++
A hallmark of geriatrics is emphasis on the functional ability of older patients. This subject is discussed from a clinical point of view in several chapters in this book. This approach recognizes that although individual diseases are important and that our system of modern medicine is oriented toward the diagnosis and treatment of specific diseases, the consequences of single and multiple diseases can be understood best by evaluating the functional status of the patient. A large body of epidemiologic work undertaken over the past two decades has treated disability as a condition that can be studied in much the same way as if it were a well-defined chronic disease by using epidemiologic tools to assess prevalence, incidence, and a wide range of risk factors. This work has led to a greater understanding of the occurrence, determinants, and consequences of disability in the older population and has provided insights into strategies for the prevention of disability.
+++
Measures of Function and Disability
++
Measures of disability were originally developed for use in the clinical setting and were aimed at quantifying the impact of severe medical conditions such as stroke on physical and mental functioning, obtaining standard information on the rate and degree of recovery from these conditions, and assessing work ability and the need for formal and informal care. These assessment tools were gradually applied in clinical research and population-based studies, and almost all research studies in older populations now assess disability status. Federal data collection efforts did not include very old populations as recently as 30 years ago, but these surveys now have no upper age limit and include various instruments to assess disability. This type of assessment is illustrated in Figure 5-17, which is based on data from the Medicare Current Beneficiary Survey. Activities of daily living (ADLs) are basic self-care tasks. Instrumental ADL (IADLs) are tasks that are physically and cognitively somewhat more complicated and difficult than self-care tasks and are necessary for independent living in the community. ADLs and IADLs are measures of disability and reflect how an individual’s limitations interact with the demands of the environment. With increasing age the prevalence of ADL disability increases rapidly, and the proportion of persons with no limitations decreases to under 30% in persons 85 years and older. Additionally, the prevalence of disability is higher in women compared to men of the same age.
++
++
Disability has been assessed with a wide variety of instruments, and even when instruments contain the same items, they may differ in how they assess specific aspects of performing the task or the severity of limitation in performing the task. For example, it is debatable whether it is better to ask people whether they actually perform a specific task or whether they should be asked to judge whether they could perform the task even if they have not done it for long time. The former approach gives more concrete information, but respondents could be classified as disabled simply because they have chosen not to do a task that they are perfectly capable of performing. The latter approach allows for classification of people regardless of whether or not they have actually done the task recently but is limited because respondents who have not done the task are required to speculate as to their ability. For this and other issues in disability assessment, there is no single best way to perform an assessment, and there is therefore no single instrument that is ideal. The lack of standardization that results from the use of multiple competing instruments makes it difficult to compare rates of disability across studies, as differences in prevalence rates result from different methods of assessment. Disability classified as having any difficulty performing a task has a higher prevalence than disability classified as requiring human assistance or being unable to perform. Disability classified as requiring human help or help from an assistive device has a higher prevalence rate than disability requiring human help only and is similar in prevalence to the classification using difficulty. In one study of individual ADL, the assessments that relied on difficulty produced estimates of disability 1.2 to 5.0 times greater than assessments that used human assistance as the criterion for disability. Thus, in evaluating research that reports on disability rates, it is critical to examine the way questions are asked and how responses are used to determine the presence of disability.
+++
Disability as an Indicator of Health Status and Prognosis
++
Disability status is one of the most potent of all health status indicators in predicting adverse outcomes. This is probably because disability measures are able to capture the impact of the presence and severity of multiple pathologies, including physical, cognitive, and psychological conditions, as well as the potential synergistic effects of these conditions on overall health status. The hierarchy of disability statuses are defined as independent in ADL, independent but reporting some difficulty with ADL, and dependent on personal assistance for ADL. For these three groups, the cumulative probabilities over a 4-year follow-up period of being admitted to a nursing home and mortality are greater for individuals classified as dependent and least for individuals who are independent. These results support the discussion in the previous paragraph that different disability criteria can identify population subgroups with different prognoses. They also support the validity of disability measurement, with level of disability being highly predictive of two important outcomes.
+++
Prevalence of Disability
++
The higher prevalence rate of disability in women shown in Figure 5-18 has been repeatedly demonstrated and is consistent across a variety of measures and in studies using both self-reporting and objective assessments. Indeed, the fact that women live longer than men in spite of their worse health and higher rate of disability at all ages is one of the most interesting paradoxes in the epidemiology of aging. Another valuable measure of disability that has received increasing attention is mobility disability and is usually evaluated using questions related to walking and climbing stairs. Figure 5-18A shows that mobility disability, defined as the inability to walk 1/2 mi or climb stairs, increases with age and has a higher prevalence in women than in men at all ages after 65 years. The dynamics of this difference in prevalence are highly instructive. Prevalence is a snapshot at a single point in time and is a function of the flow into and out of the condition being assessed. In the case of mobility disability, women have a higher incidence (new occurrence) of disability (Figure 5-18B) and are less likely to exit from the disabled state than men because they have lower mortality than men when they are disabled (Figure 5-18C) and because they are less likely to recover from disability than men (Figure 5-18D). Thus, the prevention of disability and better treatment of women who have become disabled are likely to have a beneficial effect in lowering the disability rates of women to the level of those of men. Reduction in mortality in disabled men would also contribute to reducing gender differences in disability prevalence.
++
+++
Risk Factors for Disability
++
In addition to gender, there is a wide array of risk factors for disability in older populations. The relative risk of disability for many risk factors is consistent across different disability definitions, providing strong support for the importance of these risk factors. The most commonly and consistently reported risk factors for disability are listed in Table 5-9, which comes from a review of 78 community-based longitudinal studies that assessed factors related to decline in functional status in older persons. Risk factors are divided into two sections: behavioral factors and individual characteristics, and specific chronic conditions. There is a great deal of interdependence between these risk factors. Behavioral risk factors such as physical inactivity and smoking can promote the development of a variety of diseases, which can then go on to cause disability. These behavioral risk factors may also have direct effects of their own. For example, physical inactivity may be a risk factor for the onset of specific diseases, and it also may have a direct negative impact on muscle, bone, and the central and peripheral nervous systems. These changes can move an individual closer to the physiologic and functional threshold beyond which functioning is impaired to the point that disability occurs. Figure 5-19 shows the relative contributions of specific conditions to disability according to age group. These data were obtained by asking persons who reported a limitation in their activities (work limitation or need for assistance in ADLs or IADLs) to specify the health condition that was responsible for that limitation. Arthritis and heart or other circulatory diseases make by far the largest contributions to physical limitations, but it is notable that sensory limitations are also reported to make large contributions. Dementia is reported by proxies to be a substantial contributor to activity limitations in those age 85 and older.
++
++
++
The role that cognitive impairment plays in physical disability has been generally underappreciated. For clinicians attending to severely disabled persons in nursing homes, however, it is evident that dementia itself prevents these individuals from being independent. Figure 5-20, based on data from population-based studies in Tuscany, Italy, shows estimates of the numbers of men and women with ADL disability in 1999 according to age. It also separates these people into those with and without dementia. There is an overall increase in the number of disabled persons through the mid-eighties in men and through the late eighties in women, with a drop-off in the numbers after that because of the decline in the total number of persons in the population at these advanced ages. The obvious gap in these otherwise smooth curves is a result of the low birthrate during and just after World War I, which translated into a smaller population 80 years later. In both men and women, most ADL disability before age 75 is not associated with dementia. From age 75 to 90 about half of ADL disability is accompanied by dementia, and after age 90 the majority of persons with ADL disability have dementia. These data do not prove that dementia is what caused the disability in these individuals. Serious physical impairments and diseases co-occur with dementia, and it may not be possible to always understand just what the cause of disability is. Nevertheless, it is impressive that such a large proportion of disabled persons have dementia, and it is clear that cognitive functioning must be considered when developing interventions to prevent or treat disability.
++
+++
Recovery From Disability
++
Longitudinal epidemiologic studies have revealed much about the dynamics of disability onset and progression. Contrary to previous belief, a substantial proportion of individuals who are disabled report improvement on subsequent assessments. In effect, disability is a product of the disease or diseases from which an individual suffers, sedentary lifestyle or disuse, and physiologic declines that may be the result of aging or pathologic processes that are not specific diseases but result from factors such as inflammation or endocrine changes. As these predisposing conditions change, they have an impact on the initiation of disability and on changes in the status of already established disability. Among community-dwelling persons age 70, recovery rates were high and long-term recovery was quite common among those with any disability. Recovery rates were also high in persons who had disability reported for 2 or more consecutive months (persistent disability) and 3 or more consecutive months (chronic disability). Other studies have observed high rates of improvement in disability status over intervals, ranging from 1 to 3 years despite missing transient improvement due to long time intervals between assessments. Observed improvement in function could result from inaccurate self-reporting of disability over time (unreliability of instrument), a true change in disability status, or both. In fact, both these explanations play a role, but a certain amount of real improvement in disability is consistently found. Age less than 85, good cognitive function, remaining physically mobile, absence of depression, and good social support have all been associated with a greater probability of recovery.
+++
Types of Disability Progression
++
In understanding the dynamics of disability progression it is useful to consider the pace at which disability develops. The terms “progressive disability” and “catastrophic disability” have been used, indicating a slow downhill course or a very rapid decline, respectively. Progressive disability results from one or more ongoing chronic conditions and causes disability over months or years, whereas catastrophic disability can occur in moments as a result of a stroke or a hip fracture. The prevalence of both progressive and catastrophic severe ADL disability (defined as needing help with three or more ADLs) increases with increasing age, although progressive disability rises faster than catastrophic disability. Among older persons with severe ADL disability, the proportion that has catastrophic ADL disability is much higher at younger ages, and the proportion that has progressive ADL disability is much higher at the oldest ages (Figure 5-21). A similar age pattern has been found for onset of severe mobility disability (inability to walk across a room), which is much more common in people who have three or more chronic conditions.
++
++
The dynamics of disability can also be approached by studying the pathologic changes that precede its onset. Most disability results from disease, and different theoretical pathways have been proposed to describe the changes that occur as a person proceeds from disease to disability. The theoretical pathway that has received substantial empirical support in aging research is that proposed by Nagi and endorsed by the Institute of Medicine. In this pathway, two intermediate steps, impairment and functional limitation, follow disease and lead to disability (see Verbrugge and Jette, 1994). Impairment describes the dysfunction and structural abnormalities in specific body systems that result from pathology. Functional limitation describes restrictions in basic physical and mental actions that result from impairments. Functional limitations are the basic building blocks of functioning, and the interaction of these components of functioning with environmental demands faced by an individual determines whether that person is disabled (see Guralnik and Ferrucci, 2003). Conventional study of the consequences of disease describes the physiologic organ impairments that result from specific conditions. More recent work in aging has gone on to describe further steps in the pathway. Impairments such as poor balance, muscle weakness, and visual deficits have a clear impact on functional limitations and disability. Furthermore, the relationship between functional limitations, such as reduced gait speed, and subsequent disability also supports the pathway.
+++
Physical Performance Measures
++
Objective measures of physical performance have received increasing attention as assessments that can measure functioning in a standardized manner in both the research and clinical settings. These measures can be used to represent impairments, functional limitations, or actual disability, but most are indicators of functional limitations. A short, standardized battery of performance tests was administered to a large number of participants in the Established Populations for the Epidemiologic Study of the Elderly (EPESE). This battery, which included gait speed, time required to rise from a chair and sit down five times, and hierarchic measures of balance, was used to create a summary score of lower extremity performance that ranges from 0 to 12. This measure, the short physical performance battery (SPPB), predicts mortality, the need for nursing home admission, and health care utilization in the overall older population. Furthermore, in a population that had no disability at the time the SPPB was administered, the score was highly predictive of who developed ADL and mobility disability 1 and 4 years later. These findings indicate that there is a state of preclinical disability, expressed as impairments and functional limitations, that indicates a high risk of proceeding to full-blown disability. This approach could identify high-risk older persons for whom preventive interventions may be highly effective.
++
Objective performance measures also provide a means of comparing functional status over time or across countries or cultures, where disability measures may lose comparability because of environmental differences or differential access to assistive devices. Figure 5-22 shows results from the English Longitudinal Study on Aging, a nationally representative sample of older persons in England. It demonstrates the prevalence according to age and sex of poor physical performance, documented as an SPPB score of less than or equal to 8 and gait speed of less than 0.5 m/s. Performance below these cut points is strongly associated with multiple adverse outcomes. Poor performance affects only about 10% of persons in their 60s but the prevalence rises rapidly in the 70s and attains very high levels in persons above 80. Women have higher rates of poor performance than men at all ages.
++
++
Because disability status is a good way of representing overall health status in older persons with complex patterns of disease, and because disability also has direct implications for the long-term care needs of an older person, there has been much interest in evaluating disability trends over time. Although a number of national surveys now assess disability, uniform disability assessment done over time has been available only since the mid-1980s in just a few studies with nationally representative samples. Although these studies use different assessment instruments, a decline in age- and gender-specific rates of disability was observed from the mid-1980s through the 1990s. The National Long Term Care Survey has similar assessments of ADL and IADL disability available from 1982 to 2005, and indicates that the decline in disability observed for the first 12 years of the study continued and actually accelerated from 1994 to 2005 (see Manton et al., 2006). In another study that utilized reports of functional limitations, including lifting and carrying 10 lb, climbing stairs, walking 1/4 mi, and seeing words in a newspaper, changes in prevalence were evaluated between 1984 and 1993. Declines were seen in the ability to perform all four of these tasks in the 65 years and older population over this period as well as the 80 years and older population. The functional limitations evaluated in this study, which assess more basic tasks than disability, are an excellent way to follow trends over time because they are influenced less by changing roles that can affect disability assessment (more men cooking in more recent surveys and more women managing money).
++
However, there is mounting evidence that the decline in disability prevalence among older adults is reversing. Analyses comparing two periods of NHANES data (1988–1994 vs 1999–2004) indicate that the prevalence of ADL, IALD, and mobility disability increased among those 60 to 69 years (see Seeman et al., 2010). For those 70 to 79 years, a similar pattern was reported, although the decline was not significant. The oldest group, those 80 years and older, were the only group to show a trend of declining disability, but only for functional limitations. Additionally, the change in disability prevalence was varied by race/ethnicity. Non-Hispanic blacks had a greater increase in ADL disability among those 60 to 69 years and in functional limitations for those 70 to 79 years over this period than among non-Hispanic whites of the same age groups. For those 80 years and older, the observed decline in disability did not differ by race/ethnicity, but women had a greater reduction in functional limitations over this period than did men. The reasons for the increase in disability among those aged 60 to 69 years remain unclear. However, it is suggested that this increase may be a byproduct of the rapidly growing number of overweight and obese Americans. In the analyses by Seeman et al., overweight and obese 60- to 69-year-olds reported greater increases in IADL disability than those of normal body weight in the same age group. Reciprocally, among those 80 years and older, those of normal weight were less likely to report mobility disability over this time period. The increase in disability is particularly concerning, as it appears to be occurring among the fastest-growing subgroups of older adults, those 60 years and older, non whites, and those overweight and obese. The increasing burden of disability among older adults is of public health significance for all ages. The impact of this increase could be far-reaching and negatively affect both the US health care system and economy.
++
There is an interplay among time of disability onset, duration of disability, and time of death that determines the number of years that older individuals live in the disability-free state, termed active life expectancy, and the number of years spent in the disabled state. Life table approaches have been used to calculate active and disabled life expectancy, utilizing data from population-based longitudinal studies on transitions from the nondisabled state to disability and death and from the disabled state to nondisability and death. Using this approach may provide insight into the mechanisms and risk factors that affect the quality of aging and suggest potential targets for intervention. In EPESE low education (< 12 years) was associated with shorter total life expectancy and shorter active life expectancy in both black and white women and men. However, when comparing persons with the same educational status, there were only small differences between blacks and whites. Alternative methods have been developed to estimate active life expectancy that utilize mortality data and prevalence of disability, which is easier to obtain in a national survey than 1-year transition probabilities. As more data become available to estimate both active and disabled life expectancy, we will gain more insight into the prospects for a compression of morbidity, which is the reduction in disabled life expectancy that results from compressing chronic disease and disability into a smaller number of years between disease and/or disability onset and mortality.
+++
Aging of Individuals With Life-Long Disabilities
++
In addition to the large number of persons who will develop disabilities in old age, there is a smaller cohort of persons with life-long disabilities who will enter old age with these disabilities. Persons with developmental disabilities are a heterogeneous population with varying abilities, and their disabilities may result from a variety of conditions such as cerebral palsy, mental retardation, learning disorders, autism, and epilepsy. It is estimated that in 2000 there were over 600,000 people in the United States with mental retardation and other developmental disabilities. Their numbers will double to 1.2 million by 2030 when all of the post–World War II “baby boom” generation will be in their sixties. Life expectancy of many persons with mental retardation and developmental disabilities has risen dramatically, and many people with these disorders will live into advanced old age. Down syndrome is an exception, however, with signs of accelerated aging and development of Alzheimer disease at early ages. A further consideration related to the growing number of older persons with developmental disabilities is the continuation of their care when their own parents reach old age. Nearly two-thirds of persons with developmental disabilities live with their families, and in one-quarter of these households the primary caregiver is 60 years or older. Over the next 30 years, there will be a considerable increase in the number of families where parents more than 80 years old are caring for an older child with a developmental disability.