7Indicators of Function in the Geriatric Population

Halter JB, Reuben DB.

Publication Details

Assessment of function of an elderly person can be conducted at a number of different levels and using methods which span a wide range of sophistication. On a clinical level, assessment of overall functional capability is a central theme of the field of geriatric medicine. However, with rapid advances in technology, assessment of function of individual organs or organ systems, of individual cells, and even of specific molecules will become increasingly feasible, even in the context of a household survey or evaluation. This chapter will review aspects of function at each of these levels in humans, the various types of total organism functions that are usually assessed clinically and can predict outcomes, and issues related to choice of measures of function for this population.

Measurement of function at any of these levels may define the presence of a disease state. For example, a diagnosis of congestive heart failure or renal failure is based on a certain level of functional impairment of the heart or kidneys. However, most diseases are characterized either by overt pathology like tissue damage (e.g., myocardial infarction), inflammation (e.g., infection), or invasion (e.g., cancer); the presence of defined clinical characteristics in a patient (e.g., depression or malnutrition); or a laboratory measurement which predicts subsequent pathology (e.g., high blood pressure or high cholesterol level).



Samples of tissue or body fluid provide a wide range of molecules for possible identification and characterization. Assessment of activity of certain key enzymes can provide insight into the metabolic state of a tissue sample. Molecules on the surface of circulating blood lymphocytes can provide insight about activity of the individual's immune system and predict immune system responses to certain challenges including transplantation.

Analysis of the hemoglobin molecule can provide insight about the functional characteristics of this molecule, which is critical for delivering oxygen to tissues. Abnormalities of hemoglobin can impair the appropriate function of the hemoglobin molecule and result in a serious illness such as sickle cell anemia. Exposure of the hemoglobin molecule over time to glucose, the main sugar in the blood, leads to molecular attachment of glucose, creating glycosylated hemoglobin. Thus the amount of glycosylated hemoglobin provides a quantitative index of overall body exposure to glucose over a period of time. Glycosylated hemoglobin is now a readily available clinical test for degree of elevation of blood sugar in people with diabetes (Halter, 1999). Glycosylated hemoglobin can be quantitated from a drop of blood with a rapid test that can be administered in a doctor's office or done in the home setting. It turns out that this structural change of the hemoglobin molecule also affects an important functional characteristic, its oxygen carrying capacity. The amount of glucose attached to other important body proteins can be assessed in tissue samples, as from a skin biopsy. Again, not only does such a measurement provide an estimate of exposure to increased blood sugar, but functional characteristics of the glucose-modified molecules are affected and may contribute directly to diabetes-related complications.

Advances in technology will likely make it possible to carry out molecular screening of a large number of molecules in body fluids or tissue samples that may identify genetic variation or be markers of disease processes (Burns et al., 1998). Similarly, future technology will allow rapid quantitation in such samples of the amount of messenger RNA, the template for synthesis of cell proteins. The amount and structure of specific messenger RNAs provide measures of functional capability of individual genes. Furthermore, the pattern of messenger RNA in a given cell may provide a biological fingerprint of the cell's functional response to a given set of biological circumstances.


Specific cells can be isolated from blood or tissue samples for testing of functional capability. In particular, white blood cells, some of which are responsible for initiating inflammation (polymorphonuclear leukocytes) and others for immunologic responses (lymphocytes), can be isolated from blood and tested in vitro with a variety of molecules called cytokines which can stimulate specific responses. Such measures can provide insight about an individual's risk for infection or impaired inflammatory response to injury. More generally, alterations in intercellular signaling systems may play an important role in aging processes (Roth and Yen, 1999). Function of the blood coagulation system can be assessed by measures of clotting proteins in blood and functional characteristics of platelets, which are important constituents of the blood clotting response. Tissue samples can provide access to a wider variety of cells, such as those of the skin, muscle, and fat, all of which could provide insight about functional characteristics of these cell types. A more readily available cell type is epithelial cells. However, there is currently no direct link between the functional status of these cells and disease states.


The most traditional clinical laboratory measurements of body fluids provide direct or indirect assessment of one or more organ systems. For example, blood levels of thyroid hormones provide measures of over- or under-function of the thyroid gland. Similarly, blood samples can be analyzed to provide estimates of kidney, liver, and bone marrow function. Function of other endocrine glands such as the adrenals, gonads, pituitary, and parathyroids can also be assessed. Overall function of metabolic systems controlling glucose and lipid metabolism are also available from analysis of blood samples.

Another dimension for assessment of organ system function is change over time. Current technology allows noninvasive continuous monitoring of heart rate and blood pressure for up to several days in the home setting for clinical purposes to document cardiac arrhythmias and diurnal fluctuations of blood pressure. Such testing provides substantially more information about rapidly changing variables than a single measure. Small motion detectors can be worn to quantitate the degree of physical activity over time. Sleep monitoring equipment can be used in the home setting to document nocturnal activity and sleep patterns. Noninvasive continuous monitoring of the blood glucose level will be available in the near future.

Simple mechanical devices are available to estimate pulmonary function, which can be carried out in a household survey situation. Simple measures of cardiovascular system function such as blood pressure and heart rate can also be measured readily in a home situation. However, these measures provide only indirect assessment of function of the heart or other aspects of the cardiovascular system. Advances in technology may greatly enhance the capability of quantitating organ system function. Already, portable ultrasound equipment can be used to obtain quantitative measures of cardiac function and evaluate whether there is evidence of significant atherosclerosis in major blood vessels such as the carotid artery or leg arteries. Portable equipment can also be used to quantitate body composition and to carry out sophisticated pulmonary function testing. The National Institute on Aging's Baltimore Longitudinal Study of Aging is about to send into the field a mobile research laboratory that has such capabilities. With such experience and some further refinement of technology, it is not hard to imagine a suitably trained technician in the home setting using portable equipment to carry out such measures and transmitting data via internet connection to a central site for quantitative analysis. There is little doubt that the capability for making quantitative, noninvasive measurements will increase in the future, thereby refining our ability to quantitate function of organ systems only estimated crudely with current measurements of blood products.

Total Organism

The contributions of the many organ systems are integrated to permit functioning of the total organism. Such functions may include the completion of specific tasks (e.g., movement of a limb) or be integrated to achieve a goal (e.g., preparing a meal). Integrated functioning may be directed at accomplishing personal survival tasks or towards fulfilling societal roles. Function at the level of the total organism is described in greater detail below.


Function in society is the highest level of integration of function and represents the individual's role with respect to other humans in achieving communal goals. Societal functioning includes work, family, community service, social, and recreational roles. These roles typically change with aging. As people retire they relinquish parenting and assume grand-parenting responsibilities, and their physical (and sometimes cognitive) capacity declines as the result of age-associated physiological or disease processes. Superimposed upon these changes are stereotypic expectations of aging by society and older persons themselves that influence societal roles among capable older persons. Finally, there may be a wide range of degree of personal choice in fulfilling societal roles. Some older persons have accumulated wealth in the form of savings and pensions to allow them to choose to cease certain societal roles (e.g., paid employment), whereas others have much more limited financial and social supports and so may have limited choices of function in society.



Physical functional status is an essential component of the health of older persons, and its measurement has been widely incorporated into research and clinical settings. In the strictest sense, physical function refers to voluntary motor function; however, adequate cognitive function is also required for higher-level activities. For example, severely demented persons may have the capacity for physical function but may be too cognitively impaired to successfully perform even the most basic physical task. More commonly, impaired cognition limits new learning which may be required to overcome a disability. For example, cognitive function may be a limiting factor in determining whether a stroke patient regains the ability to walk or transfer independently. A diagnosis of dementia is an important predictor of subsequent decline of functional capability of older adults (Agüero-Torres et al., 1998).

When cognition is adequate, limitations of physical function are the result of impairments of strength, coordination, flexibility, balance, and endurance, or the result of being unable to integrate these into purposeful activities. Measurement of physical functional status has been used for several purposes, including: to gauge the functional status of individuals and populations, to provide prognostic information, to determine the need for assistive services, to assess the effectiveness of specific interventions in individual persons, and to monitor the course of illness.

Physical function can be conceptualized as a series of increasingly integrated steps beginning with basic components and progressing through three levels of increasingly more integrated function (Figure 7-1). This framework can be overlaid on existing models of pathways from disease or injury to disability, and provides an expanded approach to measurement of the functional component of these models. Moreover it fits well with existing self-report and performance-based instruments (see below). The basic components (coordination/fine motor, balance, strength, flexibility, and endurance) are not functional tasks per se, but are the necessary elements that permit performance of more integrated functional tasks. At each subsequent level of integration, more than one basic component is usually necessary, and additional components unrelated to physical capability (e.g., cognitive, sensory, and affective) influence performance. Motivation, perceived self-efficacy, and physical environment also modify how the basic components are integrated into levels of higher function.

FIGURE 7-1. A framework for the hierarchy of physical functional status.


A framework for the hierarchy of physical functional status.

Examples of the basic components and integration levels are provided in Table 7-1. The first level of integration is performance of specific movements. These tasks are the most basic level at which impairments of range, strength, endurance, balance, and coordination begin to have an impact on function. At the second level, these physical movements are conjoined to complete specific goal-oriented functional tasks (e.g., instrumental activities of daily living). At the third (highest) level, goal-oriented functional tasks are integrated into behaviors aimed at fulfilling societal roles and recreational activities, the so-called advanced activities of daily living (Reuben and Solomon, 1989). At this level, function is often determined by personal choice rather than by physical capacity. For example, a person may have the physical capacity to play golf and tennis but may choose only one because of personal preference. With progressively higher level integration, the many dimensions that contribute to functional status become inextricably entwined. Consider a concert violinist whose ability to perform a concert recital depends upon each of the basic and integrated components of physical functioning, as well as vision, hearing, memory, and affect. Attempting to isolate the physical component from the remainder is a meaningless exercise.

TABLE 7-1. Examples of Physical Activities and Their Level of Integration.


Examples of Physical Activities and Their Level of Integration.


There are several major dimensions of neuropsychological function, including cognition, affect (mood), anxiety, spirituality, and personality, including self-efficacy (empowerment). Some of these have been long established as aspects of health (e.g., cognition and affect), whereas others (e.g., spirituality and self-efficacy) are becoming increasingly recognized as contributors to health.

Within cognitive function, there are also several subdimensions including attention, memory, language, visual spatial performance, and executive function. Attention is a measure of alertness and ability to interact with one's environment. It is primarily impaired in settings of acute illness (e.g., delirium) or affective disorders (e.g., depression). Nevertheless, attention is a key element of neuropsychological function, as some preservation of attention is necessary for all further neuropsychological testing (and objective testing of other dimensions) to be conducted. Inattention can dramatically affect performance and may preclude testing entirely. Although formal tests of inattention have been developed, they are not commonly performed. Rather, implicit decisions or assumptions about the level of attentiveness are made. In community-based settings, major attention deficits are uncommon and such assumptions may be appropriate.

Memory has been divided into immediate (recall), short-term (recent), and long-term (remote) (Cummings and Benson, 1992). Immediate refers to the ability to retain small amounts of material with high accuracy for very short periods of time (e.g., digit span). Short-term memory requires learning of new material (e.g., delayed recall or orientation). Long-term memory is usually tested by evaluating common historical items (e.g., recalling the last several presidents). Language refers to the content and vehicle for communication in contrast to speech, which represents the mechanical aspects of verbal communication. Commonly evaluated language functions include spontaneous verbal output, comprehension of spoken language, repetition of spoken language, naming, reading, and writing (Cummings and Benson, 1992). Although some unusual neurological diseases can cause isolated language impairment, usually these deficits are in the context of cerebrovascular disease (e.g., stroke) or dementia. Visual spatial impairments may be the result of primary visual impairment (see below) or the result of cognitive processing of vision, which in older persons is usually due to cerebrovascular disease or dementia. Commonly used tests of visual spatial function are copying objects and clock drawing. Executive function represents integrative cognitive function that requires sequencing and manipulation of information to perform tasks such as problem solving, abstract reasoning, and planning ahead. Some common tests of executive function include calculations, interpreting proverbs, and identifying similarities and differences.

A variety of objective short screens that incorporate each of these dimensions are available, many of which can be performed by nonprofessional evaluators. However, patients and their families sometimes resent these questions and tasks, even when assessed by physicians. The most commonly used screen is the Mini-Mental State Examination, a 30-item interview-administered assessment of several dimensions of cognitive function (Tombough and McIntyre, 1992). Shorter screens also have been validated (e.g., recall of 3 items at one minute, the clock drawing test, the serial sevens test, and the Time and Change test (Siu, 1991; Froehlich et al., 1998)). People with normal results on these tests are very unlikely to have clinical dementia, while abnormal results increase the odds that dementia is present. It must be recognized, however, that most of these tests rely on tasks that are not routine aspects of everyday life. Moreover, often they do not account for educational level, languages other than English, and cultural differences.

Affective status is a measure of mood and depressive symptoms. The assessment of affective status among older persons is complicated by the common presence of somatic symptoms (e.g., fatigue, loss of appetite) accompanying medical illness. These symptoms can mimic symptoms of depression and are therefore less specific for affective disorders in older persons. Affective symptoms can be assessed using a single question, “Do you often feel sad or depressed?” (Lachs et al., 1990), though this one-item evaluation is not as accurate as longer instruments. A number of other brief self-administered screens are available, including the Mental Health Index from the Medical Outcomes Study SF-36 (Ware and Sherbourne, 1992) and the Geriatric Depression Scale, which has 5-, 15- and 30-item versions (Sheikh and Yesavage, 1986; Hoyl et al., 1999).


Although aging and diseases associated with aging may affect all senses, the most commonly assessed senses are vision and hearing. Impairments of these two sensory systems are among the most prevalent disorders affecting older persons (AARP/AOA, 1998). Visual function encompasses several components including acuity (near and far), visual fields, contrast sensitivity, depth perception, and resistance to glare. Dimensions of hearing impairment include acuity, speech discrimination, and central auditory processing. In addition, the impact of these sensory impairments on integrated function can be measured. The standard method of screening for problems with visual acuity is the Snellen Eye Chart, which requires the patient to stand 20 feet from the chart and read letters, using corrective lenses. Several interviewer and self-administered instruments to detect functional problems due to visual impairment have been developed, including the “Activities of Daily Vision Scale” (Mangione et al., 1992), the VF-14 (Steinberg et al., 1994), and the National Eye Institute Visual Functioning Questionnaire (Mangione et al., 1998).

Pure-tone audiometry is the standard basic evaluation of hearing but alternative methods that are more feasible in survey settings are available. The Welch Allyn AudioscopeTM (Welch Allyn, Inc., Skaneateles Falls, NY) is a hand-held otoscope with a built-in audiometer. When administered under comparable conditions, Audioscope findings are highly correlated with those of pure-tone audiometry. Alternatives are the whispered voice test (Mulrow and Lichtenstein, 1991) and a 6-item screen for hearing impairment based on questions from the National Health and Nutrition Examination Survey (Reuben et al., 1998). Similar to vision assessment, a self-administered test of emotional and social problems associated with impaired hearing, the Hearing Handicap Inventory for the Elderly-Screening Version (HHIE-S), has been developed (Ventry and Weinstein, 1982).


A number of conceptual models describe functional status and the pathways from disease or injury to impairment to disability (Nagi, 1976; Verbrugge and Jette, 1994; WHO 1980; Fried and Guralnik, 1997). Functional status is usually assessed at three levels: basic activities of daily living (BADLs), instrumental or intermediate activities of daily living (IADLs), and advanced activities of daily living (AADLs) (Reuben and Solomon, 1989). BADLs assess the ability of the patient to complete basic self-care tasks (e.g., bathing, dressing, toileting, continence, feeding, and transferring). The loss of BADL function implies the need for direct personal assistance either by family or paid help in the home or at an institution. IADLs measure the patient's ability to maintain an independent household (e.g., shopping for groceries, driving or using public transportation, using the telephone, meal preparation, housework, handyman work, laundry, taking medications, and handling finances). Many of the community-based services (e.g., homemaker, meals-on-wheels) are aimed at providing IADL services and thereby permitting disabled older persons to remain in their homes. AADLs measure the patient's ability to fulfill societal, community, and family roles as well as participate in recreational or occupational tasks. These advanced activities vary considerably from individual to individual but may be exceptionally valuable in monitoring functional status prior to the development of disability. A commonly employed measure of more difficult functional tasks is the Rosow-Breslau scale, which measures the ability to perform heavy housework; walk half a mile; work full-time; climb stairs; and participate in activities such as going to a movie, to church or a meeting, or to visit friends (Rosow and Breslau, 1966).


Qualitative vs. Quantitative

At the most fundamental level, a qualitative measure may be sufficient. For example, if a specific gene product is completely absent it is clear that the gene in question is not present or functioning. Such qualitative assessment of genetic risk usually identifies only rare individuals with specific single gene disorders. However, in the future it is likely that qualitative evaluation of specific patterns of gene expression may be utilized to assess risk for illness or injury. At the level of the whole organism, qualitative measures are somewhat more commonly in use. Handedness or ethnic background is usually presented in qualitative terms, as is assessment of key functional characteristics such as individual activities of daily living.

However, most measurements in current use are quantitative. Thus, while clinical diagnoses are qualitative in nature (i.e., the subject either does or does not have hypertension, diabetes, congestive heart failure, coronary artery disease), the measures used to establish such diagnoses are almost always quantitative. We measure the blood pressure, blood sugar, or blood lipid levels quantitatively and use established criteria to assign a diagnostic category. Quantitative measures of overall functional capability are increasingly in use as well. For example, rather than simply determining whether or not an individual can get out of a chair without assistance, the speed and stability with which an individual gets out of a chair can be measured quantitatively (Alexander et al., 1996).

For quantitative measures which may require substantial time, effort, and cost, investigators must consider the marginal value in relation to simpler qualitative measures. The tradeoff in a household survey is to determine the relative cost versus accuracy, for example, of simply asking whether someone has diabetes versus quantitative measurement of the blood sugar; simply asking whether someone can get out of a chair without assistance versus carrying out an objective, quantitative measure of chair rise capability; or simply asking whether an individual has a memory impairment versus quantitative measurement of multiple domains of cognitive function. Particularly for asymptomatic disorders such as diabetes, hyperlipidemia, and hypertension, simply asking an individual for knowledge of presence of these conditions will lead to substantial underestimation of their true prevalence. For example, epidemiologic studies have demonstrated that 30-50 percent of people who meet criteria for diabetes mellitus are not aware that they have this condition (Harris et al, 1998). Quantitative testing of blood sugar is required to narrow this gap.

Resting vs. Challenge

A key issue in the assessment of functional capability is the relationship between function at rest versus function during a challenge. For purposes of standardization across a population, the resting state has many advantages. However, key characteristics of disease or aging processes are often not present in the resting state and are elicited only during a challenge. The term functional reserve is sometimes used to describe the difference between a functional impairment of a tissue or organ and the degree of impairment needed to cross the threshold at which clinical detection will occur. For example, only a small proportion of people with congestive heart failure or significant coronary artery disease have symptoms at rest. Clinical features become apparent with exercise or cardiac stress. We also know that there can be substantial organ system damage without perturbation of the resting state. For example, humans can lose 50 percent of kidney function without clinical findings or even abnormalities on common quantitative laboratory measures. As technology improves, it is likely that we will be able to detect organ damage using noninvasive functional imaging with increasing success prior to evidence of clinical disease. Such advances will challenge some of our concepts of disease, but may reduce the need for physiologic stress testing to elicit clinical signs.

One characteristic of aging is a loss of homeostasis, leading to a decline in adaptive capability. From this perspective aging may have little impact on the resting state, but age-related declines in functional capability become apparent during a challenge to homeostatic control mechanisms, thereby unmasking a loss of functional reserve. One example of this issue of resting versus challenge is in the area of glucose regulation. Abnormal glucose metabolism meeting criteria for diabetes mellitus can be detected in many individuals by simple measurement of the blood glucose level after an overnight fast. However, in some individuals, the fasting glucose level may not exceed diagnostic limits, but glucose values after ingesting a challenge dose of glucose (called a glucose tolerance test) may demonstrate a marked abnormality. Though not obvious clinically, such abnormality indicates a substantial loss of functional reserve of the endocrine cells of the pancreas, which produce the metabolic hormone insulin (Halter, 1999). The finding of post-challenge hyperglycemia, which is particularly common in older adults (Barrett-Conner and Ferrara, 1998), is associated with the same risk for diabetes-related vascular disease as in individuals who have fasting hyperglycemia (Barzilay et al., 1999). Thus a research project interested in risk factors and outcomes related to cardiovascular disease will underestimate the impact of diabetes if subjects do not receive an oral glucose tolerance test.

Type of Measure

Tissue or Body Fluid Sample

The type, site, and timing of a sample that is obtained from an individual can have a profound effect on interpretation of the measurement. We are just beginning to understand the complexity of some of the issues. Blood is readily accessible and the standard diagnostic material for noninvasive evaluation for many disease states or risk factors. A summary of some of the measures commonly made from blood samples is provided in Table 7-2. However, a blood sample provides only certain information about function of body systems, and results must be interpreted with caution. Just as there are sampling issues in survey research regarding how well a given study population represents the population at large, so are there also issues regarding blood sampling. The content of blood in a vein coming from heart tissue may differ from that of a vein in the arm or a vein draining a kidney. There are also sampling issues regarding the cells in a given blood sample. The population of white blood cells in the circulation may not accurately reflect the much larger population at extravascular sites in terms of subtype distribution or functional characteristics.

TABLE 7-2. Some Indicators of Function in Blood.


Some Indicators of Function in Blood.

The concentration of a substance in blood reflects the kinetics of the turnover of that substance, that is, the balance between the rate of entry of the substance into the blood and the rate of its removal. The kinetics of such turnover may be very complex, reflecting multiple sources for input (e.g., for cholesterol: GI absorption and liver production) as well as multiple tissue sites for removal (e.g., for cholesterol: skeletal muscle, smooth muscle, vascular endothelium), each with different kinetic characteristics. Furthermore, we now recognize that the concentration of many molecules in the blood is not constant, but changes over time, sometimes frequently and dramatically. Such fluctuations occur in response to exogenous physiological stimuli (e.g., environmental change, movement, food ingestion). They also occur as part of internal physiologic control systems. For example, blood levels of most hormones (i.e., molecules in the blood that carry signals from one part of the body to another, such as insulin or cortisol) fluctuate substantially over 24 hours with cycles that are circadian as well as superimposed bursts of secretion (Van Coevorden et al., 1991). Clearly, measurement of a single blood sample cannot account for this variation and may be difficult to interpret in isolation.

Analysis of urine can provide useful information regarding kidney function, salt and water metabolism, presence of infection, or body production rates of some hormones or metabolites that are excreted via the kidney. However the kinetics of kidney excretion and reabsorption are also complex. For example, glucose is normally fully reabsorbed from the kidney and does not appear in the urine. But if the glucose level is sufficiently elevated and/or there is some degree of renal tubular damage, glucose can begin to appear in the urine and provide a marker for abnormal glucose metabolism or kidney damage. However, the relationship between hyperglycemia and appearance of glucose in the urine can vary substantially within a given individual and between individuals, so glucosuria has not been a very useful marker for diagnosis or clinical management of diabetes.

Since all cells from an individual carry the same genome, any cellular source could provide suitable material for genetic analysis. Thus analysis of white cells from blood or epithelial cells scraped from the mucosa of the mouth should provide equivalent information, depending on the amount and quality of genetic material needed for a given analysis. Some analysis can be done on a sample as small as a drop of blood (obtained by pinprick), but a larger blood sample would provide many more cells to work with. White cells in blood can be kept viable for a period of time after collection and can be kept alive in culture in the laboratory, providing an opportunity to study dynamics of gene expression as well as simply documenting gene content.

While less applicable for household surveys, minor office-type procedures such as skin biopsy or needle biopsy can provide specific tissues for analysis. Such samples can complement information available from blood and address site-specific research questions. Study of fat cells can provide added information about metabolic disorders, and study of muscle cells insight about problems with mobility and neuromuscular control. There are sampling issues, since different muscles have different types of fiber distribution and contractile properties, and the metabolic characteristics of fat cells vary somewhat by location.

Overview of Testing at the Levels of Total Organism and Society

There are two general methods of measuring function at the levels of total organism and society, subjective and objective. Each of these methods can be further subdivided. Subjective measures rely on the perception of the subject (or a proxy) and may be obtained by questionnaires completed by the subject or by interview. In contrast, objective (also called performance-based) measures rely on observation by a trained observer. This evaluation may be qualitative or include instrumentation (e.g., audiometry testing). When functional status information is provided by a proxy, the distinction between subjective and objective may be less clear depending upon the nature of the measurement and the rigor of the observer.

Although many aspects of functional status can be assessed by either subjective or objective methods, some of the domains described above (e.g., neuropsychological function) must be assessed by objective testing. Subjective and objective measures may be combined to provide additional information regarding function. For example, a subject may be observed performing a task and then be asked to rate how difficult or tiring the task was.

In addition, the measurement of function in older persons in surveys must consider unique aspects of this population. Many diseases, including those that impair cognition and compromise participation and validity of responses, are more common with advancing age. Respondent burden (e.g., the length of time that a subject can be evaluated before becoming fatigued) may also be an important factor. Accordingly, the selection of measurement instruments must balance scientific accuracy with efficiency and practicality.

Subjective (Self-Report) Testing

Subjective assessment of the ability to function is the most commonly used method of evaluating physical and integrated function. This method may also be used to measure sensory function. It has the advantages of being inexpensive and portable. Moreover, subjective assessment generally focuses on items that have direct clinical relevance. When collecting population data on self-reported physical and integrated functional status, the results are quite sensitive to the wording of both the stems and response items. Among the sources of variation are differences related to duration of disability, whether human assistance is needed, whether assistive devices are used, level of difficulty, and degree of limitation. When comparing essentially equivalent ADL measures across surveys, such minor inconsistencies in wording can result in large differences. For example, the National Medical Expenditure Survey estimated that there are 60 percent more elderly people with ADL problems than did the Supplement on Aging (Wiener et al., 1990). Some items may not be relevant because of traditional gender-specific roles (e.g., some men have never cooked). When administered in written form, language and educational barriers may also preclude accurate assessment.

Finally, when assessing integrated function, subjective measures may be inaccurate because subjects may overestimate or underestimate their capabilities. Such discrepancies between capacity and actual performance are a source of concern when measuring self-reported function (Glass, 1998). Using the hypothetical tense “could do” and the enacted tense “does do,” subjects can be categorized into four groups: (1) “cannot do, doesn't do” (low functioning), (2) “cannot do, does do” (overachievers), (3) “can do, doesn't do” (underachievers), and (4) “can do, does do” (high functioning). Recently, an intermediate step between independence and dependence, “independence with difficulty,” has been identified that may help reconcile some of the discrepancies between capacity and performance (Gill et al., 1998).

Objective Testing

Over the past decade, there has been increasing interest in using performance-based measures to assess physical function. In this context, a performance-based measure can be defined as a test in which a subject (or a patient) performs a movement, behavior, or task according to a standardized protocol, which is scored by an observer. Performance-based tests rate the ability to do a behavior or task. These measures differ from tests that measure physiologic capacity or impairment within a specific dimension, such as pulmonary function or treadmill stress tests. Observations may be qualitative (e.g., rating the steadiness of balance) or quantitative (e.g., timed performance). The approach to objective measures ranges from sophisticated technology that can precisely characterize movements (e.g., joint movements) to coarse but reliable judgments of the quality of tasks (e.g., steadiness when turning 360 degrees). Much of the appeal and promise of performance instruments is the ability to provide standardized and objectively scored data that are suitable for robust statistical analysis. Performance-based data may be predictive of subsequent health events and may be useful in monitoring disease progression and response to therapy (Reuben et al., 1992; Guralnik et al., 1994). Although objective measurements have theoretical advantages (Guralnik et al., 1989), they are not necessarily superior to subjective measures (Myers et al., 1993). Rather they may be measuring different, though related, constructs (Reuben et al., 1995) and may provide complementary information.

Selecting a Measurement Method

For physical functioning, the most appropriate method of measurement depends, in part, upon the level of function that is being assessed. The basic components are best measured directly by performance testing. The remaining levels of integration may be measured by multiple methods: performance-based testing, observer rating, self or proxy report. Currently available performance-based instruments measure physical function at the basic component, specific physical movement, and task- or goal-oriented levels of integration. Many instruments measure physical functioning at more than one level. No performance-based instrument measures societal role function (except, perhaps, if competitive athletics is considered a societal role function). This level of function, highly idiosyncratic to the individual, is probably best assessed by self- or proxy-report methods.

Role of the Individual

Regardless of the functional task assessed or the method employed, the subject's motivation and immediate state (e.g., fatigue, acute illness) must be considered. The abovementioned discrepancies between self-reported “can do” and “does do” may relate to such factors or be due to misperceptions of capacity. Similarly, discrepancies between performance-based measured function and actual performance in everyday life may reflect such factors. For example, motivation to climb a flight of stairs may be higher if meals are served on the second floor. Conversely, subjects may want to impress the examiner and perform better than in the home environment.

Testing Conditions

When considering performance-based measures, several test administration factors must be considered. Among these are standardization of both the test administrator/rater (through training) and the task being performed. The same amount of encouragement and scoring rules must be enforced and subjects need to be told whether the effort is to reflect “usual” or “maximal” effort. Another important issue is that of accommodating to the test and fatiguing. The number of attempts at performing a task and how data from multiple attempts are interpreted need to be determined to account for both the “learning effect” and impaired performance due to fatigue. Finally, the testing site and equipment used can make large differences in performance-based measures. For example, chairs vary in height and flights of stairs vary in number of steps.

Conditions within a subject's home are likely to vary considerably from those in clinics or laboratories. Yet, documenting an impairment of function in the laboratory setting may have little real meaning if the subject can perform the same function at home. Patients with cognitive problems or visual loss may function surprisingly well in a familiar environment but be extremely limited in an unfamiliar clinic or laboratory setting.


  • Agüero-Torres H, Fratiglioni L, Guo Z, Viitanen M, von Strauss E, Winblad B. Dementia is the major cause of functional dependence in the elderly: 3-year follow-up data from a population-based study. American Journal of Public Health. 1998;88:1452–1456. [PMC free article: PMC1508485] [PubMed: 9772843]

  • Alexander NB, Koester DJ, Grunawalt JA. Chair design affects how older adults rise from a chair. Journal of the American Geriatrics Society. 1996;44:356–362. [PubMed: 8636577]

  • American Association of Retired Persons/Administration on Aging. A Profile of Older Americans: 1998. Washington DC: U.S. Department of Health and Human Services; 1998. PF3049 (1298) D996.

  • Barrett-Connor E, Ferrara A. Isolated postchallenge hyperglycemia and the risk of fatal cardiovascular disease in older women and men. Diabetes Care. 1998;21:1236–1239. [PubMed: 9702426]

  • Barzilay J, Speikerman CF, Wahl PW, Kuller LH, Cushman M, Furberg CD, Dobs A, Polak JF, Savage PJ. Cardiovascular disease in older adults with glucose disorders: Comparison of American Diabetes Association criteria for diabetes mellitus with WHO criteria. Lancet. 1999;354:622–625. [PubMed: 10466662]

  • Burns MA, Johnson BN, Brahmasandra SN, Handique K, Webster JR, Krishnan M, Sammarco TS, Man PM, Jones D, Heldsinger D, Mastangelo CH, Burke DT. An integrated nanoliter DNA analysis device. Science. 1998;282:484–487. [PubMed: 9774277]

  • Cummings JL, Benson DF. Dementia: A Clinical Approach. 2nd Edition. Boston: Butterworth-Heinemann; 1992.

  • Fried LP, Guralnik JM. Disability in older adults: Evidence regarding significance, etiology, and risk. Journal of the American Geriatrics Society. 1997;45:92–100. [PubMed: 8994496]

  • Froehlich TE, Robison JT, Inouye SK. Screening for dementia in the outpatient setting: The Time and Change test. Journal of the American Geriatrics Society. 1998;46:1506–1511. [PubMed: 9848810]

  • Gill TM, Robison JT, Tinetti ME. Difficulty and dependence: Two components of the disability continuum among community-living older persons. Annals of Internal Medicine. 1998;128:96–101. [PubMed: 9441588]

  • Glass TA. Conjugating the “tenses” of function: Discordance among hypothetical, experimental, and enacted function in older adults. The Gerontologist. 1998;38:101–112. [PubMed: 9499658]

  • Guralnik JM, Branch LG, Cummings SR, Curb JD. Physical performance measures in aging research. The Journals of Gerontology: Medical Sciences. 1989;44:M141–146. [PubMed: 2768767]

  • Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: Association with self-reported disability and prediction of mortality and nursing home admission. The Journals of Gerontology: Medical Sciences. 1994;49:M85–M94. [PubMed: 8126356]

  • Halter JB. Diabetes mellitus. In: Hazzard WR, Blass JP, Ettinger WH, Halter JB, Ouslander JG, editors. Principles of Geriatric Medicine and Gerontology. 4th edition. New York: McGraw-Hill; 1999. pp. 991–1011.

  • Harris MI, Flegal KM, Cowie CC, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. The Third National Health and Nutrition Examination Survey, 1988-1994. Diabetes Care. 1998;21:518–524. [PubMed: 9571335]

  • Hoyl MT, Alessi CA, Harker JO, et al. Development and testing of a five-item version of the Geriatric Depression Scale. Journal of the American Geriatrics Society. 1999;47:873–878. [PubMed: 10404935]

  • Lachs MS, Feinstein AR, Cooney LM, et al. A simple procedure for general screening for functional disability in elderly patients. Annals of Internal Medicine. 1990;112:699–706. [PubMed: 2334082]

  • Mangione CM, Berry S, Spritzer K, Janz KN, Klein R, Owsley C, Lee PP. Identifying the content area for the 51-item National Eye Institute Visual Function Questionnaire: Results from focus groups with visually impaired persons. Archives of Ophthalmology. 1998;116:227–233. [PubMed: 9488276]

  • Mangione CM, Phillips RS, Seddon JM, et al. Development of the “Activities of Daily Vision Scale:” A measure of visual functional status. Medical Care. 1992;30:1111–1126. [PubMed: 1453816]

  • Mulrow CD, Lichtenstein MJ. Screening for hearing impairment in the elderly: Rationale and strategy. Journal of General Internal Medicine. 1991;6:249–258. [PubMed: 2066832]

  • Myers AM, Holliday PJ, Harvey KA, Hutchinson KS. Functional performance measures: Are they superior to self-assessments? The Journals of Gerontology: Medical Sciences. 1993;48:M196–206. [PubMed: 8366262]

  • Nagi SZ. An epidemiology of disability among adults in the United States. Milbank Memorial Fund Quarterly. Health and Society. 1976;54:439–467. [PubMed: 137366]

  • Reuben DB, Siu AL, Kimpau S. The predictive validity of self-report and performance-based measures of function and health. The Journals of Gerontology: Medical Sciences. 1992;47:M106–110. [PubMed: 1624692]

  • Reuben DB, Solomon DH. Assessment in geriatrics: Of caveats and names. Journal of the American Geriatrics Society. 1989;37:570–572. [PubMed: 2654261]

  • Reuben DB, Valle LA, Hays RD, Siu AL. Measuring physical function in community-dwelling older persons: A comparison of self-administered, interviewer-administered, and performance-based measures. Journal of the American Geriatrics Society. 1995;43:17–23. [PubMed: 7806733]

  • Reuben DB, Walsh K, Moore AA, Damesyn M, Greendale GA. Hearing loss in community-dwelling older persons: National prevalence data and identification using simple questions. Journal of the American Geriatrics Society. 1998;46:1008. [PubMed: 9706892]

  • Rosow I, Breslau N. A Guttman Health Scale for the aged. The Journal of Gerontology. 1966;21:556–559. [PubMed: 5918309]

  • Roth J, Yen CJ. The role of intercellular communication in diseases of old age. In: Hazzard WR, Blass JP, Ettinger WH, Halter JB, Ouslander JG, editors. Principles of Geriatric Medicine and Gerontology. 4th edition. New York: McGraw-Hill; 1999. pp. 45–59.

  • Sheikh JI, Yesavage JA. Geriatric Depression Scale: Recent evidence and development of a shorter version. Clinical Gerontology. 1986;5:165.

  • Siu AL. Screening for dementia and investigating its causes. Annals of Internal Medicine. 1991;115:122–132. [PubMed: 2058860]

  • Steinberg EP, Tielsich JM, Schein OD, et al. The VF-14: An index of functional impairment in patients with cataract. Archives of Opthalmology. 1994;112:630. [PubMed: 8185520]

  • Tombaugh TN, McIntyre NJ. The Mini-Mental State Examination: A comprehensive review. Journal of the American Geriatrics Society. 1992;40:922–935. [PubMed: 1512391]

  • Van Coevorden AV, Mockel J, Laurent E, Kerkhofs M, l'Hermite-Balériaux M, Decoster C, Nève P, Cauter EV. Neuroendocrine rhythms and sleep in aging men. American Journal of Physiology. 1991;260:E651–E661. (Endocrinol. Metab. 23) [PubMed: 2018128]

  • Ventry IM, Weinstein BE. The hearing handicap inventory for the elderly: A new tool. Ear and Hearing. 1982;3:128–134. [PubMed: 7095321]

  • Verbrugge LM, Jette AM. The disablement process. Social Science and Medicine. 1994;38:1–14. [PubMed: 8146699]

  • Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36) Medical Care. 1992;30:473–483. [PubMed: 1593914]

  • Wiener JM, Hanley RJ, Clark R, Van Nostrand JF. Measuring the activities of daily living: Comparisons across national surveys. The Journals of Gerontology: Social Sciences. 1990;45:S229–237. [PubMed: 2146312]

  • World Health Organization. The International Classification of Impairments, Disabilities and Handicaps. Geneva: World Health Organization; 1980. [PubMed: 6449782]