There is some confusion in the psychiatric literature regarding the incidence, prevalence, and illness course of psychiatric disorders in late life. This can be partly attributed to a paucity of studies focused on determining the epidemiology of psychiatric illnesses in older adults (+2). Moreover, there is a common belief (and some empirical evidence) that psychiatric disorders (except dementia) are less prevalent in older adults than in younger populations (+3). Yet, these studies have certain limitations. Behavioral and neurovegetative symptoms of psychiatric illness frequently overlap with those of general medical conditions, and increasing age is generally accompanied by increased medical comorbidity, which creates diagnostic conundrums in the evaluation of older adults. In addition, age-related changes in the body and brain, as well as cohort effects, can lead to atypical manifestations of psychiatric illness, resulting in inaccurate or overlooked psychiatric diagnosis (+4). Increasingly, however, it is recognized that special care needs to be given to psychiatric assessment of the older patient. Awareness and subsequent clinical recognition of the differences in symptoms between older and younger age groups may help reduce both psychiatric and medical comorbidity. +Table 1 outlines age-related differences in the clinical features of several major psychiatric illnesses. Estimates of incidence and prevalence as well as prognosis and illness course are included.

It is important to note that available information on psychiatric disorders in older adults is based on research conducted using diagnostic criteria from either the DSM or ICD, both of which were almost invariably developed from models of illness in mostly younger or middle-aged adults. However, it has become increasingly evident that use of these criteria may result in an underestimation of the prevalence of psychiatric disorders in older persons, largely because these criteria have not been validated in older populations (+25). Symptoms of psychiatric illness in older adults do not always correspond with criteria described in the standard diagnostic manuals, a phenomenon also seen and more widely recognized among children and adolescents. Ultimately, as research on age-related changes in the brain expands, we will gain a more sophisticated understanding of differences in clinical symptoms of psychiatric illness as a function of age. Creating age-specific diagnostic criteria on the basis of evolving knowledge of this process will allow clinicians to better identify symptoms, diagnose illness, and devise treatments for management of late-life psychiatric disorders. For now, awareness of variable presentations of psychiatric disorders in older adults, as reviewed in +Table 1, enables the clinician to be more attuned to the psychiatric needs of older adults.

Unlike the disorders described in +Table 1, cognitive disorders are unequivocally more prevalent as age increases, and competency in assessing cognition is an essential skill for clinicians caring for older adults. This encompasses the use of cognitive screening assessments, early detection and education about dementias, treatment of comorbid disorders that masquerade as dementia, attention to family system and caregiver issues, referral to and collaboration with other professionals such as neurologists and neuropsychologists, and reassurance when cognitive changes are the result of normal aging.

There is a common public (and even professional) misconception that dementia is a sine qua non of aging. Although age is the most prominent risk factor for dementia, half or more of adults older than age 85 retain normal cognitive functioning (+26). Cognitive impairments that meet the diagnostic criteria for dementia should never be dismissed as normal aging. Nonetheless, it is also valuable to recognize that there are some changes in cognition that are age-related and not considered pathological. Examples of such age-related cognitive changes include some decline in free recall and consistently slower processing speed. However, more crystallized abilities (vocabulary and general knowledge) tend to remain stable or even, in some circumstances, increase with age (+27). When appropriate, reassurance about benign cognitive changes of aging can allay significant anxiety for older adults, who are increasingly cognizant of illnesses such as Alzheimer's disease.

The most widely recognized standardized cognitive assessment used in clinical settings is the Folstein Mini-Mental State Examination (MMSE) (+28). The MMSE is useful primarily as a screening tool, although it may also be used as a crude proxy for illness progression in persons with dementia. Cutoffs for "normal" versus "abnormal" scores vary by age and educational attainment (+29), in addition to possible cultural variations (+30). Criticisms of the MMSE have included a relative lack of assessment of certain cognitive domains (e.g., executive functioning) and insensitivity to early dementia, especially in older adults with high intelligence and educational attainment (+31).

Other screening cognitive tests have been developed to address possible limitations of the MMSE. The Mini-Cog Test, which uses only three-item registration and recall separated by the Clock Drawing Test, is somewhat quicker to administer than the MMSE, and is useful in certain settings such as primary care and the emergency room, and appears to have sensitivity and specificity for dementia similar to that of the MMSE (+32). Other examples include the Saint Louis University Mental Status examination (+33) and the Montreal Cognitive Assessment (+34) (+Figure 1), which have more extensive memory and executive function assessment than the MMSE.

The gold standard of cognitive assessment is formal neuropsychological testing, typically performed by a specialist in neuropsychology. Referral for neuropsychological testing may be particularly useful if diagnosis is uncertain (e.g., early Alzheimer disease versus dementia of depression), as various forms of dementia exhibit somewhat distinct profiles of relative strengths and impairments early in the course of illness (+35). +Table 2 lists the most common variants of cognitive disorders among older adults, along with clinically distinguishing features including typical neuropsychological profiles. In moderate-to-severe stages, most forms of dementia eventually cause diffuse and nonspecific impairment across multiple cognitive domains. One must also keep in mind that the most troubling symptoms of dementia are often the accompanying psychological and behavioral changes, which affect up to 80% of persons with dementia across the illness course and which frequently increase caregiver burden and the likelihood of patient institutionalization (+55). Evaluation for depressive symptoms, psychosis, sleep disturbance, and agitation is a crucial part of the evaluation of cognitive disorders in older adults, and progress is being made toward establishing reliable diagnostic categorization of such neuropsychiatric symptoms of dementia (+56).

There are several traditionally distinct (although admittedly sometimes overlapping) cognitive domains that can be assessed with neuropsychological testing, and abbreviated assessments of each domain are also possible in the clinical setting. Some of the most frequent cognitive domains as conceptualized by neuropsychology today are described below:

Cognitive testing of older adults should be conducted with several caveats in mind. For example, there is a lack of age-appropriate normative data on many tests for the oldest-old age groups and certain cultural groups. Many older people have sensory or motor impairments that may interfere with certain tasks independently from cognition per se. Subtle cognitive deficits may be very difficult to detect, especially in high-functioning or highly educated older adults. Even the best of cognitive testing may leave questions unanswered, and in these circumstances the trajectory of change in cognition over time can be very informative.

Clinicians may often be asked to evaluate the decision-making capacity of older adults. Competency usually refers to an evaluation derived from legal adjudication. Decisional capacity is the preferred term when describing a clinician's evaluation of a patient's decision-making abilities. Evaluation of decisional capacity is most commonly prompted when a patient declines a recommended treatment (+73). Two other common scenarios involve capacity to drive and to decide on one's living environment (+74). Clinicians are required in a few and allowed in most states to report cases of dementia to motor vehicle departments. There is no clinical consensus for how to determine when a person should have his or her driver's license revoked. Although the safety of the public and patient is clearly important, this must be balanced with the possible adverse consequences of license revocation on a person's sense of independence and practical mobility. On-road driving tests may represent the best available option in questionable cases. Equally contentious at times is the dilemma of when an older adult needs to enter a structured living environment (i.e., assisted living or nursing home), also reflecting basic ethical conflicts between paternalism and respect for autonomy.

Older adults are certainly at increased risk for impaired decisional capacity. Dementia is often the most prominent illness in clinicians' minds when considering impaired decisional capacity. Psychosis and depression may also impair decisional capacity in older adults via emotional factors, such as paranoid delusions or severe hopelessness. However, psychiatric illness and dementia do not invariably impair decisional capacity (+75, +76). Several qualities have been consistently described as necessary for decisional capacity. The four most commonly cited criteria are Because of the increased risk for impaired decisional capacity, it is vitally important to encourage older patients to execute advanced directives proactively before such impairments occur.

Psychiatry is perhaps unique among medical specialties in that it has traditionally emphasized functional impairment as part of its disease constructs, but functional assessments are especially germane in the evaluation of older adults and are an essential component of what distinguishes geriatrics as a subspecialty (+78). Functional impairment increases with age and is often overlooked or marginalized in importance if one focuses too narrowly on symptomatology. All psychiatric assessments of older adults should include a basic history of functioning in both instrumental (e.g., driving, cooking, and managing finances) and basic (e.g., grooming, eating, and walking) activities of daily living. Standardized assessments are available as well, including the Lawton Instrumental Activities of Daily Living scale (+79) and the Older Americans Resources and Services (OARS) Multidimensional Functional Assessment Questionnaire (+80).

Given the link between the use of most psychotropic medications and increased risk for falls (+81), specific assessments of mobility may also be useful for psychiatrists caring for older adults. Perhaps the simplest and most sensitive screening for fall risk is the Timed Get-Up and Go Test (+82). This test entails asking the patient to rise from a chair without using his or her arms, walk about 10 feet, turn around, and sit back down (again without using his or her arms). If the patient takes more than 10 seconds to complete the examination or exhibits obvious unsteadiness, then further evaluation of gait and balance is indicated.

Nowhere is the mind-body connection more evident than in the psychiatric assessment of older adults. Although a medical assessment is an integral part of the initial evaluation for all persons presenting with psychiatric symptoms, it is especially important for older adults. Older adults are at increased risk for delirium (+83), polypharmacy (+84), and many medical conditions, all of which can present primarily or even exclusively with psychiatric symptoms. As providers with both psychological and medical training, psychiatrists are uniquely positioned to provide an integrated medical and psychological evaluation of the older adult presenting with behavioral or psychological symptoms. Even when a medical disorder is not the direct physiological cause of psychiatric illness, the impact of chronic medical illness on quality of life and functional ability is often apparent in the lives of older adults and relevant to psychiatric assessment and treatment (+85). For instance, medical illness is associated with increased risk for suicide among older adults (+86).

As with any medical or psychiatric assessment, the best approach is to start with a thorough history. A medical history is, of course, part of any psychiatric evaluation. We will therefore highlight some special considerations to keep in mind in the initial psychiatric assessment of older adults. Always consider obtaining collateral information, particularly when patients exhibit any cognitive impairment (+87). Even cognitively intact older adults may withhold or fail to recognize important historical information that family members can provide, although balancing the relative validity of contradictory perspectives requires consideration of several possible competing motivations and family dynamics.

Given the prevalence of delirium and psychiatric illness due to a medical condition in late life, it is important for a clinician to keep these diagnoses at the top of the differential diagnosis when conducting psychiatric evaluations of older adults (+88). Obtaining a thorough understanding of the time course of the psychiatric symptoms can be one of the best clues as to whether delirium or a medical condition is implicated in the etiology (e.g., increased suspicion for a medical etiology when symptom onset is acute or when new physical symptoms overlap chronologically with new psychiatric symptoms) (+87). Some cases may be readily apparent (e.g., confusion and hallucinations accompanying 2 days of fever and cough associated with pneumonia), but many require a higher level of vigilance (e.g., depression associated with hypercalcemia due to an occult parathyroid adenoma) (+89). There may also be complex reciprocal relationships between medical conditions and psychiatric symptoms (e.g., the interplay between anxiety and dyspnea in chronic obstructive pulmonary disease) (+90).

The nature and pattern of psychiatric symptoms may also yield clues as to whether they may be due to medical illness. For example, visual, olfactory, or tactile (versus auditory) hallucinations increase suspicion for medical causes (+91). Early morning awakening is a classic symptom of melancholic depression, whereas repeated nighttime awakenings often occur with nocturia due to benign prostatic hypertrophy, persistent pain, or severe gastroesophageal reflux (+92). In addition, cognitive impairment accompanied by inattention and an altered level of consciousness points toward delirium, often underrecognized as a medical emergency (+83).

Most primary psychiatric illnesses (other than cognitive disorders) manifest well before old age, such that the first episode of an illness after age 40 significantly raises suspicion for an underlying medical problem (+93). As in the history of present illness, one should assess whether prior symptoms or episodes could have been related to medical illness. If no prior link to medical illness is identified, it is, of course, important to remember—and remind our medical colleagues—that previous psychiatric illness does not preclude delirium or psychiatric illness due to a medical condition being part of the current presentation.

Although often overlooked in older adults, substance abuse, particularly alcohol and prescription drug abuse (+94, +95), is an important potential contributor to medical and psychiatric problems. Thus, it is always worth evaluating whether substance intoxication, withdrawal (especially postoperatively), abuse, or dependence could be contributing to an older adult's presenting psychiatric symptoms. A lifetime of substance use may lead to multiple medical sequelae in older adults (+96, +97), including various cancers, nutritional deficiencies, and organ failure, all of which may themselves cause secondary cognitive and psychological symptoms.

Some of the most relevant disorders to investigate, because of their contribution to neuropsychiatric symptoms and/or their high comorbidity with psychiatric illnesses include endocrine disorders (e.g., hypo-/hyperthyroidism, hypo-/hyperparathyroidism, and diabetes mellitus) (+98), cardiovascular disorders (e.g., coronary artery disease and cerebrovascular disease) (+99), and rheumatological disorders (e.g., rheumatoid arthritis, osteoarthritis, and polymyalgia rheumatica) (+100). Persistent pain disorders are quite common in older adults, are often comorbid with anxiety and depression, and may increase suicide risk (+101, +102). Taking a sexual history may seem uncomfortable or unnecessary with older adults, but this reluctance reflects an ageist attitude. Sexual functioning is often an important component of the physical and emotional health of older adults, and many psychotropic medications may adversely affect sexual desire and/or performance.

One of the best contributions a physician can make to an older adult's care is a thorough review of all his or her medications (+84). Iatrogenic psychiatric symptoms can arise from medication side effects and drug-drug interactions but often go undetected (+103). Medication assessments should include obtaining a full list of all medications the patient is taking including over-the-counter and herbal medications and supplements, assessing dosing schedules, determining how often patients are really taking medications (looking for over- and underuse), asking about side effects, and checking for drug-drug interactions (+104). If potential problems, such as unnecessary or redundant medications or possible psychiatric side effects, are identified, working with a patient's primary care physician or team to minimize these problems can be invaluable in improving an older adult's medical and psychiatric condition. Particular attention should be given to minimizing the cumulative anticholinergic burden of medications in older adults, often the result of several concomitant medications whose anticholinergic effects are underrecognized (+105). Geriatric care is often as much or more about removing iatrogenic culprits of illness as prescribing additional treatments.

Although psychiatrists typically do not do physical examinations as part of their initial medical assessment, these may be appropriate under certain circumstances. Certainly routine vital signs, including pain severity, weight, and orthostatic blood pressure measurements, may identify medical conditions that may alter one's diagnosis or treatment as well as potentially serious side effects of psychotropic drugs (e.g., anticholinergic or proadrenergic drug-induced tachycardia predisposing to cardiac complications, antiadrenergic-induced orthostasis predisposing to falls, and medication-associated weight gain or loss predisposing to a host of adverse medical consequences) (+106, +107). Although vital signs may be an important clue to medical emergencies, one must also keep in mind age-associated changes in the physiology of vital signs (e.g., infection in the absence of fever or increased carotid sinus sensitivity) (+108, +109).

Examinations for tardive dyskinesia and parkinsonism are particularly important when a clinician is prescribing antipsychotics for older adults, given that age increases the risk of these motor complications (+110). The mental status examination of older adults consists largely of the same components included in the assessment of younger adults, with the notable exception that more extensive cognitive assessment, including evaluation for delirium, is often needed. Evaluation for suicidal ideation should be at the forefront of clinical assessments in light of the elevated rates of completed suicide among older adults, mostly accounted for by white men (+111).

Depending on an older adult's presentation and the treatment options being considered, a variety of laboratory tests may be indicated in the medical assessment of an older adult presenting with psychiatric symptoms. Laboratory tests to consider routinely obtaining include the following: 1) thyroid function tests (e.g., thyroid-stimulating hormone), because thyroid disease may cause depression, anxiety, and cognitive impairment (+112); 2) electrolyte levels, because alterations of these may significantly affect the central nervous system (e.g., hyponatremia-induced delirium due to the syndrome of inappropriate antidiuretic hormone or selective serotonin reuptake inhibitor treatment) (+113); 3) renal function tests (blood urea nitrogen and serum creatinine), because renal function is often reduced by age or age-associated illness, which may significantly affect the pharmacokinetics of medications (+114); 4) hepatic function tests (e.g., alanine aminotransferase and aspartate aminotransferase), because psychopathological conditions may significantly contribute to chronic liver disease via alcohol use and comorbid hepatitis C infection (+115); 5) complete blood count, because anemia may explain fatigue/depression, leukocytosis may point to an undetected infection, and many psychotropic drugs have hematopoietic side effects (+116); 6) lipid panel and fasting glucose, particularly if a patient is already taking or considering starting medications that could cause metabolic syndrome (+117); 7) rapid plasma reagin to test for syphilis and HIV antibodies if the patient has risk factors and consents, because prejudicial views of older adults may lead clinicians to dismiss their current or past sexual activity (+118); 8) urine drug screen to assess for substance use and breathalyzer or blood alcohol level, especially in emergency room settings, where, again, age may inappropriately lower clinicians' vigilance for substance use (+95); 9) urinalysis, because urinary tract infection may present exclusively with mental status changes in older adults (+119); 10) vitamin B₁₂ and folate levels, because deficiencies of these increase with age and may contribute to both depression and cognitive impairment (+42); and 11)drug levels of any prescription drugs patients are taking, especially those with low therapeutic indices (e.g., lithium, anticonvulsants, digoxin, and tricyclic antidepressants).

Other diagnostic tests to be considered in certain situations include the following: 1) chest radiograph for respiratory symptoms, signs of delirium, or fever; 2) neuroimaging (magnetic resonance imaging or computed tomography) to evaluate for stroke, mass lesions, or normal pressure hydrocephalus (+120); 3) electrocardiogram for acute mental status change, cardiorespiratory symptoms, or consideration of psychotropic agents that may alter cardiac conduction (e.g., lithium, tricyclic antidepressants, and certain antipsychotics) (+121); 4) electroencephalogram for paroxysmal symptoms that could be due to seizures, unusual or subacute dementias (e.g., Creutzfeld-Jacob disease), or diagnostic confusion between delirium versus dementia (although generalized slowing is present in both delirium and more advanced cases of dementia) (+122); and 5) lumbar puncture in patients with atypical cognitive disorders, fever without a clear etiology, or suspected neurosyphilis (+123).

Many invasive (e.g., lumbar puncture) or expensive (e.g., magnetic resonance imaging) tests may be unnecessary with a careful history and physical examination. Although the use of neuroimaging is clinically routine in the assessment of delirium, dementia, and late-onset psychiatric disorders, there is debate as to its cost-effectiveness and clinical yield in the absence of historical (e.g., acute onset) or physical examination findings (e.g., focal neurological symptoms) that raise suspicion for certain neurological disorders (+124, +125). The explosion of research in functional neuroimaging has yet to yield significant clinical applications, although this is anticipated in upcoming years with promising techniques such as in vivo amyloid imaging (+126). Currently, the only psychiatric application of functional neuroimaging approved by Medicare is the use of positron emission tomography in differentiating Alzheimer-type versus frontotemporal dementia in patients with clinically unclear illness (+127).

Although medical disorders and neurobiological changes associated with aging are at the forefront of considerations in psychiatric assessments of older adults, a balanced biopsychosocial perspective remains crucial as well. The comfort level of older adults (at least the current cohort) with mental illness and psychiatric evaluation must be considered to develop the rapport needed for any other aspect of assessment (+128). Older adults also tend to face somewhat unique psychosocial stressors that may have tremendous relevance for the symptoms at hand. For instance, acting as a caregiver for a disabled relative is associated with high rates of depression and is often used in research as a prototypical model of stress (+129). Accumulating medical (and at times psychiatric or cognitive) illnesses often result in disability and loss of independence (including forced moves to senior living communities) that challenge the resilience and self-concepts of older adults. Other forms of loss, including deaths of friends and family, are common, and grief reactions may ensue that warrant careful attention for clinically significant depressive symptoms (+130). DSM-IV-TR (+131) outlines several features that are believed to help distinguish normal versus abnormal bereavement, such as suicidal ideation, persistent worthlessness, and severe, persistent functional impairment. Such losses and personal disabilities may trigger psychological struggles regarding one's own mortality that may be very relevant in psychotherapy with older adults, but clinicians often avoid this topic because of discomfort (+132). Another sometimes surprising stressor for older adults is retirement, which may challenge persons whose self-identity and structured activities center around their occupation (+133). Although these challenges confronting older adults may seem daunting, it is also worthwhile to note the positive coping strategies used by many older adults so that clinicians also work to facilitate the person's innate strengths. Possible psychological benefits of increased age include, on average, better emotional regulation (+134) and, according to many cultural beliefs, increased wisdom (+135).

For these and other reasons, social support is an important predictor of the health outcomes of older adults (+136, +137). Psychiatric assessment and care of older adults thus requires careful attention to social networks and often includes family members in the assessment and treatment of the older adult in a manner akin to that used for children and adolescents (with important similarities and differences in the dynamics encountered, including struggles for autonomy and role reversals). The increasing diversity of the aging population in the United States also warrants increasing attention to culturally competent psychiatric assessments, including awareness of linguistic barriers, differing cultural concepts of both aging and illness, and possible roadblocks to health care access (+138).

In summary, psychiatric assessment of older adults is both challenging and rewarding. Therapeutic nihilism and ageism may taint the enthusiasm some clinicians have for working with older adults affected by neuropsychiatric illness, but careful attention to the issues summarized briefly in this review can help ensure that such clinical experiences are rewarding for both older patients and the clinicians who care for them. Psychiatric assessment of older adults should remind psychiatrists of why they are often called upon to set the example for biopsychosocial care in clinical medicine. Medical comorbidity requires attention to clinical issues that overlap with other fields of medicine, unique psychosocial stressors, and the stigma of mental illness associated with aging require careful attention to the human relationship between clinician and patient, and the importance of social support and extended family involvement call for a broad, systems-based approach to the psychiatric care of older adults. Integrating these various components of assessment and diagnosis will be crucial for the wide variety of psychiatrists who will be increasingly called upon to care for the rapidly expanding population of older adults.