Cardiorespiratory fitness, commonly referred to as exercise tolerance, declines with increasing age, typically beginning in the third decade. This decline has been attributed to disease processes, disuse, and aging (Bortz, 1982; Fries and Crapo, 1981). A minimum level of exercise tolerance is necessary to perform many tasks that are important to independent functioning. Low-level cardiorespiratory fitness is a major cause of functional difficulties in old age (Bruce, 1985); it is estimated that basic self-care tasks can use as much as 90 percent of the exercise capacity of a sedentary older adult (Blessey, 1986).
Body weight and body composition also have important implications for physical functioning of older persons. Both high and low weight have been associated with functional difficulties and disability in old age (Ensrud et al., 1994; Galanos et al., 1994; Harris et al., 1989; Launer et al., 1994; Pinsky et al., 1985). Heavy weight is also a major risk factor for several diseases and conditions, including diabetes mellitus and coronary artery disease, which in themselves can affect function. Excess body weight can also exacerbate symptoms associated with particular conditions, for example, osteoarthritis of the knee (Ettinger et al., 1994). Low weight, particularly when it results from weight loss in old age, can be indicative of poor or declining health and is a risk factor for mortality (Fischer and Johnson, 1990).
This chapter presents data on exercise tolerance and body composition in disabled older women participating in the Women's Health and Aging Study (WHAS).
Exercise tolerance was estimated using two different approaches: (1) a questionnaire-based assessment derived from the Specific Activity Scale developed by Goldman et al. (1981) and (2) direct measurement using the seated step test developed by Smith and Gilligan (1983), administered by a nurse in the participant's home.
The Specific Activity Scale used in the WHAS was modified to improve its applicability to older disabled women. A few of the very demanding activities were excluded and some low-level activities were added. The modified scale consists of 18 activities organized into six groups according to the estimated metabolic equivalents (METs) required to perform the activity. (One MET is equal to 3.5 milliliters of oxygen consumed per kilogram of body weight per minute, the average value for oxygen consumption at rest). The following groups of activities describe exercise tolerance from lowest to highest: level 1 (< 2 METs): sit quietly in a chair; level 2 (2 to 2.3 METs): dress, iron, stand for 2 hours, play cards; level 3 (3 to 5 METs): strip and make a bed, mop floors, handwash clothes, walk 2.5 mph, bowl; level 4 (4.5 to 5.2 METs): walk down a flight of stairs; level 5 (5 to 6 METs): carry a light parcel up stairs, garden, dance a fox trot, walk 4 mph; and level 6 (7 to 9 METs): carry 24 lbs. up 8 steps, shovel snow or spade soil, walk 5 mph or jog.
To determine exercise tolerance, participants were asked if they could do the first task in a group, regardless of any symptoms they experienced while doing the task. If they could not do the first activity in a group, they were asked if they could do the second activity and so on. If participants stated they could do any task in a group, they were classified as able to perform activities at the MET level the group represents. If they could not do any activity in a group they were classified as having exercise tolerance below that level. The exercise tolerance score is the highest activity level from 1 to 6 that the participant could perform regardless of symptoms. The pattern of questioning allows participants to skip over activities in a group that they do not or cannot do for reasons unrelated to cardiorespiratory fitness. To achieve a particular score, the participant need only report that she could do any one of the activities in a group.
Table 10.1 gives the exercise tolerance level of study participants as measured by the modified Specific Activity Scale. Sixty-six percent of the population had an exercise tolerance level between 3 and 6 METs (score of 3 to 5), indicating adequate cardiorespiratory fitness to perform activities equivalent to walking at least 2.5 miles per hour (normal walking pace) on level ground. Twenty-three percent had higher and 11 percent had lower levels of exercise tolerance. This 11 percent had barely adequate fitness to bathe (2.5 to 3.5 METs) and dress (2.4 to 4.0 METs) (Blessey, 1986).
Exercise tolerance level declined with increasing age and increasing severity of disability. Sixty-three percent of participants age 65 to 74 years could perform activities requiring 5 METs, compared with 57 percent of those 75 to 84 years and only 40 percent of women 85 years and older. The proportion of women with extremely low exercise tolerance among those age 85 years and older (19 percent) was more than twice that found in the youngest participants (8 percent). The disparity in fitness level across groups was even greater by disability level. While 71 percent of the moderately disabled and 56 percent of those with difficulty in activities of daily living (ADLs) who did not receive help could perform activities requiring 5 METs, only 26 percent of women who required help with ADLs had this level of exercise tolerance. Similarly, while 3 and 8 percent of the less severely disabled had extremely low exercise tolerance (score 1 or 2), 41 percent of women who required help in ADLs had extremely low fitness capacity.
For many tasks, particularly those in which speed is not a factor, the level of effort expended can vary tremendously across individuals. The Specific Activity Scale therefore represents only a crude approximation of fitness level. Objective measurement of exercise tolerance, that is, observed performance of standardized exertional tasks with known energy requirements, may be a more accurate assessment of fitness. For this reason, a graded exercise test was included in the nurse's baseline examination. Because of the low functional level and poor health status of a substantial proportion of the study population, a low-level graded exercise step test, performed in a seated position (Smith and Gilligan, 1983), was selected.
The seated step test is a four-stage, graded low-level test of exercise tolerance developed specifically for use in older adults (Smith and Gilligan, 1983). Stage 1 of the test goes to 2.3 METs, and stages 2, 3, and 4 go to 2.9, 3.5, and 3.9 METs, respectively. For comparison, 3.5 METs of energy expenditure is equivalent to walking 3.5 miles per hour, a fairly brisk pace. The test begins with the subject seated in a straight-backed chair with both feet flat on the floor. A step is placed in front of the subject such that when the leg is extended, the heel can reach the top of the step and the ankle is even with the edge. A metronome, set for 1 second beats, is used to keep time. On the first beat, the subject touches the front edge of the step with the arch of one foot, on the second she returns her foot to the floor, alternating feet. To ensure the safety of study participants, American College of Sports Medicine (ACSM) guidelines for exercise testing of older adults (ACSM, 1991) were used to establish (1) exclusion criteria for participation in the seated step test and (2) stopping criteria for persons who started the test. During testing, the participant wore an ambulatory electrocardiograph cabled to a notebook computer from which a rhythm strip was run during each stage, an oximeter finger probe that displayed pulse rate and oxygen saturation, and a blood pressure cuff. The Borg Perceived Exertion Scale (Borg and Linderholm, 1974) was used to ascertain how hard the participant felt she was working.
Each stage lasts 3 minutes. Stage 1 of the test uses a 6-inch step. After 2 minutes (beginning at 1 minute 45 seconds), heart rate and oxygen saturation were recorded, a 30-second rhythm strip was run, blood pressure was checked, the Borg scale was presented, and stopping criteria were evaluated. If no stopping criteria were reached and the participant was willing, the test continued for another minute. After 3 minutes, the nurse stopped and cleared the stopwatch, and again recorded Borg perceived exertion, heart rate, and oxygen saturation. If no stopping criteria were reached and the participant was willing, the test continued on to stage 2, using a 12-inch step. The participant was instructed to keep stepping between stages. The procedures for each stage are identical. Stages 3 and 4 use an 18-inch step, and stage 4 adds arm movement. Whenever stopping criteria were reached, the nurse took a blood pressure; ran a rhythm strip; and recorded heart rate, oxygen saturation, Borg perceived exertion, the presence of any signs and symptoms, and the stage at which the test was terminated.
Table 10.2 lists the exclusion criteria for the seated step test and the number and percentage of participants meeting each criterion. The most common reasons for exclusion included loud systolic murmur detected in the nurse examination; electrocardiogram (ECG) abnormalities, particularly wide QRS and atrial fibrillation; severe leg weakness as determined by the nurse; elevated blood pressure; and angina. The majority of the women who were excluded met only one of the exclusion criteria. A total of 442 participants (44.1 percent) met one or more exclusions; an additional 23 (2.3 percent) did not feel they could do the test. The rate of exclusion increased stepwise with increasing age from 34 percent of women age 65 to 74 years to over 55 percent of participants age 85 years and older. The rate of exclusion varied by disability level as well; about 40 percent of the moderately disabled and those with ADL difficulty who received no help and nearly 62 percent of the most severely disabled were excluded from the seated step test. complete the first 2 minutes of a stage, it was assumed she completed 1 minute. A total of 537 of the 1,002 participants (53.6 percent) attempted the test; only 33 were able to complete the entire test. The likelihood of stopping was about the same during a stage as between stages. Forty-five percent of the total study population (84 percent of those who started the test) completed the first stage which had an estimated MET value of 2.3; 24 percent (44 percent of those who started) completed the second stage with an estimated MET value of 2.9; 7 percent (13 percent of those who started) completed the third stage with an estimated MET value of 3.5; and 3 percent (6 percent of those who started) completed the fourth stage with an estimated MET level of 3.9. The highest drop-out rates occurred during the second stage and between the second and third stages.
Table 10.3 lists the reasons for stopping the test. Five hundred and four participants, 94 percent of those who started, stopped the seated step test before the end of the fourth stage. The two most frequent reasons for stopping, accounting for 70 percent of those who stopped, were related to participant reports of discomfort; 42 percent felt they could not continue and 28 percent perceived their exertion level to be 8 or greater. An additional 4 percent experienced shortness of breath. Very few participants met any of the clinical or objective stopping criteria. Less than 10 percent had an elevated heart rate, less than 3 percent had an elevated blood pressure, and less than 2 percent had abnormal ECG readings. Only one person (0.2 percent) with an abnormal ECG showed ST depression exceeding 1 millimeter, which resolved with cessation of exercise.
Figures 10.2 and 10.3 display the performance of participants who attempted the seated step test by age group and disability level, respectively. Participants age 65 to 74 and 75 to 84 years exhibited nearly identical patterns of test performance. Participants age 85 years and older stopped the test sooner and only one attempted (but did not complete) the fourth stage. Similar patterns of exercise performance were observed for disability level. The moderately disabled and women with ADL difficulty who received no help showed nearly identical test results; the most severely disabled group, women who received help with ADLs, exhibited much poorer performance on the seated step test. The most disabled women tended to stop the test sooner than the less disabled women; 61 percent completed the first stage and 13 percent completed the second stage, in contrast to the less disabled women, of whom 75 percent and 31 percent completed the first and second stages, respectively. Differences across age groups and disability levels in actual test performance were relatively small owing to the strict exclusion criteria.
Body composition and fatness are represented by body mass index (BMI), which is derived from measured weight in kilograms divided by height in meters squared, and by triceps skinfold thickness. Height and weight were measured with the participant standing in stocking feet wearing light indoor clothing. The participant's head was positioned against a level doorway using a Frankfort plane, and height was measured to the nearest centimeter using a stadiometer. Weight was measured in kilograms using a bathroom-type digital scale. Knee height was measured using a mediform sliding caliper (Medical Express, Beaverton, OR) with the participant in a semi-recumbent position with her left knee and ankle bent at 900 angles. When standing height was unavailable, height in centimeters was estimated from the average of two knee height measurements using the following formula: 84.88 + 1.83 (knee height) - 0.24 (age) (Chumlea et al., 1985).
Triceps skinfold thickness is a standard measure of body fatness and nutritional status. Skinfold thickness was measured at the midpoint of the upper right arm with Holtain skinfold calipers (Seritex, Carlstadt, NJ) to the nearest 0.2 millimeters in accordance with standard procedures (Lohman et al., 1988). If the difference between the first and second measure exceeded 2.0 millimeters, a third reading was done. Skinfold thickness values represent the mean of the first and second measures except when a third measure was taken which occurred for 70 (7 percent) of the participants. When the third measure fell within 2.0 millimeters of either the first or second measure, skinfold thickness was determined by averaging the third reading with the next closest one. If the third measure was within 2.0 millimeters of both the first and second measures, all three were averaged.
Table 10.4 gives the mean and median values of BMI and the percentages under- and overweight based on the 15th and 85th percentiles, respectively, of BMI derived from the first National Health and Nutrition Examination Survey (NHANES I) for females age 70 to 74 years (Must et al., 1991). Data are presented by age group, disability level, and race. For the total study population, the median BMI was 27.4 and the mean was 28.9, a reflection of extreme obesity in a small number of participants. The prevalence of overweight in this population was greater than in the general population of women age 70 to 74 years, with over 28 percent having a BMI above the 85th percentile value. Overweight, including extreme obesity (BMI greater than 35) was more prevalent in the Black than the White population, with over 37 percent having a BMI greater than the 85th percentile.
The prevalence of under- and overweight varied greatly by both race and age group. Women age 65 to 74 years had the highest rate of overweight, with 34 percent of the White and 49 percent of the Black women having a BMI above the 85th percentile. Underweight in this age group was rare. The prevalence of overweight dropped to 23 percent in women age 75 to 84 years and to 12.5 percent in the oldest participants. This decline was observed for both Blacks and Whites, although it was somewhat steeper for Blacks. Mirroring the decline in overweight was an increase in the prevalence of underweight with increasing age, from 6 percent in those age 65 to 74 years up to 20 percent in women age 85 years and older. Extreme low weight was not common in this population and underweight was no more prevalent in these disabled women than in community-dwelling women of similar age (Must et al., 1991).
For the most part, the prevalence of extremes of BMI did not vary with severity of disability. The two exceptions were the relatively high rate of underweight (20 percent) in the most disabled White women and the relatively low rate of underweight (4.6 percent) in the most disabled Black women.
Table 10.5 presents the mean and median values of triceps skinfold thickness and the percentages of low and high values based on the 15th and 85th percentiles, respectively, of skinfold thickness derived from NHANES I for females age 70 to 74 years (Must et al., 1991). Data are presented by age group, disability level, and race. The mean thickness was 21.9 millimeters and the median was 20.7 millimeters which reflects the presence of extreme obesity in a small proportion of participants. Racial differences in skinfold thickness were less pronounced than those for BMI; Black participants had only slightly higher values than White participants and more similar percentages at the lower extreme.
Triceps skinfold thickness gives a somewhat different impression of the relative fatness of the study population than does BMI. On the basis of skinfold thickness, there does not appear to be a higher than expected prevalence of obesity; only 15 percent had a skinfold measurement at the 85th percentile or above. In contrast, more than 25 percent had a skinfold thickness at the 15th percentile or below. Analogous to the trend seen with BMI, the prevalence of high skinfold thickness was lower and the prevalence of low skinfold thickness was higher in the older age groups. Skinfold thickness did not vary much with disability level, with the exception of a moderately higher percentage of women with low values in the most disabled subgroup.
This population of disabled older women exhibits a broad range of exercise capacity, as assessed by both self-report and objective testing. The modified Specific Activity Scale (Goldman et al., 1981) gave a higher estimate of exercise tolerance than the seated step test (Smith and Gilligan, 1983), particularly considering that nearly half the study participants (those most likely to have the lowest levels of cardiorespiratory fitness), were excluded from the seated step test. Exercise tolerance assessments based on self-reported capacities, like the Specific Activity Scale, may overlook the role of compensatory strategies, such as reduced speed of performance and amount of work performed, and frequent rest periods (Fried et al., 1991). Extremely low exercise tolerance was common, yet low cardiorespiratory fitness did not consistently underlie functional limitation and disability in this population.
Amount of body fatness was also broadly distributed. On the basis of BMI, the study population appeared disproportionately overweight, particularly the African American women and those in the youngest age group. This observation is consistent with the well-established association between overweight and functional difficulties in older women (Ensrud et al., 1994; Galanos et al., 1994; Harris et al., 1989; Launer et al., 1994; Pinsky et al., 1985). The prevalence of severe underweight was relatively low in this disabled population, somewhat contrary to expectations (Galanos et al., 1994) and the conceptualization of frailty (Buchner and Wagner, 1992). The oldest old, however, tended to be thinner and had a markedly different distribution of BMI than younger women, suggesting an alteration in the relationship between physical function and body weight with increasing age.
American College of Sports Medicine. (1991). Guidelines for Exercise Testing and Prescription, 4th ed. Philadelphia, PA:Lea & Febiger.
Blessey RL. (1986). Energy cost in the physically impaired geriatric population. Topics Geriatr Res 2:33-43.
Borg G, Linderholm H. (1974). Perceived exertion and pulse rate during graded exercise in various age groups. Acta Med Scand 472:194-206.
Bortz WM. (1982) Disuse and aging. JAMA 248:1203-1208.
Bruce RA. (1985). Functional aerobic capacity, exercise, and aging. In: Andres R, Bierman EL, Hazzard WR, eds. Principles of Geriatric Medicine. New York: McGraw-Hill, Inc.
Buchner DM, Wagner EH. (1992). Preventing frail health. Clin Geriatr Med 8:1-17.
Chumlea WC, Roche AF, Steinbaugh ML. (1985). Estimating stature from knee height for persons 60 to 90 years of age. J Am Geriatr Soc 33:116-120.
Ensrud KE, Nevitt MC, Yunis C, Cauley JA, Seeley DG, Fox KM, Cummings SR. (1994). Correlates of impaired function in older women. J Am Geriatr Soc 42:481-489.
Ettinger WH, Davis MA, Neuhaus JM, Mallon KP. (1994). Long-term physical functioning in persons with knee osteoarthritis from HANES I: Effects of comorbid medical conditions. J Clin Epidemiol 47:809-815.
Fischer J, Johnson MA. (1990). Low body weight and weight loss in the aged. J Am Dietetic Assoc 90:1697-1706.
Fried LP, Herdman SJ, Kuhn KE, Rubin G, Turano K. (1991). Preclinical disability: Hypotheses about the bottom of the iceberg. J Aging Health 3:285-300.
Fries JF, Crapo LM. (1981). Vitality and Aging. New York: WH Freeman and Co.
Galanos AN, Pieper CF, Cornoni-Huntley JC, Bales CW, Fillenbaum GG. (1994). Nutrition and function: Is there a relationship between body mass index and the functional capabilities of community-dwelling elderly? J Am Geriatr Soc 42:368-373.
Goldman L, Hashimoto B, Cook EF, Loscalzo A. (1981). Comparative reproducibility and validity of systems for assessing cardiovascular functional class: Advantages of a new Specific Activity Scale. Circulation 64:1227-1234.
Harris T, Kovar MG, Suzman R, Kleinman JC, Feldmen JJ. (1989). Longitudinal study of physical ability in the oldest-old. Am J Public Health 79:698-702.
Launer LJ, Harris T, Rumpel C, Madans J. (1994). Body mass index, weight change, and risk of mobility disability in middle-aged and older women. JAMA 271:1093-1098.
Lohman T, Roche A, Mastasell R. (1988). Anthropometric Standardization Reference Manual. Champagne, IL: Human Kinetics Books.
Must A, Dallal GE, Dietz WH. (1991). Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht2) and triceps skinfold thickness. Am J Clin Nutr 53:839-846.
Pinsky JL, Branch LG, Jette AM, Haynes SG, Feinleib M, Cornoni-Huntley JC, Bailey KR. (1985). Framingham disability study: Relationship of disability to cardiovascular risk factors among persons free of diagnosed cardiovascular disease. Am J Epidemiol 122:644-656.
Smith EL, Gilligan C. (1983). Physical activity prescription for the older adult. Physician Sports Med 11:91-101.