National Institute on Aging > Research > Conferences, Workshops & Meetings
Print this page E-mail this page

Unexplained Fatigue in the Elderly

An Exploratory Workshop sponsored by the National Institute on Aging

June 25-26, 2007




    “Fatigue” or “exhaustion” is a common clinical complaint among older adults for which multiple definitions and measurement instruments exist. Fatigue is often a symptom of underlying medical or psychiatric illness.  Cancer, heart disease, chronic lung disease, hypothyroidism, multiple sclerosis, and rheumatoid arthritis are just a few of the many diseases or conditions that may present with fatigue; however, for many older persons, no physiologic or psychological explanation is identified, and fatigue becomes a diagnosis or a syndrome with which the elderly must attempt to cope across all activities of daily living. In addition to its direct effects on mood and function, persistent fatigue may also elevate risk for later adverse outcomes [1]. In the absence of overt medical conditions that could explain fatigue, it is likely that multiple factors may play a role in its pathophysiology. Metabolic, muscular, inflammatory, neurologic, and psychological factors may all contribute to unexplained or idiopathic fatigue. Understanding these factors may lead to identification of subtypes of fatigue, for which specific interventions may be proposed. Furthermore, it is clear that describing fatigue with qualifiers, such as cognitive fatigue or muscular fatigue, can facilitate a better understanding of this broad concept. 
    In June, 2007, the NIA convened this exploratory workshop to initiate a broad-based scientific dialogue exploring the clinical problem of “idiopathic” or “unexplained” fatigue in older adults. Participants represented a diverse array of scientific and research fields, spanning clinical, translational, and basic scientists from a variety of biological, psychological, and social science backgrounds. The working objectives of the program were to 1) describe the scope of the problem of fatigue in older adults; 2) review current definitions, dimensions, and correlates of fatigue; 3) describe diseases or conditions with which fatigue is associated, exploring common characteristics, pathophysiologic mechanisms, and interventions; 4) explore potential pathophysiologic mechanisms underlying unexplained fatigue in older adults; and 5) assess existing instruments for measuring fatigue and related constructs such as functional measures and physiologic variables.


Neil Alexander, MD
University of Michigan

Kirsten Avlund, PhD
University of Copenhagen

Marcas Bamman, PhD
University of Alabama School of Medicine

Shalender Bhasin, MD, PhD
Boston Medical Center

Dedra Buchwald, MD
University of Washington

Paulo Chaves, MD, PhD
Johns Hopkins University

Christopher Christodoulou, PhD
State University of New York Stony Brook

John DeLuca, PhD, ABPP
UMDNJ-New Jersey Medical School

Basil Eldadah, MD, PhD
National Institute on Aging

Bill Evans, PhD
University of Arkansas for Medical Sciences

Roger Fielding, PhD
Tufts University

Bret Goodpaster, PhD
University of Pittsburgh

Jack Guralnik, MD, PhD
National Institute on Aging

Evan Hadley, MD
National Institute on Aging

Susan Hardy, MD, PhD
University of Pittsburgh

Paul Jacobsen, PhD
Moffitt Cancer Center and Research Institute

Jane Kent-Braun, PhD
University of Massachusetts Amherst

Lauren Krupp, MD
State University of New York Stony Brook

Andrea Lacroix, PhD, MPH
University of Washington

Christiaan Leewenburgh, PhD
University of Florida College of Medicine

Jan Moynihan, PhD
University of Rochester

Susan Murphy, ScD, OTR
University of Michigan

Susan Nayfield, MD, MSc
National Institute on Aging

Anne Newman, MD, MPH
University of Pittsburgh

Barry Oken, MD
Oregon Health and Science University

Marco Pahor, MD
University of Florida College of Medicine

Denise Park, PhD
The University of Illinois

Catherine Sarkisian, MD, MSPH
University of California Los Angeles

Michael Sharpe, MA, MD
University of Edinburgh

Arthur Stone, PhD
State University of New York Stony Brook

Stephanie Studenski, MD, MPH
University of Pittsburgh School of Medicine

George Taffet, MD
Baylor College of Medicine

Joseph Verbalis, MD
Georgetown University Medical Center

Sonja Vestergaard, PhD
National Institute on Aging

Jeremy Walston, MD
Johns Hopkins University School of Medicine

Ellen Whyte, MD
University of Pittsburgh School of Medicine

Alex Zautra, PhD
Arizona State University



    Several definitions of fatigue exist, ranging from broader definitions that encompass subjective and objective dimensions, to narrower definitions that consider discrete physiological mechanisms. Some definitions are based on self-reported symptoms of tiredness or exhaustion, while others are based on objective measurements of physical or cognitive performance. Arguably, fatigue should result in some kind of functional alteration.
    The measurement of fatigue is complex. As articulated by Dittner et al. [2], “No two scales measure exactly the same thing. Some measure phenomenology and others measure fatigue severity or impact, while many assess a mixture of all these.” Fatigue results in functional impairment and is also caused by functional impairments. When an individual has less ability, it takes more energy to perform a task. 
    Fatigue can be studied in relation to level of activity, which is a product of effort and time. Two individuals may report the same level of fatigue in relation to different levels of activity; for example, one individual may be fatigued after performance of an activity of daily living (ADL), while another may be fatigued only after a full day’s work. This distinction is important for intervention studies, as outcomes should consider fatigue in the context of function. For example, consider an intervention that improves patients’ function such that they can perform more work before reporting the same level of fatigue. If the outcome measure is function, the intervention appears successful. If the outcome measure is fatigue level, the intervention appears ineffective. Conversely, consider a subject who performs the same amount of work after an intervention but reports a lower level of fatigue. If the outcome measure is function, the intervention appears ineffective; if the outcome measure is fatigue level, the intervention appears successful. Therefore, a useful approach for evaluating efficacy may be to assess fatigue per level of activity.

Epidemiology of Fatigue
    Prevalence estimates of fatigue can be influenced by several factors, including varying methods used to classify fatigue, arbitrary cutpoints for scales, shifts over time in diagnostic conventions, population characteristics (inclusion of special populations), and external events, disasters, and major stressors. Without consideration of these factors, asking what the prevalence of fatigue is may be like asking how long a piece of string is [3].
    Fatigue prevalence in published studies ranges from 5% to almost 50%. Most studies of fatigue prevalence show a female preponderance, with females reporting fatigue 1.2 to 2.3 times more frequently than males. Age is a risk factor for chronic fatigue in some studies. Clark et al. [4] demonstrated that the risk for persistent (> 2.5 years) chronic fatigue is associated with age greater than 38 years at initial presentation. Cognitive impairment, anxiety, inflammation-related chronic conditions (eg. cancer, rheumatoid arthritis), debilitating symptoms (eg. dyspnea) and end-of-life scenarios can identify special populations of older adults at high risk of fatigue. Depression, disease burden, and functional status are likely to interact in the perception and impact of fatigue in later life. 

A Brief Historical Overview of Fatigue Research
    Fatigue was generally unrecognized in the medical literature until the mid-1800s with industrialization. George Beard coined the diagnostic term “neurasthenia” as a disease characterized by general malaise, debility, poor appetite, fugitive neuralgic pains, hysteria, insomnia, hypochondriasis, disinclination for consecutive mental labor, and severe and weakening attacks of headaches. It was viewed as a disease of the nervous system resulting from “overexertion of the brain.”
    Around 1900, clinicians were already struggling with two issues still confronting researchers today: the blurred distinction between fatigue and depression, and the difficulty of differentiating between perceived and observed exhaustion. At the same time, psychiatry was moving away from the disease concepts of Kraepelin and more toward a symptom-based nosology. As a result, fatigue became a symptom synonymous with tiredness. Fatigue as a disease was included in the Diagnostic and Statistical Manual (DSM) versions I and II, but was removed in the DSM-III (1980). It remains in ICD-10 as neurasthenia.
    Other fatigue-based illnesses have also been described throughout modern history. Irritable heart, effort syndrome, and neurocirculatory asthenia were recognized during wartime and were also attributed to psychiatric etiology. Others include myalgic encephalomyelitis, epidemic neuromyasthenia, Iceland disease, atypical poliomyelitis, and chronic fatigue syndrome (formally recognized in 1988 in the US).
    The multidimensional nature of fatigue has been appreciated for a long time. Mosso [5] described four components of fatigue—behavior (decrement in performance), feeling state (perception), mechanism, and context. He found that perception does not correlate with performance. The dichotomy of central vs. peripheral fatigue dates to least the turn of the 19th century. According to a current description [6] peripheral fatigue involves an end organ (muscle), while central fatigue is a “failure to initiate and/or sustain attentional tasks (‘mental fatigue’) and physical activities (‘physical fatigue’) requiring self motivation (as opposed to external stimulation).” The distinction may not always be clear. 



Fatigue can be a symptom of underlying medical disease. Commonly cited conditions include cancer, multiple sclerosis, rheumatoid arthritis, lupus, sleep disorders, HIV/AIDS, depression, heart failure, chronic pain conditions, and following stroke or traumatic brain injury. The manifestations and mechanisms of fatigue in each of these conditions are probably diverse; however, many conditions may share common pathophysiologic processes. For this workshop, only a few of many conditions could be explored.

    Fatigue associated with cancer can present as a disease symptom, as an acute side effect of treatment, or as a persistent symptom after completion of treatment. Cancer-related fatigue (CRF) syndrome is a more recently defined entity consisting of clinically significant fatigue of 2 weeks’ duration associated with 5 additional symptoms (including non-restorative sleep and active disruption of daily activities by fatigue), and is not a consequence of cancer, cancer treatment, or psychiatric disorder according to clinical judgment[7]. Among 40 studies of cancer-related fatigue, the majority found no relationship between age and fatigue [8]. Possible explanations are that 1) older patients are not more susceptible to fatigue; 2) older patients are more susceptible to fatigue but they may not report it because they have different expectations for it (e.g., they expect to feel tired or they ascribe it to aging); or 3) older patients are more susceptible to fatigue, but they may be less bothered by it because they have fewer competing demands for their energy (self-pacing). 
    Possible causes of cancer-related fatigue include anemia [9], physical inactivity/deconditioning [10], central nervous system toxicity of cancer treatment (“chemobrain”) [11]; altered immune function [12]; or cancer-related symptoms themselves (for example, menopausal symptoms in breast cancer [13].)
    Different chemotherapeutic agents may be associated with different levels of fatigue, but the challenge is to separate concomitant side effects from specific properties of the chemotherapeutic agent. The most reliable finding in chemotherapy-associated fatigue is a dose-response relationship—if a drug is associated with fatigue in a particular individual, a greater dose is associated with greater fatigue in that individual.

Future questions:

  • What are the mechanisms underlying the multiple putative causes of cancer-related fatigue?
  • What factors predict development of persistent fatigue in cancer survivors? Is family history of depression or fatigue one such factor? Do physician-patient interactions play a role?
  • Can genetic variability explain differences between individuals in susceptibility to treatment-related fatigue?
  • How can we develop empirically-based approaches to treatment selection for management of cancer-related fatigue?

Multiple sclerosis and other neurological disorders
    Fatigue is a major symptom of many neurological disorders of both the central nervous system (e.g., multiple sclerosis, Parkinson’s disease) and peripheral nervous system (e.g., muscular dystrophy, post-poliomyelitis syndrome). Fatigue associated with neurological disorders is challenging to study because its presence and severity are under-recognized, it is difficult to measure, its pathophysiology is unclear, and its treatments are incomplete.
    Self-report measures of fatigue are easy to administer and readily available, but they are entirely subjective. In addition, patients often have limited insight to distinguish different types of fatigue (mental, physical, social, etc.) Performance measures are objective, but invariably, in the field of multiple sclerosis, performance-based measures do not correlate with self-reported fatigue. This is particularly true with regard to cognitive performance. Patients with multiple sclerosis demonstrate impaired sustained motor force generation, impaired recruitment of motor pathways, metabolic changes in skeletal muscle, and cognitive fatigability, but none of these correlate well with self-reported fatigue.
    Independent cross-sectional predictors of fatigue include depression, pain, quality of life, disrupted sleep, sense of loss of control over one’s environment, and lower education level [14-16]. Longitudinal predictors of fatigue after one year include baseline fatigue, pain, mood, and neurologic impairment. All together, these account for only 36% of fatigue variance, so a large amount of fatigue variance is due to unidentified factors.
    Proposed mechanisms underlying fatigue in multiple sclerosis include cytokine disturbances (particularly, α-interferon, β-interferon, γ-interferon, interleukin-6, interleukin-1, and tumor necrosis factor-α); hypothalamic-pituitary-adrenal axis abnormalities; axonal injury; and brain atrophy. Neuroimaging studies of patients with MS show differences in brain metabolism between fatigued and non-fatigued individuals as assessed by glucose utilization and functional MRI.
    Management of fatigue includes education (encouraging exercise, practicing energy conservation, avoiding isolation, counseling); addressing co-morbidities (e.g., depression, anxiety, sleep disturbance, pain); and medication: modafinil, amantadine, pemoline, and aspirin have all been shown to have modest, though inconsistent, benefit.

Future questions:

  • On which bases can a typology of fatigue be established? For example, according to measurement (self-report vs. performance, unidimensional vs. multidimensional)? Locus (motor vs. cognitive)? Etiology (primary vs. secondary)?
  • Can clinical situations where fatigue occurs predictably, acutely, and reversibly be used to study underlying mechanisms of fatigue?

Depression and sleep disorders
    Fatigue is a common complaint among patients with major depression. Fatigue is most strongly associated with deficits in social functioning and work productivity. Pathophysiologic studies of major depression in chronic fatigue syndrome have shown increased perfusion in the right thalamus (suggestive of increased vigilance to cognitive and motor tasks), increased frontal subcortical white matter hyperintensity, limbic dysfunction, lower levels of corticotropin releasing hormone (CRH), and increased levels of interleukin-1, interleukin-2, interleukin-6, and tumor necrosis factor [17;18]. Modafinil improves fatigue acutely in late life major depression, but this effect disappears after 6-8 weeks of treatment.
    Fatigue, depression, and sleep disturbances are intimately connected. Depression may lead to deconditioning and sleep disturbance, which in turn can lead to fatigue; however, sleep disturbance can also lead to further depression, possibly via inflammatory mediators. Alternatively, depression and fatigue may feed each other, with both arising from a common third factor. One group examined four models describing the relationships among disease severity, sleep disturbance, depression, and fatigue in multiple sclerosis [19]. The best-fitting model showed depression, sleep disturbance, and disease severity leading independently to fatigue.

    Fatigue is common in stroke survivors and is separable from post-stroke depression. It is rated as one of the worst residual symptoms in up to half of adult stroke survivors [20] and predicts dependency in activities of daily living (ADLs) and death [21]. Possible mechanisms of post-stroke fatigue include lesions in basal ganglia, thalamus, or brainstem, diffuse neuronal injury, inflammation, abnormalities of the hypothalamic-pituitary-adrenal axis, and physical deconditioning.

Chronic pain conditions
    Chronic pain affects half of community living older adults and three-fourths of long term care residents. Pain associated with osteoarthritis is the leading cause of disability among older adults [22]. Contributors to functional decline in osteoarthritis also ADL disability [23], depression and sleep disturbance [24], and physical inactivity and anxiety [25]. All of these factors are also associated with fatigue. Retrospective self-report of fatigue can be influenced by peak or recent symptoms, so real-time assessment may be more accurate—what has been called ecological momentary assessment [26]. Most interventions for chronic pain conditions tend to focus exclusively on pain, but more attention should be given to treating associated symptoms such as fatigue.

Future questions:

  • What are the main predictors of fatigue in older adults with chronic pain with or without co-morbidities?
  • What is the relationship between sleep quality and daytime momentary pain and fatigue?



Chronic Fatigue Syndrome (CFS)
    Chronic fatigue syndrome is often defined according to the case definition of Fukuda et al. [27] CFS affects 1-2 million in the US; 60-70% are female; patients are usually between 20-50 years old. Risk factors include female gender, less education, lower social class and income, ethnic minority group membership. Most symptoms reported in CFS are, in fact, complex symptoms requiring more exploration than simply noting their presence. For example, poor sleep can be associated with defined sleep disorders such as restless leg syndrome or sleep state misperception (a discordance between actual sleep measured by polysomnography and a subject’s perception of that sleep), as well as other entities, such as pain, environmental sensitivity (e.g., noise), mental health problems, and polypharmacy. Psychiatric disorders are ~3-5 times more prevalent in CFS than in the general population. Predictors of resolution of CFS are younger age, shorter duration of fatigue, and absence of psychiatric co-morbidity.
    Patients with CFS are often misdiagnosed. Psychiatric disorders and sleep disorders are the most common misdiagnoses. In addition, drowsiness and true muscle weakness must be distinguished from fatigue. It is challenging to distinguish performance declines due to fatigue from declines from other aging-related conditions. Physicians are often reluctant to diagnose CFS because many believe that doing so could be disabling and self-fulfilling to patients; however, the majority of patients feel that a diagnosis is enabling.
    It is important to consider the context in which fatigue occurs. For example, environmental, family, social, occupational, and personal factors and stressors may all contribute to an individual’s fatigue.
    The population with CFS comprises a small portion of a much larger population of individuals affected by prolonged fatigue. CFS overlaps with several other disorders including fibromyalgia, irritable bowel syndrome, temperomandibular disorder, interstitial cystitis, multiple chemical sensitivity, tension headache, chronic low back pain, chronic pelvic pain, post-concussion syndrome, and chronic Lyme disease. Each entity tends to fall under the purview of a particular medical specialty. All of these entities encompass varying degrees of fatigue, sleep disorder, pain, and disability.
    Studies attempting to find biomarkers of CFS have inconsistently shown higher antibody titers, natural killer cell dysfunction, or HPA axis abnormalities in CFS. Recent neuroimaging studies point to possible structural or functional differences compared to controls, such as decreased density of serotonin transporters in anterior cingulated gyrus, decreased volume of gray matter which may be linked to reduced physical activity, and more extensive activation of brain regions to process challenging auditory information.
    Many factors have been proposed to explain CFS, including amplification or perceptional disturbance, atypical psychiatric disorder, stress-distress disorder, autonomic hyperactivity, neuroendocrine disorder, genetic vulnerability, central nervous system disorder, viral infection, and immune dysfunction. No theory explains all or even most cases.
    Among randomized controlled trials for CFS, the only interventions demonstrated consistently to have clear benefit are exercise, cognitive behavioral therapy, and possibly steroids.

Future questions:

  • What are the effects of aging on outcomes in CFS?
  • Can highly active or energetic individuals serve as a clinical model for studying fatigue?

    Frailty refers to a clinical syndrome whose hallmark is increased vulnerability to stressors secondary to decreased physiologic reserve. Frail older adults are at high risk for major outcomes, including disability, morbidity, and mortality.
    Fatigue has been regarded as an integral part of the phenotype of frailty in older adults.  Operational definitions of frailty have included a component of self-reported exhaustion or, more broadly, fatigue. In the Cardiovascular Health Study [28], the fatigue component of the phenotype was measured according to two items from the Center for Epidemiological Studies—Depression scale [29]: “everything I did was an effort” and “I could not get going.” In the Women’s Health and Aging Studies [30], the fatigue component was considered present based on the presence of low usual energy level (score 3 in an visual analog scale ranging from 0–10), self-report of “I felt unusually tired in last month” (most or all the time), or “I could not get going in the last week” (most or all the time). 
    There may be synergy of risk factors for frailty; for example, cardiovascular disease or anemia alone are moderate risk factors for frailty, but both combined they substantially enhance the likelihood of frailty. Community-dwelling older subjects with anemia report being exhausted with equal frequency as those without anemia; however, however, they are significantly more likely to report other components of frailty, including decreased physical activity and weakness [29]. This lack of association between non-severe anemia and fatigue may be due an effect modification by physical activity; i.e., anemia may promote behavioral modifications, including reduction in spontaneous physical activity levels, which in turn may lead to resolution of fatigue although while simultaneously promoting further deconditioning and acceleration of the frailty process. This gradual pathway, which remains to be more fully studied, is consistent with recent data showing that in the frailty process hierarchy, exhaustion is often preceded by reduction in physical activity and weakness [31].

    Self-reported tiredness in response to specific daily activities can be assessed by the Mobility-Tiredness scale and Lower Limb-Tiredness Scale [32-34]. These measures are qualitative and do not include information about severity of each disability. Self-reported tiredness in daily activities is predictive of onset of disability, walking limitations, use of health and social services, decline in physical activity, and mortality. This is true in both young-old and old-old populations, and in different geographic localities [33;35]. Factors related to tiredness, but that do not explain the association between tiredness and above outcomes, include social position, comorbidity (specific diseases, depressed mood, medications), and physiologic and functional impairment (muscle strength, pain, cognitive performance, aerobic capacity, walking limitations) [36].
    An intervention study evaluated  interdisciplinary education of general practitioners and other community figures and of health care professionals involved in preventative home visits for 75+ year old individuals. The intervention consisted of education focused on recognizing tiredness in daily activities and intervening when possible, a multidimensional geriatric assessment, and referral to general practitioner when appropriate. There was less functional decline, less self-reported tiredness, and less use of nursing homes. The effect was greater in 80 year-olds vs. 75 year-olds, and in women more than men [37].

Future questions:

  • Which social, psychological, physiological, or health factors explain the association between tiredness and functional decline?
  • Is tiredness related to other indicators of aging, such as biological indicators?
  • Is tiredness in midlife related to functional decline in old age?
  • Is the predictive power of measures of function and fatigue the same among different ages or between sexes?



      Multiple biological and psychosocial factors may contribute to or influence fatigue. Biological mechanisms include changes in skeletal muscle function, cardiovascular impairment, anemia, dehydration and electrolyte disorders, inflammatory mediators, and nutritional deficiencies. Psychosocial mediators include depression, pain, positive and negative affect, and interpersonal processes. A few of these factors were considered for further exploration.

Skeletal muscle changes
    Skeletal muscle is the single largest organ in the body, comprising 45% of body mass. Skeletal muscle is also a significant reserve of energy, comprised by amino acids stored as protein. Therefore, it is possible that a reduction in skeletal muscle mass could influence the availability of energy, the ability to perform work, and the feeling of fatigue with effort.
    The functional unit of muscles is the sarcomere, a series of actin and myosin filaments where contraction takes place in response to cytosolic levels of calcium. Three isoforms of myosin determine speed of contraction: type I (slowest), type IIa (intermediate) and type IIx (fastest). The maximum capacity to generate force is dependent on muscle cross-sectional area, intrinsic force-generating ability of the muscle fiber, and ability of the nervous system to recruit motor units.
    Sarcopenia is defined simply as age-associated loss in muscle mass, but it is operationalized by different researchers in different ways. One such definition is appendicular skeletal muscle mass, normalized to height, less than 2 standard deviations below the mean of a younger sex-matched reference group [38]. Sarcopenia begins around the 5th or 6th decade, but functional precursors of sarcopenia begin much earlier. Factors contributing to diminished muscle mass include malnutrition, anorexia, cachexia in disease states, deconditioning, low testosterone, and growth hormone.
    In longitudinal and cross-sectional studies, muscle strength decreases with age, but sarcopenia does not represent the entire spectrum of age-related muscle pathophysiology. Loss in strength does not parallel loss in muscle mass—strength declines later than does skeletal muscle mass. Furthermore, sarcopenia does not address qualitative changes in muscle; eg. specific force, attenuation associated with increased accumulation of intramyocellular lipid, velocity components that occur with age, or phenotypic changes in muscle (e.g.,. loss of type II units with preservation of type I units). There is debate over whether strength or power is the most appropriate measure of muscle function. Age-related declines in muscle strength (force) occur later in life than do declines in muscle power (force times velocity). In older adults, leg power is more closely associated with gait speed and self-reported physical disability than leg strength.
    Aerobic capacity represents the rate of oxygen consumption during activity, and is determined predominantly by skeletal muscle function. With aging, the oxygen cost of activities becomes a greater percentage of an individual’s maximum aerobic capacity (VO2max). For example, a sedentary 75 year-old healthy woman has a VO2max of approximately 1.2 L/min, but the oxygen cost of climbing a flight of stairs is approximately 1.5 L/min. Both younger and older subjects self-pace activity to about 50% of their VO2max. Therefore, reduced aerobic capacity should translate to reduced spontaneous activity with age. However, peak VO2 is only modestly related to level of daily physical activity [39].

Future questions:

  • What is the relationship between muscle mass (sarcopenia) and fatigue?
  • What is the relationship between adiposity and fatigue? To what extent is the effect of adiposity on fatigue mediated by fat mass alone vs. metabolic mediators; for example, adipokines?

    Mitochondria are the major source of energy production in the body. ATP generated by mitochondria is the biological currency of energy used ultimately to power all metabolic activities. Mitochondrial dysfunction can lead to reduced energy production, which could result in fatigue. In addition, decreased oxidative metabolism in mitochondria leads to increased anaerobic metabolism and lactic acid production. Acidosis in skeletal muscle can be perceived as muscular fatigue. 
    The mitochondrial theory of aging proposes that oxidants or free radicals produced in mitochondria lead to mitochondrial DNA damage, leading to mutations, defective encoded proteins, and impaired mitochondrial function. A mutant mouse devoid of DNA replication proofreading ability, which led to increased mitochondrial DNA mutation frequency, exhibited an aging phenotype including sarcopenia, osteoporosis, and reduced aerobic capacity at older age.
    There are two populations of mitochondria in skeletal muscle: subsarcolemmal (near muscle cell membrane) and intrafibrillar (embedded deep within myofibrils); the latter are located near fat stores, have higher oxygen consumption, but also higher free radical production and DNA damage and greater antioxidative adaptations to counteract damage. Previous literature did not make the distinction between these two populations, which may explain discrepancies in findings.
    Mitochondria under oxidative stress release cytochrome-C and apoptosis-inducing factor leading to apoptosis of cells. This process may be affected by upregulation of antiapoptotic proteins or expression of inflammatory mediators. Damaged mitochondria are removed from otherwise healthy cells through autophagy; however, the efficiency of this process decreases with age. 
    Exercise leads to increased mitochondrial size and number. Different components of mitochondrial structure or electron transport chain have different responses to exercise. In ~70 year old subjects after 16 weeks of moderate exercise, there was a ~50% increase in electron transport chain activity, which was thought to be due to increased proliferation of mitochondria [40].

Future questions:

  • Does aging lead to mitochondrial dysfunction? How should “dysfunction” be defined?
  • Does exercise-induced proliferation of mitochondria propagate damaged mitochondrial DNA?
  • How can we develop standardized methods for assessing mitochondrial function in vivo?
  • What is the relationship between mitochondria and acute or chronic skeletal muscle fatigue?

Neuromuscular mechanisms
    Neuromuscular fatigue is a focused conceptualization of fatigue based on skeletal muscle performance. Neuromuscular fatigue can be defined as a decline in skeletal muscle function with repeated effort.         Muscle function can be characterized in terms of strength, power, activation (central vs. peripheral), contractile properties (fast vs. slow), energetics (oxidative vs. glycolytic), fatigue resistance (fall of maximal force), and endurance (time to task failure).
    Changes in muscle function with old age include decreased number of motor units, decreased maximum motor unit discharge rate, no changes in ability to fully activate muscle; decreased muscle size, decreased muscle strength, increased intramuscular fat (both intramyocellular and fat deposits outside muscle cell), equivocal changes in specific strength (force per cross-sectional area); no change in peripheral activation (excitability of neuromuscular junction in muscle membrane), decreased type II:type I fiber area, decreased contractile speed, equivocal change in oxidative capacity (matched for health and physical activity level), and unclear changes in blood flow.
    Central activation is assessed by measuring the difference in force generated between maximum voluntary contraction and electrically stimulated contraction. Peripheral activation is assessed by the electrical signal in response to a single stimulus across the belly of a muscle. Energetics is assessed by phosphorous magnetic resonance spectroscopy (MRS). Contractile function is assessed by the amount of force produced by muscle in response to single or multiple stimuli, which provides information indirectly about excitation/contraction coupling and calcium handling. Phosphorus MRS can assess pH, concentrations of α-, β-, and γ-ATP, phosphocreatine, inorganic phosphate, ADP, AMP, and diprotonated inorganic phosphate (H2PO4-), a metabolite reputed to cause fatigue.
    There is disagreement among multiple studies on muscle fatigue resistance and endurance in older vs. younger subjects. Across studies, fatigue resistance has been shown to be poorer, better, or the same in older subjects; endurance has been shown to be better or the same but not worse in older subjects. More recent literature suggests greater fatigue in older muscle under conditions of high velocity. Differences are probably attributable to variation in subjects, muscle group, and type of contraction. In older adults who show relatively less fatigue, there is less acidosis in muscle and less concentration of other metabolites associated with fatigue.
    Therefore, central activation and peripheral activation are not the source of age-related fatigue differences; also there is no evidence for impairment of electrochemical coupling or calcium kinetics in vivo as an explanation. What is left, then, is energetics. There is a similar Vmax for oxidative phosphorylation in older vs. younger subjects, but a much lower peak glycolytic rate in older subjects. Therefore, it appears that older adults rely on oxidative metabolism more than anaerobic metabolism, which is consistent with pH data.
    Neuromuscular fatigue has been associated with symptomatic fatigue in certain diseases where fatigue is prominent, such as cancer, multiple sclerosis, amyotrophic lateral sclerosis, and end-stage renal disease.

Future questions:

  • If older adults rely more on oxidative metabolism, does reduced delivery of oxygen to muscle (in peripheral vascular disease, for example) impair muscle function more in this population?
  • What is the relationship between muscle function and the symptom of fatigue? Can older pre-frail adults serve as a model to examine the sequence of events leading to fatigue?
  • What is the role of the central nervous system in altered muscle function and symptomatic fatigue?

Dehydration and electrolyte disturbances
    Mild hyponatremia (<135 mmol/L) is the most common electrolyte abnormality in any patient group, whether hospitalized or chronically diseased ambulatory. Incidence increases markedly with age, and is associated with higher mortality. Body fluid homeostasis is significantly affected by three age-associated changes: body composition, kidney function, and brain function. Sarcopenia leads to a 5-10% decrease in total body water and a 15-20% decrease in plasma volume. As a result, older adults exhibit greater fluxes of electrolyte concentrations and osmolality, and are likely to develop greater hyperosmolality with the same level of dehydration. Decreased glomerular filtration rate with aging leads to a decreased ability to conserve water and sodium. Finally, older adults have decreased thirst perception and increased vasopressin secretion. All of these mechanisms can contribute to symptoms of fatigue. Altered blood volume, decreased heat tolerance, and elevated core temperature may be important consequences of body fluid derangements.

Inflammatory mediators
    An evolving definition of inflammation in geriatrics is a low grade activation of the innate immune system that leads to chronic production of inflammatory mediators. Chronic inflammation influences adverse outcomes through symptoms such as fatigue, pathological changes in multiple tissues, worsening of chronic diseases, and activation of new diseases. It is associated with frailty, disability, and early mortality.
    Commonly studied inflammatory mediators include IL-6, C-reactive protein, TNF-alpha, and NF-kappa-B. Inflammatory mediators increase with age, probably as a result of increased chronic diseases, altered body composition, increased free radicals, and decreased anabolic hormones. Multiple studies link increased cytokine levels with frailty, disability, mortality, disease states, and decline in muscle function. Inflammatory mediators affect skeletal muscle function through inhibition of myogenic differentiation and skeletal muscle regeneration (TNF-α), and possibly effects on muscle stimulation capacity.
    IL-6 and CRP are significantly associated with vital exhaustion in older women after recent myocardial infarction [41]. IL-6, CRP, and IL-1 receptor antagonist are associated with fatigue in cancer patients [42]. Multiple clinical and rodent models show a relationship between pro-inflammatory cytokines and sickness behavior, of which fatigue is one component. 
    Potential mechanisms of inflammatory mediators leading to fatigue include indirect mechanisms via disease states and multiple system changes triggered by chronic inflammation, and direct mechanisms via action on the central nervous system, specifically the HPA axis. IL-6 appears to be the main endocrine communicator between the periphery and brain. IL-6 acts on brain vasculature to produce prostaglandin E2, leading to altered HPA axis activity and behavioral responses such as anorexia and hypoactivity. 
    There may be clinical utility in blocking inflammatory mediators to treat fatigue. Aspirin and ibuprofen are already in widespread use for sickness symptoms, such as with upper respiratory infections or influenza. It is unclear, however, whether there is a specific effect on fatigue.

Future questions:

  • Are there “best markers” for fatigue? Can we develop aggregate inflammatory and HPA axis markers that are most predictive of symptoms and adverse outcomes?
  • Can we more appropriately and directly target specific components of inflammatory pathways?
  • Can we identify genotypes/phenotypes “at risk” for fatigue and target individuals for intervention?

Psychoneuroimmunologic mechanisms
    Adverse events, chronic stress, and depression are interpreted or perceived by the central nervous system, leading to changes in both central (e.g., cognition, behavior, and affect) as well as peripheral functions (e.g., physiologic and immunologic responses).  Communication between these central and peripheral components is bidirectional.  Traditionally, it was thought that the hypothalamic-pituitary-adrenal axis was the main mediator of this communication.  Currently, CNS output is known to affect immune function through over 20 different mediators.  A hypothetical model of fatigue in aging proposes that psychosocial factors (e.g., stress and depression) plus individual differences (e.g., personality and neuroticism) can lead to fatigue directly or indirectly through inflammation.  Both fatigue and inflammation interact with, and potentially modify, disease.
    “Sickness behavior” is a syndrome of anhedonia, anorexia, fatigue, sleep changes, fever, and exaggerated responses to pain.  Sickness behavior can be elicited experimentally through administration of bacterial lipopolysaccharide (LPS), which also causes a large-magnitude release of IL-1, TNF-α, and IL-6.  Aged mice have an increased IL-6 response and prolonged sickness behavior after LPS injection compared to younger adult mice [43], and pre-administration with IGF-1 abolishes this sickness behavior response [44].  As such, cumulative environmental and immunological burdens may lead to increased inflammatory mediators and decreased IGF-1, which may lead to increased diseases of aging, fatigue, and mortality.
    In five clinical studies (Moynihan et al., submitted), fatigue was measured using three questions from the CES-D (inertia, psychomotor retardation, tiredness).  In all five populations, depressive symptoms (as measured by the CES-D-R minus the three fatigue items) were correlated with reported fatigue, as has been shown by others [36;45-49].  As well, trait Neuroticism and poor sleep quality were also significantly associated with fatigue; however, after controlling for depression, these relationships were no longer significant.  Given the strong link between Neuroticism and depression [50], this observation may suggest a mediating relationship in which dispositional Neuroticism functions as a “distal” risk factor for fatigue, its effects being mediated more proximally by depression.  Poor sleep quality, high levels of perceived stress, loneliness, and poorer social and emotional functioning are also common correlates of depression, and their apparent association with fatigue appears to be mediated by depressive symptoms also.  This suggests that while multiple factors may mark increased fatigue, depression may function as a “final common pathway” through which dispositional stress-reactivity, situational stress, social isolation, and poor sleep operate to induce fatigue.  Unexpectedly, the oldest age groups reported the lowest fatigue according to questions from the CES-D.  There was no significant correlation between fatigue and plasma IL-6 or IGF-1.

Psychosocial influences
    Fatigue may be best characterized as the absence of energy and the mind-body response to that absence. In order to obtain a comprehensive picture of fatigue, it is essential to explore the interaction of biological (e.g., illness), psychological (e.g., affect), and social factors (positive and negative interactions). Between-persons analyses identify “who” has more fatigue, while within-person analyses over time show “when” people have more fatigue; analyses that combine both approaches show who has the greatest fatigue when experiencing pain, stress, or other events. 
    In a study of women with osteoarthritis, rheumatoid arthritis, or fibromyalgia, subjects kept daily diaries that included fatigue ratings throughout the day. Fibromyalgia patients reported the greatest levels and fluctuations of fatigue throughout the day [22]. Physical functioning was predicted by fatigue and pain combined, but not by pain alone. In addition, fatigue was predicted by the presence of negative affect as well as the absence of positive affect, with the effect size of the latter double that of the former. (Increasing evidence supports the view that positive affect and negative affect are independent constructs rather than two ends of a continuum.)
    Compared to non-depressed patients, increased daily fatigue in depressed patients is associated with greater increases in daily pain, greater decline in daily enjoyment of interpersonal events, and greater daily negative affect. Stressful or positive interpersonal events predict daily fatigue, even after accounting for pain, depression, and sleep. Fatigue in rheumatoid arthritis patients with greater social engagement is unaffected by negative interpersonal events. In fibromyalgia or rheumatoid arthritis, positive interpersonal events led to less fatigue on that day, but paradoxically higher fatigue on the following day. 
    Interpersonal stress can be imposed acutely in a laboratory setting. Discussing a recent negative interpersonal stressor is associated with increases in heart rate and systolic blood pressure. Greater perceived stressfulness is associated with increased fatigue, while ability to preserve positive affect (emotional well-being) was associated with less fatigue.
    Resilience is the natural capacity to recover from stress. In patients with osteoarthritis or fibromyalgia, experimental stress followed by an induced positive mood improved joviality in both patients groups; however, stress followed by neutral mood improved joviality in osteoarthritis, but not in fibromyalgia. Therefore, patients with osteoarthritis may have greater resilience.

Future questions:

  • Is it more useful for studies to focus on the presence of fatigue or the absence of energy?
  • Can exploration of central dopaminergic pathways help to explain the neurobiology of resilience?
  • Can interventions be developed that increase positive affect without costly energy; for example, meditation or relaxation?



Self-report measures
    Given the subjective nature of fatigue, self-report may provide insights into the nature of an individual’s fatigue that are difficult to ascertain through objective measures. Self-report instruments can measure multiple dimensions (temporal characteristics, severity, impact) and/or manifestations (physical, cognitive, emotional, behavioral). Although there are several validated tools for measurement of fatigue, there is no gold-standard.
    Measurement of fatigue has many parallels with pain, and it may be useful to refer to the pain literature to examine the manner in which similar issues have been addressed regarding multidimensionality and measurement. The ideal self-report instrument should have the smallest possible number of items to complete (acceptable to patients), validity and reliability demonstrated within the target population, a scaling method that is easy for patients to use, easily understood items, straightforward scoring and interpretation, easy translation for cross-cultural validity, and clinical utility.
    Examples of unidimensional scales include the Avlund Tiredness Scales, Brief Fatigue Inventory, Fatigue Severity Scale, Functional Assessment of Chronic Illness Therapy – Fatigue  (FACIT-F), Global Vigour and Affect Scale, and Schedule of Fatigue and Anergia (SOFA). Multidimensional scales include Fatigue Assessment Instrument, Fatigue Impact Scale, Fatigue Symptom Inventory, Multidimensional Assessment of Fatigue and the Global Fatigue Index, Multidimensional Fatigue Inventory, Multidimensional Fatigue Symptom Inventory, Piper Fatigue Scale / Revised Piper Fatigue Scale, and visual analogue scales.
    Each type of measure has its own strengths and weaknesses. Unidimensional measures are easy to administer and interpret, but they suffer from questionable reliability, especially for single-item measures, and lack of information about symptom duration, interference, quality, or variability. Multidimensional scales provide more in-depth assessment and potential for greater clinical utility, but they suffer from questionable evidence for their proposed dimensional structures, and many questions within scales yield redundant information. Syndromal measures, such as cancer-related fatigue syndrome, can reliably estimate prevalence of clinically significant symptomatology and have the potential to distinguish cancer-related fatigue from other forms of fatigue, but the diagnostic criteria used are of questionable validity and their utility in clinical decision-making is unclear.
    When assessing self-reported fatigue in older individuals, it is important to consider who comprises the comparison group. Subjects may report different levels of current fatigue depending on whether they are comparing themselves to others of their own age, to themselves at a particular time in the past, or to themselves at their optimal level in the past.
    While self-report of fatigue symptoms can be a powerful tool, it can sometimes also be problematic. There are multiple synonyms for fatigue, but they are not necessarily interchangeable. In a community sample, self-reported “fatigue” was twice as common as “exhaustion”, 10 times more common than “feeling generally run down”, and up to 10 times more common than “weakness” [51]. Prevalence differs depending on how the question is asked: ~40% report “general fatigue” but only ~8% report “tired all the time” [52]. Furthermore, self-report scales include questions that may have face-validity for fatigue, but not necessarily validity. Muscle weakness may sound related to fatigue, but the 2 terms are not always related. In multiple sclerosis, there is increased motor fatigue and muscle weakness, but no significant correlation between motor fatigue and muscle weakness [53]. Scales must be reliable in order to be valid; convergent validity is meaningless if there is a correlation among multiple scales that are not reliable. Furthermore, definitions can be arbitrary, in CFS, fatigue is present for ≥ 6months, while in multiple sclerosis, fatigue of chronic type is ≥ 6 weeks.
    The NIH Patient-Reported Outcomes Measurement Information System (PROMIS) is a consortium of 7 universities operating under a cooperative agreement through the NIH Roadmap. It was designed to develop, validate, and standardize item banks to measure patient-reported outcomes relevant across common medical conditions in 5 domains: pain, fatigue, social functioning, physical functioning, and emotional functioning. PROMIS uses item response theory and computer adaptive testing to assess outcomes efficiently and reliably. Item banks will be developed for public access so that researchers can construct their own individual instruments.

Future questions:

  • In existing datasets that include fatigue-related items not originally intended to measure fatigue, how can these items be used validly and appropriately?
  • Can the multiple questions and dimensions of the various fatigue scales be consolidated or distilled into one or a few representative scales?

Measures of energy utilization
    One approach to measuring energy utilization is to determine the rate of oxygen consumption with activity. Submaximal oxygen consumption exercise tests may be easier and safer for older adults, may better predict functional ability, and can predict health outcomes, especially in congestive heart failure. Measurement of oxygen uptake before, during, and after submaximal treadmill walking yields two parameters: oxygen deficit (delay to reach steady state oxygen uptake at beginning of walk), and oxygen debt (delay to return to baseline oxygen uptake during recovery period).  Elderly with lower peak VO2 have greater oxygen deficit and oxygen debt with submaximal walking [54].
    Another approach to understanding energy utilization is to measure metabolic rate. Resting metabolic rate (RMR) is the minimal amount of energy intake needed to maintain energy balance and stable weight, measured at rest, but awake. It is determined largely by skeletal muscle mass, plus smaller contributions from highly metabolically active organs (heart, kidney, brain, liver). High metabolic rate is characteristic of short-lived animals. In humans, metabolic rate is increased with illness, and higher metabolic rate may be associated with shorter lifespan and greater mortality. Factors associated with higher RMR include younger age, male gender, non-African American ethnicity, higher calorie intake, greater physical activity, and greater physiological stress (eg. trauma, chronic illness). Earlier studies showed an age effect on RMR, but newer studies show minimal age effects.
    Extra-muscle lean body mass is more metabolically active than non-active muscle mass, but because muscle mass is such a large proportion of total body mass, RMR is largely explained by changes in muscle mass; therefore, caution is needed in interpreting data when RMR is normalized to muscle mass alone. In order to estimate RMR more precisely, methods are needed to measure non-skeletal muscle organs that are highly metabolically active. Furthermore, more precise measures of muscle mass are needed, as the assumptions inherent with dual-energy X-ray absorptiometry (DXA) or 24-hour urine creatinine could be inaccurate. Some investigators currently use MRI to evaluate organ mass.

Future questions:

  • How does aerobic function relate to mobility performance and symptoms of exertion and fatigue?
  • What is the relationship between rating of perceived exertion (Borg scale) and self-reported fatigue during activity? There is a high correlation between heart rate and Borg scale rating. Does the same relationship hold for heart rate and self-reported fatigue?
  • How can performance-based measures of fatigue be distinguished from measures of the consequences of fatigue?

Functional correlates
    The relationship between functional ability and self-reported fatigue was studied in the InChianti dataset, a population-based study of older adults in Tuscany, Italy [55]. This analysis involved 1,055 non-cognitively impaired adults, age 65+ years. Fatigue was assessed according to two fatigue-related items of the CES-D. Prevalence of fatigue increased with age among men, but not women. Adjusted for age, the fatigued group among men was older, had greater self-reported inactivity, and reported more diseases than the non-fatigued group; these differences were absent among women. Among both men and women, the fatigued group had greater disability, weaker handgrip strength, poorer functional performance as measured by the Short Physical Performance Battery (SPPB), lower walking speed (400 m), more depressive symptoms, poorer sleep, and poorer self-rated health than the non-fatigued group. After further adjusting for health behaviors, diseases, inflammatory markers, and thyroid function, fatigue remained associated with SPPB, lower walking speed, inability to walk 400 meter, and disability in instrumental activities of daily living.

Physiologic mechanisms of fatigue perception
    In older patients with congestive heart failure, self-reported fatigue is highly correlated with patients’ perceived symptom burden (lower limb swelling) [56], but the correlation may not be as great with the clinician’s assessment of symptom burden. The traditional model of symptom generation in heart failure is that failure of muscle perfusion leads to fatigue and breathlessness. However, the sensing mechanisms that alert to diminished muscle perfusion and/or increased exertion are unclear. 
    Previous studies have used breathlessness as a surrogate for fatigue. Brachial arterial occlusion following exercise caused increased ventilation compared to non-occluded controls, providing the first evidence for the existence of a mechanism for sensing work performed by muscle, dubbed “ergoreceptor.” In heart failure patients, there was much higher ventilation during exercise, and ventilation remained high with arterial occlusion, suggesting hypersensitive ergoreceptors in heart failure. Exercise training decreased ventilation post-occlusion closer to controls, suggesting that exercise can desensitizes ergoreceptors, which may also be an intervention for improving fatigue and quality of life in heart failure [57]; however, these were not consistent findings. 
    Ergoreceptor hyper-responsiveness correlates with reduced muscle mass, poor exercise tolerance, NYHA functional class, lower ejection fraction, and cachexia [58]. In cycle-based exercise, subjects stopped more often because of fatigue, while with treadmill exercise, subjects stopped more often stop because of breathlessness. It is unclear whether weight bearing leads to greater ergoreflex activation or simply causes more muscles to be metabolically active? 
    Ergo receptors may be metaboreceptors (group IV afferents, unmyelinated, vanilloid-1 receptors, capsaicin augmented) and/or mechanoreceptors (group III afferents, thinly myelinated, P2X receptors (purinergic), ATP-augmented). In addition, they may be activated by substance P and tachykinin. Triggers of ergoreceptors might include inactivity, renin-angiotensin-aldosterone system activation, sympathetic nervous system activation, inadequate nutrient blood flow, cytokines, and other factors.

Future questions:

  • Can a typology of fatigue include level of ergoreceptor activity? Fatigue in heart failure may involve ergoreceptor overactivity, while in multiple sclerosis, there may be ergoreceptor underactivity.
  • Given that older adults have impaired thermoregulatory responses, could faster increases in core temperature amplify the ergoreceptor response?

Assessing cognitive fatigue
    Cognitive aging can be characterized in terms of the building blocks of cognition: speed of cognitive processes, capacity (working memory), durability of memory traces, and accrued effects of experience (world knowledge). Speed, capacity and durability of memory decrease steadily with age, but world knowledge increases steadily [59]. Older adults have more trouble ignoring or inhibiting irrelevant information and are more likely to remember events through reliance on feelings of familiarity rather than explicit information. There are age-related declines in brain volume in multiple brain regions.
    Unlike cognitive function, brain neural activity increases with age. In cognitive tasks, older subjects recruit more brain regions than younger subjects for the same task. Furthermore, cognitive decline with aging is magnified in the setting of increased physical fatigue; therefore, physical fatigue may limit cognitive resources.
    Objective measurement of cognitive fatigue can be performed in several ways depending on the question: over an extended time (eg. across a workday), during sustained mental effort, after challenging mental exertion, or after challenging physical exertion. So far, the best evidence for demonstrating fatigue under these conditions is with during sustained mental effort. 
    Neuroimaging of fatigue has revealed no evidence of structural changes (eg. brain volume, lesion load, atrophy), but functional changes include hypoactivation of cortical and subcortical regions including basal ganglia and frontal lobes in multiple sclerosis subjects during a simple motor task, and activation of more brain regions during simple and complex cognitive tasks. In the latter, subjective fatigue after task completion was correlated with performance on more complex tasks. Multiple other studies have also shown increased brain activation in fatigued subjects for the same task. Fatigue does not necessarily result in decreased performance, but it may require more effort.
    Fibromyalgia is a chronic pain syndrome associated with fatigue. Patients with fibromyalgia performed the same as age-matched controls in speed of processing information; however, they performed like older controls on tasks of working memory, long term memory, and verbal fluency. In addition, patients with fibromyalgia have poorer access to vocabulary than older controls or age-matched controls. Fibromyalgia patients recruit more brain regions than age-matched controls on a complex cognitive task involving visuospatial processing, language, and working memory [60]

Future questions:

  • What are the sources of age-related cognitive fatigue?
  •  Is the age-related increase in brain activity compensatory for a declining brain? Is it another symptom of inefficient and declining neural function?
  • Do older people require increased mental effort to overcome age-related fatigue?
  • Do older adults get more tired after the same amount of mental effort than younger adults?
  • Can fatigue arise when individuals cannot inhibit default cognitive patterns?



Multiple studies have found that slowly graded exercise therapy improves fatigue in chronic fatigue syndrome with no significant adverse effects [61]. Potential mechanisms causing unexplained fatigue include psychological causes (fear of exertion, abnormal perception of symptoms) and functional causes (muscular deconditioning, abnormal reaction to exercise). As such, physical exercise may treat both causes. Randomized controlled studies with long-term follow-up are needed to determine whether patients who respond to exercise maintain health. Standardization of outcomes and definitions of fatigue across studies is essential.

Future questions:

  • Can measurement of non-exercise activity help to describe more comprehensively the effect of interventions for fatigue?
  • Can examination of social and psychological effects of exercise or other interventions help to paint a more complete picture of the effect of the intervention?
  • Can extending widely used tests to expose decrements in performance with effort or time bridge the divide between self-report and objective measures? For example, in a 400-meter walk test, would it be informative to compare walking speed at the end of the walk with the beginning, while assessing concomitantly self-reported fatigue at both times?

    Evidence for pharmacologic therapy for fatigue is of poor quality. It is difficult to make a recommendation because 1) fatigue is poorly defined, leading to heterogeneity in subject selection; 2) therapies are empirically selected, leading to poor linkage between pathophysiology, mechanisms, and therapy; 3) studies often have small sample sizes, heterogeneous outcome measures and scales, and high placebo response rates; and 4) little attention is given to functional impact; i.e., physical activity or fatigue-activity relationships. Furthermore, pharmacologic interventions for fatigue have not typically followed the usual steps in the drug discovery process of target identification, preclinical proof of concept, biomarker development, preclinical toxicology, and finally, clinical trials.
    Potential pathophysiologic contributors to fatigue include cardiovascular, ventilatory, musculoskeletal, and central mechanisms. Pharmacologic agents that have been evaluated for fatigue include centrally acting drugs (methylphenidate, caffeine, SSRIs, cholinesterase inhibitors, modafinil), promyogenic agents (testosterone, growth hormone), agents that improve aerobic capacity/endurance (erythropoietin, growth hormone), immune modulators / anti-inflammatory agents (anti-TNFα, hydrocortisone, aspirin, interferon-α, immunoglobulin), and others (DHEA, ACE inhibitors). Of approximately 50 randomized, non-randomized, and open-label trials, there is no consistent evidence for successful treatment effect of any drug. Of note, fluoxetine improves mood in chronic fatigue syndrome, but not fatigue.
    The best evidence for pharmacologic effect on fatigue is with erythropoietin in cancer fatigue, but more recently appreciated complications of treatment (thromboembolic events, red cell aplasia, etc.) and target hemoglobin concentrations need to be considered.
    Testosterone increases lean body mass in older adults, increases grip strength, and increases self-reported physical function. There are inconsistent effects on observed physical function. Preliminary data suggests testosterone increases mitochondrial area. In open-label trials in hypogonadal men, testosterone improved mood and increased energy or pep. One potential explanation for testosterone’s effects on fatigue is increased hemoglobin concentration.
    Pharmacologic interventions for fatigue require well-designed, adequately powered randomized controlled trials with rigorous and uniform entry criteria and outcomes. Initial trials should be conducted in discrete, well-defined patient populations. Selection of therapeutic agents should be guided by disease pathophysiology and mechanism of drug action, and efficacy should be judged by both self-reported measures of fatigue perception and impact, and objective measures of physical activity, physical function, and fatigue-activity relationships.

Cognitive Behavioral Therapy
    Fatigue is probably a CNS phenomenon and may be a final common pathway acting as a fundamental “alarm” that forces an individual to stop activity and understand what is wrong. Fatigue is rarely “explained” by identifiable disease. Only a fraction of the symptoms presented in primary care settings (fatigue, chest pain, dizziness, headache, edema, back pain, dyspnea, insomnia, abdominal pain, and numbness) are found to have a cause in identifiable disease pathology [62]. Even in chronic medical conditions, symptoms are poorly related to severity of the pathology
    The traditional medical model focuses on pathology giving rise to symptoms. Consequently, because of their disease focus, physicians find it more difficult to help patients who have symptoms that are not easily explained by an organic cause [63]. An alternative approach to understand symptoms such as fatigue is to start with the symptom and attempt to identify biological, psychological, behavioral, social cultural, and interpersonal factors that contribute it. The contributing factors can be categorized into predisposing, precipitating, or perpetuating factors. Cognitive behavioral therapy (CBT) attempts to address perpetuating psychological factors, such as worrying about fatigue or seeing fatigue as uncontrollable or a barrier to activity.
There are different kinds of CBT (adaptive vs. rehabilitative), but all forms help patients identify and change unhelpful symptom-perpetuating thoughts and behaviors. Older adults tend to be more compliant with CBT, possibly due to greater conscientiousness and/or fewer distractions.
    Multiple studies have demonstrated efficacy of CBT on fatigue in chronic fatigue syndrome and cancer-related fatigue. Of note, beneficial effects may continue and even increase after treatment stops. There is modest evidence for effect from brief CBT or psycho-educational interventions, though the strongest effect if from more intensive CBT.
    Exercise has been thought to improve fatigue through improvements in physical fitness; however, recent studies suggest that exercise itself may help patients to focus less on their fatigue [64]. As such, it may be misleading to categorize treatments as either exercise or CBT, as both may work in the same way.

 Future questions:

  • Does CBT work for all types of fatigue?
  • What is the active ingredient of CBT? Or does CBT need several ingredients?
  • Does CBT do more than simply make patients more adherent to increased activity?
  • Is exercise really a psychological intervention?
  • Can very brief and simple interventions work?
  • Can we prevent fatigue?

Complementary and Alternative Medicine Approaches
    Complementary and alternative medicine modalities are classified into mind-body medicine, biologically-based therapies, manipulative and body-based systems, and energy therapies. Yoga is a form of mind-body medicine. In a randomized controlled trial of yoga or walking in patients with multiple sclerosis, both yoga and walking improved energy/vitality (SF-36) compared to wait-list controls, but there was no effect on depression. Furthermore, fatigue measures were not markedly associated with baseline functional impairment [65].
    In a randomized controlled trial of yoga or walking in healthy 65-85 year olds who did not need to be sedentary for inclusion, yoga, but not exercise, improved energy/vitality (SF-36), but there was no effect on general fatigue (Multidimensional Fatigue Inventory) [66].
Although perception of fatigue is important, objective measures are needed because many subjects lack insight into their deficits. For example, in multiple sclerosis patients, self-reported cognitive difficulty was correlated with depression, but not with measured cognitive ability [67]. Examples of objective measures include EEG and motor strength with repetition; for example, serial handgrips.

Future questions:

  • What is the underlying mechanism mediating benefit of mind-body interventions for fatigue?
  • Can certain medications that cause fatigue as a side-effect be used as a clinical model to test fatigue interventions? Anti-epileptic drugs may be one such example.



1.   Avlund K, Pedersen AN, Schroll M (2003) Functional decline from age 80 to 85: influence of preceding changes in tiredness in daily activities. Psychosom Med 65:771-777
2.   Dittner AJ, Wessely SC, Brown RG (2004) The assessment of fatigue: a practical guide for clinicians and researchers. J Psychosom Res 56:157-170
3.   White PD (2007) How common is chronic fatigue syndrome; how long is a piece of string? Popul Health Metr 5:6
4.   Clark MR, Katon W, Russo J, Kith P, Sintay M, Buchwald D (1995) Chronic fatigue: risk factors for symptom persistence in a 2 1/2-year follow-up study. Am J Med 98:187-195
5.   Mosso A (1904) Fatigue. Swan Sonnenschein and Co., London,
6.   Chaudhuri A, Behan PO (2000) Fatigue and basal ganglia. J Neurol Sci 179:34-42
7.   Sadler IJ, Jacobsen PB, Booth-Jones M, Belanger H, Weitzner MA, Fields KK (2002) Preliminary evaluation of a clinical syndrome approach to assessing cancer-related fatigue. J Pain Symptom Manage 23:406-416
8.   Prue G, Rankin J, Allen J, Gracey J, Cramp F (2006) Cancer-related fatigue: A critical appraisal. Eur J Cancer 42:846-863
9.   Gabrilove JL, Cleeland CS, Livingston RB, Sarokhan B, Winer E, Einhorn LH (2001) Clinical evaluation of once-weekly dosing of epoetin alfa in chemotherapy patients: improvements in hemoglobin and quality of life are similar to three-times-weekly dosing. J Clin Oncol 19:2875-2882
10.   Courneya KS, Mackey JR, Bell GJ, Jones LW, Field CJ, Fairey AS (2003) Randomized controlled trial of exercise training in postmenopausal breast cancer survivors: cardiopulmonary and quality of life outcomes. J Clin Oncol 21:1660-1668
11.   van Dam FS, Schagen SB, Muller MJ, Boogerd W, vd WE, Droogleever Fortuyn ME, Rodenhuis S (1998) Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer: high-dose versus standard-dose chemotherapy. J Natl Cancer Inst 90:210-218
12.   Bower JE, Ganz PA, Aziz N, Fahey JL (2002) Fatigue and proinflammatory cytokine activity in breast cancer survivors. Psychosom Med 64:604-611
13.   Weitzner MA, Moncello J, Jacobsen PB, Minton S (2002) A pilot trial of paroxetine for the treatment of hot flashes and associated symptoms in women with breast cancer. J Pain Symptom Manage 23:337-345
14.   Lerdal A, Celius EG, Moum T (2003) Fatigue and its association with sociodemographic variables among multiple sclerosis patients. Mult Scler 9:509-514
15.   Bakshi R, Shaikh ZA, Miletich RS, Czarnecki D, Dmochowski J, Henschel K, Janardhan V, Dubey N, Kinkel PR (2000) Fatigue in multiple sclerosis and its relationship to depression and neurologic disability. Mult Scler 6:181-185
16.   Schwartz CE, Coulthard-Morris L, Zeng Q (1996) Psychosocial correlates of fatigue in multiple sclerosis. Arch Phys Med Rehabil 77:165-170
17.   MacHale SM, Lawrie SM, Cavanagh JT, Glabus MF, Murray CL, Goodwin GM, Ebmeier KP (2000) Cerebral perfusion in chronic fatigue syndrome and depression. Br J Psychiatry 176:550-556
18.   Johnson SK (2005) Depression and fatigue. In: John DeLuca (ed) Fatigue as a Window to the Brain. MIT Press, Cambridge, MA, p 157
19.   Strober LB (2005) An examination of four models predicting fatigue in multiple sclerosis. Archives of Clinical Neuropsychology 20:631-646
20.   Ingles JL, Eskes GA, Phillips SJ (1999) Fatigue after stroke. Arch Phys Med Rehabil 80:173-178
21.   Glader EL, Stegmayr B, Asplund K (2002) Poststroke fatigue: a 2-year follow-up study of stroke patients in Sweden. Stroke 33:1327-1333
22.   Zautra AJ, Fasman R, Parish BP, Davis MC (2007) Daily fatigue in women with osteoarthritis, rheumatoid arthritis, and fibromyalgia. Pain 128:128-135
23.   Pincus T, Swearingen C, Wolfe F (1999) Toward a multidimensional Health Assessment Questionnaire (MDHAQ): assessment of advanced activities of daily living and psychological status in the patient-friendly health assessment questionnaire format. Arthritis Rheum 42:2220-2230
24.   Wolfe F, Hawley DJ, Wilson K (1996) The prevalence and meaning of fatigue in rheumatic disease. J Rheumatol 23:1407-1417
25.   Martin JC (1996) Determinants of functional health of low-income black women with osteoarthritis. Am J Prev Med 12:430-436
26.   Stone AA, Shiffman S (2002) Capturing momentary, self-report data: a proposal for reporting guidelines. Ann Behav Med 24:236-243
27.   Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A (1994) The chronic fatigue syndrome: a comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med 121:953-959
28.   Fried LP (2001) Frailty in older adults: evidence for a phenotype. The journals of gerontology Series A, Biological sciences and medical sciences 56:-56
29.   Radloff LS (1977) The CES-D scale: a self-report depression scale for research in the general population. Applied Psychological Mesaurement 1:385-401
30.   Bandeen-Roche K, Xue QL, Ferrucci L, Walston J, Guralnik JM, Chaves P, Zeger SL, Fried LP (2006) Phenotype of frailty: characterization in the women's health and aging studies. J Gerontol A Biol Sci Med Sci 61:262-266
31.   Xue Q-L, Bandeen-Roche K, Zhou J, Varadhan R, Fried LP (2008) Initial manifestations of frailty criteria and the development of frailty phenotype in the Women's Health and Aging Study II. J Gerontol A Biol Sci Med Sci (In press):
32.   Schultz-Larsen K, Avlund K, Kreiner S (1992) Functional ability of community dwelling elderly. Criterion-related validity of a new measure of functional ability. J Clin Epidemiol 45:1315-1326
33.   Schultz-Larsen K (2007) Tiredness in daily activities: a subjective measure for the identification of frailty among non-disabled community-living older adults. Archives of Gerontology and Geriatrics 44:83-93
34.   Avlund K, Kreiner S, Schultz-Larsen K (1996) Functional ability scales for the elderly: a validation study. Eur J Public Health 6:43-48
35.   Avlund K, Rantanen T, Schroll M (2006) Tiredness and subsequent disability in older adults: The role of walking limitations. J Gerontol A Biol Sci Med Sci 61:1201-1205
36.   Avlund K, Rantanen T, Schroll M (2007) Factors underlying tiredness in older adults. Aging Clin Exp Res 19:16-25
37.   Avlund K (2007) Educational intervention toward preventive home visitors reduced functional decline in community-living older women. Journal of clinical epidemiology 60:954-962
38.   Baumgartner RN (1998) Epidemiology of sarcopenia among the elderly in New Mexico. American Journal of Epidemiology 147:755-763
39.   Brochu M, Starling RD, Ades PA, Poehlman ET (1999) Are aerobically fit older individuals more physically active in their free-living time? A doubly labeled water approach. J Clin Endocrinol Metab 84:3872-3876
40.   Menshikova EV (2006) Effects of exercise on mitochondrial content and function in aging human skeletal muscle. The journals of gerontology Series A, Biological sciences and medical sciences 61:534-540
41.   Janszky I, I (2005) Self-rated health and vital exhaustion, but not depression, is related to inflammation in women with coronary heart disease. Brain, behavior, and immunity 19:555-563
42.   Schubert C, Hong S, Natarajan L, Mills PJ, Dimsdale JE (2007) The association between fatigue and inflammatory marker levels in cancer patients: a quantitative review. Brain Behav Immun 21:413-427
43.   Godbout JP, Moreau M, Lestage J, Chen J, Sparkman NL, Connor JO, Castanon N, Kelley KW, Dantzer R, Johnson RW (2007) Aging Exacerbates Depressive-like Behavior in Mice in Response to Activation of the Peripheral Innate Immune System. Neuropsychopharmacology
44.   Dantzer R (1999) Central administration of insulin-like growth factor-1 inhibits lipopolysaccharide-induced sickness behavior in mice. Neuroreport 10:289-292
45.   Ganz PA, Bower JE (2007) Cancer related fatigue: a focus on breast cancer and Hodgkin's disease survivors. Acta Oncol 46:474-479
46.   Samaha E, Lal S, Samaha N, Wyndham J (2007) Psychological, lifestyle and coping contributors to chronic fatigue in shift-worker nurses. J Adv Nurs 59:221-232
47.   Skerrett TN, Moss-Morris R (2006) Fatigue and social impairment in multiple sclerosis: the role of patients' cognitive and behavioral responses to their symptoms. J psychosom Res 61:587-593
48.   Ter Wolbeek M, van Doornen LJ, Kavelaars A, van de Putte EM, Schedlowski M, Heijnen CJ (2007) Longitudinal analysis of pro- and anti-inflammatory cytokine production in severely fatigued adolescents. Brain Behav Immun
49.   Weisbord SD, Fried LF, Arnold RM, Fine MJ, Levenson DJ, Peterson RA, Switzer GE (2005) Prevalence, severity, and importance of physical and emotional symptoms in chronic hemodialysis patients. J Am Soc Nephrol 16:2487-2494
50.   Kendler KKSG (2006) Personality and major depression: a Swedish longitudinal, population-based twin study. Archives of general psychiatry 63:1113-1120
51.   Wessely S, Hotopf M, Sharpe D (1998) Chronic Fatigue and Its Syndromes. Oxford University Press, New York
52.   Lewis G (1992) The epidemiology of fatigue: more questions than answers. Journal of epidemiology and community health 46:92-97
53.   Schwid SR, Tyler CM, Scheid EA, Weinstein A, Goodman AD, McDermott MP (2003) Cognitive fatigue during a test requiring sustained attention: a pilot study. Mult Scler 9:503-508
54.   Alexander NB, Dengel DR, Olson RJ, Krajewski KM (2003) Oxygen-uptake (VO2) kinetics and functional mobility performance in impaired older adults. J Gerontol A Biol Sci Med Sci 58:734-739
55.   Vestergaard S, Nayfield SG, Patel KV, Eldadah B, Cesari M, Ferrucci L, Ceresini G, Guralnik JM (2008) Fatigue in a Representative Population of Older Persons and its Association with Functional Impairment, Functional Limitation and Disability. J Geron Med Sci (in press).
56.   Barnes SSG (2006) Prevalence of symptoms in a community-based sample of heart failure patients. Journal of Pain and Symptom Management 32:208-216
57.   Piepoli M (1996) Contribution of muscle afferents to the hemodynamic, autonomic, and ventilatory responses to exercise in patients with chronic heart failure: effects of physical training. Circulation 93:940-952
58.   Piepoli MF (2006) Reduced peripheral skeletal muscle mass and abnormal reflex physiology in chronic heart failure. Circulation 114:126-134
59.   Park DC (2002) Models of visuospatial and verbal memory across the adult life span. Psychology and aging 17:299-320
60.   Park DC (2001) Cognitive function in fibromyalgia patients. Arthritis and rheumatism 44:2125-2133
61.   Whiting P (2001) Interventions for the treatment and management of chronic fatigue syndrome: a systematic review. JAMA 286:1360-1368
62.   Kroenke K, Mangelsdorff AD (1989) Common symptoms in ambulatory care: incidence, evaluation, therapy, and outcome. Am J Med 86:262-266
63.   Carson AJ (2004) Patients whom neurologists find difficult to help. Journal of neurology, neurosurgery and psychiatry 75:1776-1778
64.   Moss-Morris R (2005) A randomized controlled graded exercise trial for chronic fatigue syndrome: outcomes and mechanisms of change. Journal of health psychology 10:245-259
65.   Oken BS, Kishiyama S, Zajdel D, Bourdette D, Carlsen J, Haas M, Hugos C, Kraemer DF, Lawrence J, Mass M (2004) Randomized controlled trial of yoga and exercise in multiple sclerosis. Neurology 62:2058-2064
66.   Oken BS, Zajdel D, Kishiyama S, Flegal K, Dehen C, Haas M, Kraemer DF, Lawrence J, Leyva J (2006) Randomized, controlled, six-month trial of yoga in healthy seniors: effects on cognition and quality of life. Altern Ther Health Med 12:40-47
67.   Lovera J (2006) Correlations of Perceived Deficits Questionnaire of Multiple Sclerosis Quality of Life Inventory with Beck Depression Inventory and neuropsychological tests. Journal of rehabilitation research and development 43:73-82

    Page last updated Feb 19, 2009