lack of efficacy, drug–drug interactions, complexity of dosing regimens, medication cost, inadequate insurance coverage, and lack of awareness about the sequelae and chronicity of mood disorders. Although medication adherence is associated with favorable outcomes, approximately 40% of patients of all ages with major depression and 50% with bipolar disorder are nonadherent to antidepressant therapy. Compared with nondepressed persons, depressed medical patients are three times more likely to be nonadherent to medication regimens, exercise, diet, other health-related behaviors, vaccinations, and appointments. These observations suggest the need to improve collaboration with families and care providers and develop therapeutic alliances to improve treatment adherence. Treatment studies need to move progressively to primary care and community populations to treat representative older adult populations.

Studies of Risk and Protective Factors for Late-Life Mood Disorders as a Basis for Selective Prevention

Depression and bipolar disorder in the elderly are not necessarily associated with the same risk factors as in younger adults. It is necessary to identify temporal relationships, correlates, and causal patterns to develop clinically useful profiles for high-risk patients and to inform prevention and intervention studies and health care policy.

Late-life depression and mania often occur in the context of chronic medical illness and the medications used to treat them. Medical illness, chronic pain and insomnia, inflammatory processes, functional disability, family and personal histories of mood disorders, social isolation, bereavement, caregiving, and other losses frequently co-occur with depression. Temporal and causal relationships between depression and such correlates are not known with precision and are likely to be bidirectional. Risk factors for depression may also be consequences of depression. Social isolation may contribute to depression risk, but depression can also lead to loneliness and isolation. Correlates of risk for late-life bipolar disorder are less well studied but include comorbid neurological illness, cerebrovascular disease, and a family history of affective disorder. Depression and cardiovascular and cerebrovascular diseases are also strongly correlated. Stroke is a risk factor for depression, and, conversely, depression predicts a poor outcome after stroke. Depression occurs in roughly one third to one half of individuals after stroke. A few intervention studies on antidepressant treatment of post-stroke depression are promising and suggest a pathway for larger prevention and intervention studies.

Late-life depression and depressed mood are common correlates of cognitive impairment and may be prodromes for subsequent cognitive dysfunction and dementia. The presence of mild cognitive impairment during an episode of major depression in late life may precede the development of dementia, particularly if the depression is severe, untreated, and associated with psychomotor retardation or psychosis. Depressive symptoms, especially those with a late-life onset, can herald the onset of Alzheimer disease in nondemented patients. Some cognitive abnormalities such as abnormal performance in response inhibition or in initiation and perseveration tasks and structural neuroimaging indices (e.g., white matter hyperintensities and hippocampal atrophy) may predict a treatment-resistant depression. These findings are preliminary and require further study and replication.

Mood disorders in late life are a major correlate of suicidal behavior. In the United States, suicide is more common in elderly White men than in the general population. Although medical illness and functional disability also are associated with late-life suicide, the risk associated with these factors is mediated by the presence of depression. Other risk factors such as personal or family history of mood and anxiety disorders, family history of suicide, loneliness, hopelessness, access to handguns, and insufficient social support systems are also important. Suicidal intent or behavior often is not recognized by physicians, and many older adults who commit suicide have visited a physician in the month before their suicide. Suicide prevention for elderly persons must be a primary focus of intervention efforts in the community.


Despite increased awareness of older age mood disorders in recent decades, there are still a number of questions that remain unanswered. Based on the rapid development of neuroimaging, genomic, and computer technologies, we anticipate a shift in research priorities that will address immediate needs of the field. Better understanding should encompass better methods of assessment, more appropriate treatment options for older people and their families, and public education about these conditions. Future research will also seek to translate findings into clinical practice to improve the well-being of older people. Given increasing economic globalization, international efforts in mental health research will focus on alleviation of the global burden of depression and associated disability, and they are likely to utilize simple algorithm-based approaches to the management and prevention of common mental disorders in primary care settings.

Current and Future Research Priorities

  • Longitudinal studies that assess interactions between psychosocial and biological factors and inform the development of clinically useful high-risk profiles for late-life mood disorders, including better delineation of acute versus chronic risk factors for suicide

  • Selective and indicated prevention and intervention studies to identify and reduce risk for first-episode and recurrent late-life mood disorders

  • Studies of genetic liability to mood disorders with late-life onset with integration of biomarkers of inflammation and biological and cognitive aging

  • Studies of depression subtypes, including dimensional diagnosis of depression and anhedonia, vascular depression, depression associated with inflammation, depression of Alzheimer’s disease, dysthymia, and late-onset depression, and development of clinically useful diagnostic tools that facilitate recognition and appropriate treatment

  • Studies of the mechanisms by which mood disorders predispose to dementing illnesses

  • Studies of depression and suicide prevention, screening, and diagnosis in primary care settings and in underserved minorities

  • Longitudinal, early-treatment studies (including psychotherapy and combined pharmacotherapy and psychotherapy) that address prevention of treatment resistance, relapse, recurrence, persistence, and chronicity

  • Studies to measure the effect of psychotherapeutic and psychoeducational interventions on treatment adherence

  • Development of treatment algorithms specifically tailored to the elderly, including guidelines to optimize treatment response in partially or slowly responding and nonresponding patients and to guide long-term treatment planning

  • Studies of pharmacogenetic and pharmacodynamic sources of treatment response variability

  • Development of new classes of therapeutic agents for late-life depression (including vascular depression, inflammatory mechanisms, biological aging, and cognitive impairment) and bipolar disorder

  • Development of a public health model for mental health care that encompasses case finding, facilitation of access to and delivery of care, and evaluation of reimbursement for service innovations to improve outcome

  • Development of partnerships among the research community, health care providers (including agencies that provide home-based nursing care), and health care payers such as the Centers for Medicare and Medicaid Services to design economically sustainable models of mood disorder care in the general medical sector and to further effectiveness and cost-effectiveness studies of evidence-based, clinical disease management models

  • Development of partnerships between researchers and religious and other community agencies for outreach to underserved elderly people and for overcoming mistrust of research and mental health care among minorities

  • Development of international collaborations for global mental health initiatives