OVERVIEW

Depression is among the most common health conditions affecting older adults. Depression may present in a phenotypically distinct manner in older, as opposed to younger, individuals. For example, in the older adult depression is less likely to present with dysphoria and more likely to be accompanied by irritability and somatic symptoms.¹ In addition, psychosocial stressors predisposing to the development of affective symptoms are different with advancing age (e.g., the stressor of losing a loved one becomes more common).¹ Still, one of the major factors affecting the inception and prognosis of depression in the older adult is the co-occurrence of medical conditions.

In this chapter, we will address common medical conditions with which coexistent depression is likely to occur in the older adult. We also will give attention to biological and psychosocial mechanisms that might give rise to both the depressive symptoms and the medical condition, and emphasize how this co-occurrence predicts poorer health outcomes for both conditions. We will then survey methods for screening older adults with medical illness for depression, and review methods of differentiating it from symptoms that could simulate it, but which are more likely due to the medical condition. Next, we explore pharmacologic and nonpharmacologic treatment approaches for major depressive illness in the setting of a medical condition, and survey the considerations that must be taken into account in older adults. Finally, we review future methods of treating depression in this clinical setting, including novel techniques for preventing depression in the medically ill older adult (Table 21-1).

EPIDEMIOLOGY OF DEPRESSION IN OLDER ADULTS AND ROLES OF MEDICAL COMORBIDITY

Overall, it is clear that the burden and health impacts of depression among older adults, particularly those with medical illness, are high, and that substantial numbers of cases occur in late-life. Current (12-month) prevalence of major depressive disorder and persistent depressive disorder (dysthymia)—is estimated at 2% to 4%.^2,3 However, when one considers presentations of subsyndromal or minor depressions, the current prevalence is much higher for older adults. In a comprehensive meta-analysis of many studies of community-based older adults, the current prevalence of any depressive disorder—major or minor—was estimated to be 13%.² Lifetime prevalence is much higher than current prevalence. As is observed in younger persons, there is a gender difference in the lifetime burden of depression such that the ratio of women to men affected is approximately 2:1. For example, one study estimated the lifetime prevalence of depression among men aged 65+ years at 9.6%; while, the lifetime prevalence among women of the same age was 20.4%.³

Another important epidemiological consideration is whether depression in the older adult is new in onset, or whether it represents a recrudescence of a prior mood disorder. In the former case, it is incumbent upon the clinician to search for medical conditions/medications that may be mimicking depression (see below), as well as to evaluate unique psychosocial factors related to aging. The distinction can also be important for prognostic reasons, as new onset depression in the elderly has been associated with greater treatment resistance and a worse overall prognosis.⁴ An impactful study by Luijendijk et al.⁵ demonstrated incidence rates of seven cases per 1,000 person-years for major depression and 19 cases per 1,000 person-years for combined minor and major depression. These rates are comparable to incidence rates of other major diseases in this age group, such as myocardial infarction, stroke, and breast cancer. Furthermore, older people are affected by high recurrence rates of depression—given that depressive episodes in earlier life are a risk factor for depression later in life. Thus, recurrence outpaces incidence by 3 to 4 to 1; overall recurrence rates were 27 cases per 1,000 person-years for major depression and 66 cases per 1,000 person-years for minor and major depressions combined.⁵

This high prevalence of depression in older adults has substantial implications for healthcare costs. For example, in a study of Medicare beneficiaries,¹⁰ patients with depression had significantly higher total healthcare costs than those without ($20,046 compared to $1,196; p< .01). Importantly, the magnitude of the difference in healthcare costs associated with depression was higher with increasing levels of medical comorbidity, as measured by the Charlson index.¹⁰ The costs associated with depression treatment per se were only a small fraction of the incremental healthcare costs. Therefore, results support that, in addition to their benefits for the quality of life, adequate treatment of depression in medically ill older persons could be an important moderator of health care costs.

CO-OCCURRENCE OF MEDICAL ILLNESS AND DEPRESSION AMONG OLDER ADULTS

Major depression commonly coexists with other medical conditions. Depression and medical illness often share common biopsychosocial antecedents, and each can impact the clinical course of the other in a deleterious manner. Katon⁸ outlined several of the bidirectional influences of depression and medical illness. First, depression is a risk factor for the development of medical illness. In some cases, this causal influence is readily apparent. For example, depression may give rise to poor health habits (such as smoking, an insalubrious diet and/or a sedentary lifestyle), which can serve as risk factors for the development of cardiovascular and cerebrovascular disease. Second, medical illness may catalyze the onset of depression. Thus, the presence of at least one chronic medical condition significantly increases the risk of having depression.¹¹ In some cases, depression may be an effective response to the psychological stress of having a medical condition, particularly chronic, painful, and debilitating or life-threatening conditions.¹² Alternatively, certain medical conditions likely induce depression through neurobiological mechanisms. This is supported by the fact that select medical illnesses appear to produce depressive symptoms significantly more commonly than others causing similar amounts of disability and mortality. Moreover, coexistent neurological illnesses can have direct effects upon corticolimbic brain networks that regulate mood. In addition, the medications used to treat a medical illness may themselves have depressogenic actions. Finally, a given medical condition might induce depression through indirect effects on the central nervous system (CNS). For example, proinflammatory cytokines such as interleukin-6 (IL-6), which may be elaborated in the setting of medical disease, are capable of traversing the blood–brain barrier and of stimulating the hypothalamic–pituitary–adrenal (HPA) axis.⁷ Given the high proportion of older adults with medical morbidities, all of the above-mentioned potential paths to depression are noteworthy in this population.

IMPLICATIONS OF COMORBIDITY AMONG OLDER ADULTS

In the older adult, the interplay between depression and medical illness is amplified. Depression is increasingly prevalent with advanced age, especially in hospitalized elders.¹³ Greater medical comorbidity is associated with more severe depression in the elderly, and with greater utilization of medical resources.^13,14 Several factors increase the strength of this relationship between depression and medical illness in the older adult: (1) the increased likelihood of having one or more chronic medical conditions in individuals over the age of 65;⁸ (2) the increased likelihood of functional impairment resulting from a given medical condition; (3) medication factors such as polypharmacy, changes in pharmacokinetics and increases in drug–drug interactions; (4) a higher risk that a given medical illness will directly and/or indirectly impact affective brain networks; (5) changes in neuroendocrine and neuroinflammatory function¹² as well as changes in neuroplasticity¹⁴ with advanced age; (6) changes in sleep architecture;¹⁵ and (7) changes in the nature and extent of certain psychosocial stressors in later life.

Identifying depression in the older adult with medical illness can be challenging, in part because many medical conditions cause symptoms that mirror depressive somatic symptoms (such as apathy, nonspecific physical complaints, fatigue, and anergia). Nonetheless, increasing awareness of the association between depression and medical illness in the older adult is critical, because depression increases the morbidity and mortality of the comorbid medical illness, and because chronic medical conditions are known to increase the morbidity of co-occurring depression.

SPECIFIC MEDICAL CONDITIONS WHICH COEXIST WITH DEPRESSION IN THE OLDER ADULT

Depression in Neurological Illness

Certain neurological conditions can induce depressive symptoms by directly disrupting homeostatic neurobiological mechanisms that regulate mood. Specifically, this could occur through alteration of neurotransmitter levels, disruption of anatomical and/or functional connections among corticolimbic networks, and modulation of neuroplasticity (e.g., by changing the expression of trophic factors). Neurological conditions affecting such neurobiological changes become increasingly common with advanced age.

For example, the association between neurodegenerative disease and depression is well supported. When coexistent with depression, cognitive symptoms and signs can be mistaken for a dementing illness—a condition referred to historically as “pseudodementia.” Cognitive deficits induced by mood symptoms can be distinguished from a true neurodegenerative condition by a patient’s neuropsychological profile. Depression is associated with dysfunction of frontal-subcortical circuits, and presents with psychomotor slowing, frontally based memory deficits and executive dysfunction¹² In contrast, Alzheimer disease (AD) is typically also attended by deficits with memory consolidation, language deficits and loss of semantic knowledge. In addition, the extent of cognitive impairment can be helpful to differentiate depression from dementia. Finally, the two conditions can be differentiated through the use of biomarkers, such as structural and functional neuroimaging and CSF analysis.

Depressive symptoms can also co-occur with diagnostically established neurodegenerative conditions. In some cases, this occurs because depression and dementia affect shared neural circuits. For example, apathy is a common symptom in depression, frontotemporolobar degeneration and AD. Full depressive symptomatology can also accompany neurodegenerative illnesses. For instance, approximately 35% of patients with Parkinson disease¹⁶ have clinical depression,⁶ and the incidence of this is higher than in other chronic debilitating disorders. One explanation for this association is that changes in monoaminergic neurotransmission, specifically dysfunction of dopaminergic basal ganglia relays, predispose PD patients to depressive symptoms (e.g., anhedonia). Lewy body deposition in limbic regions may also contribute to depressive symptoms.⁶ The presence of depression in PD is associated with greater physical disability and increased cognitive decline.¹⁷ Finally, it is now well recognized that major depression can be a prodromal condition of PD, predating its presentation by years or decades.⁶ Comorbidity between depression and AD is less well established, with some studies suggesting that premorbid depression is a strong predictor of the development of amnestic dementia, and others finding no such relationship.¹⁸ It does appear, however, that individuals with AD are more likely than matched controls to be depressed. Also, similar to PD, the presence of depression in AD is predictive of greater functional decline and overall disability.

Cerebrovascular disease is another factor that is consistently linked to the onset of late life depression. This has given rise to the “vascular depression hypothesis.”⁷ Cerebrovascular disease giving rise to depression can either take the form of a clinical stroke or occult subcortical white matter hyperintensities revealed on T2-weighted MRI. Several reports have found that late-life depression is more common in adults with high microvascular ischemic disease burden, as compared to age-matched controls. Also, microvascular ischemic disease burden has been preferentially tied to late-onset (as opposed to early-onset) depression, and the extent of this disease burden has been linked to depression severity.⁷ Because leukoaraiosis tends to be highest in concentration in prefrontal subcortical white matter, depression due to microvascular ischemic disease presents with a similar neuropsychological profile of frontal-subcortical dysfunction. With respect to clinical strokes, post-stroke patients are more depressed than controls matched for level of functional disability, suggesting a neurobiological, as opposed to psychological, effect of the stroke on mood. Often, these strokes affect the left hemisphere preferentially and tend to involve the prefrontal cortex, the basal ganglia, or cortical-subcortical relays between these structures.¹⁹ Still, following a clinical stroke it is often difficult to disentangle depressive symptoms, such as fatigue or apathy, from other common sequelae of cerebral infarction.²⁰ Moreover, the exact incidence of post-stroke depression, and the time course of the development of depressive symptoms after a stroke, vary considerably across studies-though it appears to be more evident in the months following an infarct.²⁰

Depression and Cardiovascular Illness

Along similar lines, a rich literature now exists detailing a bidirectional relationship between cardiovascular disease and major depression. A myriad of studies have demonstrated that depression is a risk factor for the development of coronary artery disease (CAD), and that it is associated with increases in both the morbidity and the mortality of CAD. Depression predicts myocardial infarction and death, even when controlling for confounding factors which give rise to both CAD and depression (e.g., smoking and other behavioral patterns.¹⁵ Conversely, up to one-third of adults suffering a myocardial infarction will develop depressive symptoms in the year following their cardiac event.¹⁵

Several mechanisms have been proposed to explicate this bidirectional association, including psychosocial and biological explanations. With respect to the former, older depressed individuals may have less healthy lifestyles, thus predisposing them to heart disease; and depressed individuals with established heart disease may be less likely to adhere to disease-modifying interventions. From a biological perspective, depression could catalyze the development of heart disease through centrally mediated changes in heart rate variability, increased recruitment of the HPA axis, increased inflammatory activation, increases in platelet aggregation or through increases in catecholamine release and sympathetic tone.^12,15 Finally, heart disease and depression may well share common genetic predeterminants, such as alterations in the methylenetetrahydrofolate reductase (MTHFR) gene, which is involved in homocysteine metabolism.¹²

Depression and Cancer

As is the case with other medical conditions, depression in the older adult can worsen the morbidity of cancer. Depression is more likely when a diagnosis of cancer is accompanied by the recent loss of a spouse, the prospect of poorly controlled pain or significant functional disability. Certain types of cancer are also more likely to be accompanied by depression in the older adult, including cancers of the pancreas, head and neck, and lung.⁹ Several medications used to treat cancer in the older adult can induce depression (and cognitive changes). These include vincristine, vinblastine, procarbazine, corticosteroids and interferons. Radiotherapy, particularly whole brain radiation, can promote the development of neuropsychiatric symptoms as well.⁹

Depression and Other Medical Illnesses

An assortment of endocrine, pulmonary, renal, and gastrointestinal diseases tend to coexist with depression,¹⁵ and many of these are exacerbated in the older adult. Although somewhat controversial, some authors have posited that declining testosterone levels in older men play a causative role in the development of late-life depression. Others have ascribed a role to decreases in thyroxine levels. Several studies indicate that diabetes and insulin resistance are risk factors for depression.²¹ Depression is also more common in older adults suffering from chronic obstructive pulmonary disease (COPD), especially in the setting of oxygen dependence and during periods of COPD exacerbation.¹⁵ Moreover, depression is quite common in end-stage renal disease (ESRD) with prevalence rates of approximately 21%.¹⁵ There have also been links between depression and irritable bowel disease, osteoporosis, and obstructive sleep apnea.¹² Finally, depression has been linked to hearing and visual impairment in the older adult.²¹

The discussion below specifically focuses on screening for depression in elderly patients. A discussion of depression screening in general is found in Chapter 3.

DEPRESSION SCREENING FOR OLDER ADULTS IN PRIMARY CARE

There are important clinical benefits of screening for depression among older adults in medical settings, such as primary care. First, older adults have the highest risk of suicide of all age groups, which is a major public health concern for communities and the nation,²² and most of these suicidal patients experience depression episodes prior to suicidal attempt. Nearly 50% to 75% of older adults who commit suicide have seen their primary care doctors for medical care during the preceding month, and about 40% saw their doctors during the week prior to their death.²³ Therefore, screening for depression in primary care practice and treating it is a central component of suicide prevention.²² Second, screening may help clinicians identify older patients earlier in their course of depression. Third, screening may identify patients who may have been previously diagnosed with depression but were ineffectively treated—suggesting the need for modification of treatment.

Depression in older adults can be difficult to diagnose in primary care for a number of reasons:^24,25

• 
  

  Multiple medications: Many medications can cause symptoms of depression, such as fatigue, loss of energy insomnia or hypersomnia, appetite change, weight loss, and difficulty concentrating.

  
  
• 
  

  Multiple comorbidities: Depressive screening for older adults is often complicated by comorbid medical conditions that increase in incidence with longevity

  
  
• 
  

  Cognitive or functional impairment: Communication deficits such as hearing impairments, poor vision, and speech problems, or early onset of dementia can all complicate the screening process.

  
  
• 
  

  Reporting bias: Reluctance of older people or family members to complain about problems because of concerns of being ignored, stigmatized, or generating additional, burdensome medical care costs.

CURRENT EVIDENCE AND RECOMMENDATIONS FOR DEPRESSION SCREENING

As many as three-quarters of primary care settings performed mental health screening, including screening for depression for older adults²⁶ in 2010. However, other studies suggest that up to 30~40% of depression cases are missed by primary care providers.²⁷

Depression screening for older adults in clinical practice has been recommended.²⁶ This recommendation is based on some evidence that screening improves the accurate identification of depressed patients in primary care and that treatment of these patients can decrease clinical morbidity. However, trials evaluating the effect of screening on clinical outcomes have shown mixed results: simply providing screening results to primary care clinicians resulted in small benefits, though the communication of screening results in combination with coordinated follow-up and treatment leads to larger benefits.

USE OF SCREENING TOOLS FOR ASSESSING DEPRESSION IN OLDER ADULTS

Given the high caseloads of primary care settings, both clinicians and patients prefer efficient depression screening tools. Below are some screening instruments that have been successfully used for detecting depressive symptoms in older adults during primary care.

PHQ-9

The 9-item Patient Health Questionnaire (PHQ-9) is an effective and efficient depression screening tool for older adults in primary care settings.²⁸ The PHQ-9 was also successfully used to identify depression in homebound older adults.²⁹ Medicare integrated the full PHQ-9 into the revised Minimum Data Set assessments (MDS 3.0) for nursing home residents.³⁰

CES-D

The 20-item standard Center for Epidemiologic Studies Depression Scale (CES-D) has been extensively used and studied, and is considered a reliable valid instrument and a widely recognized research tool for assessing levels of depressive symptoms in adult populations. Specifically in older adults, the CES-D is excellent a screening instrument for major depression.³¹ The CES-D can also be used to screen for both depression and anxiety disorders at the same time in very old populations.³² The abbreviated 10-item version of the CES-D³³ has excellent properties for use as a screening instrument in older adults, while requiring 5 minutes less than the standard 20-item CES-D. However, it should be noted that the CES-D items do not assess the diagnostic criteria items of appetite, anhedonia, guilt, or suicidality.

GDS-15

The Geriatric Depression Scale is a self-report instrument measuring depressive symptoms in older adults.³⁴ The original 30-item version³⁵ has been reduced to a short version (GDS-15) with a yes/no response format to decrease fatigue or lack of focus seen in the elderly.³⁶ The scale is designed to identify symptoms that distinguish depression in the elderly from dementia. The GDS is comprised of the affective (e.g., sadness, apathy, crying) and cognitive (e.g., thoughts of hopelessness, helplessness, guilt, worthlessness) symptoms of depression in the elderly. It contains none of the somatic items (e.g., disturbances in appetite, sleep, energy level, and sexual interest) that are potentially important confounds in older adults.²⁴ Compared with younger depressed adults, depressed elderly tend to present more often with somatic symptoms (e.g., general and gastrointestinal somatic symptoms), agitation and hypochondriasis, and less guilt and loss of sexual interest.³⁷

BDI (or BDI-II)

The Beck Depression Inventory (BDI) is a valid instrument for the diagnosis of depression in older adults.³⁸ It is composed of cognitive, affective, somatic, and vegetative symptoms of depression. The 7-item BDI–PC is an abbreviated primary care version of the BDI–II, but it has not been specifically tested in elderly medical patients.³⁹ This self-report scale is generally regarded as a cost-effective option for screening depression in elderly medical patients, but there are concerns about overlapping symptoms between medical conditions and the depressive somatic symptoms in BDI.⁴⁰ Moreover, some elderly individuals may not be able to complete the scale due to reading difficulty, physical debility, or compromised cognitive functioning.⁴¹

APPROACH TO THE DIAGNOSIS OF DEPRESSION IN THE MEDICALLY ILL ODER ADULT

One of the most challenging tasks in detecting depression in the medically ill older adult is distinguishing the symptoms of true affective illness from those of medical disorder. Fatigue, anergia, anorexia, sleep disturbance and apathy can all be due to medical conditions in the elderly. This can lead to misattribution of the medical symptoms to depression. As such, routine tests to exclude a medical condition causing a patient’s symptoms are recommended. Among these are evaluations of anemia, hypothyroidism and B12 deficiency.¹ Of equal concern, a diagnosis of depression can be missed if a provider attributes these symptoms solely to the comorbid medical condition, since depression often presents in the older adult with prominent somatic symptoms.⁹ One useful approach is to probe for more specific symptoms, such as anhedonia, hopelessness (e.g., hopelessness for the future), frequent crying, pronounced self-reproach/guilt, psychosis, and passive or active suicidal ideation. Structured clinical interviews and selected diagnostic batteries can be helpful in discriminating depression from the symptoms commonly associated with advanced aging, and those of comorbid medical conditions.

CURRENT APPROACHES TO DIAGNOSIS AMONG OLDER ADULTS IN PRIMARY CARE

Screening instruments are not sufficient for diagnosing depression, but do indicate the need for more detailed follow-up by a clinician to determine whether the individual meets diagnostic criteria for a depressive disorder, to explore other possible causes for depression (such as medication or substance use), and assess for co-existing psychiatric disorders.

In some patients, it can be difficult to attain an accurate diagnosis of depression in older adults. There are several things to consider:

(1) Fully use the results from assessment tools: Depression screening instruments inform the clinicians about the likelihood of having depression. Some instruments, such as PHQ-9 or GDS-15, have clear cutoff points that, based on empirical evidence, are highly predictive of the diagnosis of major depression. If the screening score is very far above or below the cutoff point, the chance is high that the screening result can be used to obtain or eliminate a diagnosis, with the understanding that it does not replace a formal diagnosis. If the screening result is close to the cutoff point, further assessment often is needed.

(2) Clinical diagnostic interview: With adequate training, and if time is available, primary care providers can conduct a clinical interview, using either an observer-rated instrument, such as the SCID (Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders [DSM]),⁴² or following a reliable process for assessing DSM diagnostic criteria. Many primary care physicians may not have the time or the requisite skills, and prefer a referral to a mental health or geriatric specialist.

(3) Assessing risk factors for depression: At least half of older adults who present with depression have no previous history of depression, suggesting that different pathological mechanisms may be involved in “late-onset depression.” Late-onset depression may be associated with specific comorbid diseases, poor physical health, cognitive impairment and structural changes in the brain. Older people with depression who have experienced an episode earlier in life are more likely to have a family history of mental illness. Factors that have been associated with increased risk of depression in the elderly are presented in Table 21-2.