Evaluation of a group exercise program for elderly women.

Evaluation of a Group Exercise Program for Elderly Women Ample intuitive and scientific information indicates that physical performance decreases as age increases. For example, cross-sectional studies indicate that aerobic power decreases at the rate of 0.40 to 0.45 mL [O.sub.2..kg.sup.-1..min.sup.-1] per year, and longitudinal studies have found a decrease of 1.0 mL O.sub.2..kg.sup.-1..min.sup.-1] per year. [1] Active individuals, however, show less decrease in aerobic power (cardiopulmonary endurance) and skeletal muscle strength than sedentary individuals. [2-4] A relatively slow rate of decline in physical performance is likely when malnutrition, disease, genetic defects, and hypokinetic life styles are eliminated or minimized. [5,6] Given these circumstances, we are likely to observe minimal performance deficits up to the age of 85 years and perhaps higher. [5] Certainly, evidence exists that performance deficits are minimal in active individuals up to the age of 65 years. [7] After age 65 to 75 years, a combination of factors typically occurs to contribute to a more rapid rate of decline. These factors are a sedentary life style, the effects of aging, and the cumulative effects of chronic health problems. Further study is needed to more clearly describe the rate of decline in physical performance and the trainability of persons over the age of 75 years.

Most research concerning the effectiveness of exercise for counteracting the ill effects of a hypokinetic life style on cardiopulmonary function has been conducted on relatively young individuals. In the past, it was customary to classify anyone over 55 or 60 years of age as "old." In addition, previous studies have been limited by restricting the types of exercises to jogging, treadmill exercise, or bicycle ergometer exercise. Given the rapidly increasing number of elderly individuals, the need for inexpensive exercise in this population will continue to expand. Information on the effectiveness of exercises that can be performed by groups of elderly subjects without expensive equipment, large gymnasiums, or outdoor facilities is critical. The purpose of this study was to evaluate the physical training effect of a specific set of calisthenics performed by a supervised group of elderly women. We hypothesized that eight weeks of physical training with the specific set of calisthenics would result in decreases in heart rate (HR), systolic blood pressure (SBP), and rate-pressure product (RPP) (SBP X HR) responses to given submaximal work rates and to increases in predicted maximal aerobic power (MAP).

Review of Literature

Repetitive physical exercise using the lower extremities or a muscle mass greater than one-seventh of the total muscle mass has been demonstrated to be effective for physical training when performed at appropriate intensities and durations by healthy young and middle-aged adults and by patients with cardiac disease. [8-17] Recently, several reviews and recommendations for physical training programs for the elderly have been published. [1,6,18-20] The consensus is that exercise for the elderly is feasible and is likely to be effective for improving cardiopulmonary endurance. The evidence, however, should be examined carefully before making inferences concerning the effectiveness of specific programs.

Most physical training programs evolved from theories or programs originally developed for middle-aged individuals at risk for cardiac disease or from specific cardiac rehabilitation programs. For example, Adams and deVries studied the effectiveness of 12 weeks of calisthenics, walking-jogging, and muscle stretching performed for 60 minutes, three times pr week. [21] Seventeen women with an average age of 65.9 years participated in the physical training program, and 6 women with an average age of 66.7 years who did not participate in the training program constituted the control group. All subjects were evaluated during pretraining and posttraining sessions. They used a bicycle ergometer to perform submaximal graded exercise tolerance (GXT) tests that included the monitoring of HR, pulmonary ventilation, and oxygen uptake ([Vo.sub.2]). The physical work capacity (watts achieved on the GXT tests) increased significantly for the trained subjects but not for the control subjects. Training resulted in greater efficiency of cardiovascular responses to the GXT tests. Trained subjects consistently demonstrated greater decreases in HR responses (pretraining minus posttraining values) to each submaximal work rate of the GXT tests than the untrained control group subjects. After training, however, the [Vo.sub.] at 85% of the predicted maximal HR measured at the final stage of the GXT tests was not significantly different between the trained group and the control group. Blood pressures were not reported for GXT test results.

Because previous investigators had shown little or no increase in aerobic power as a result of low- to moderate-intensity physical training of the elderly, Seals et al conducted a study of the effects of low- and high-intensity training. [22] Low-intensity training consisted of continuous walking for 20 to 30 minutes at least three times per week and encouragement to increase general daily activity. High-intensity training consisted of a 10- to 15-minute warm-up period of muscle stretching and slow walking followed by 30 to 45 minutes of endurance exercise at a HR of 75% of HR reserve. This endurance exercise consisted of level walking progressed to jogging, bicycle ergometer exercise, or grated tradmill walking. Pretraining and posttraining GXT tests were conducted on a treadmill using the Bruce protocol. Physiological criteria were used to determine [Vo.sub.2]. Eleven subjects (7 men, 4 women) aged 62 [+ or -] 2 years, completed the training, which consisted of six months of low-intensity training followed by six months of high-intensity training. A statistically significant increase in [Vo.sub.2] occurred after low-intensity training (0.8 MET, (*1) p [is less than] .05) and again after high-intensity training (1.34 METs, p [is less than] .01). Overall, the subjects' [Vo.sub.2] increased 2.14 METs, from an average of 7.26 to 9.4 METs. One MET is the average [Vo.sub.2] during rest in a sitting position.

The elderly subjects of the studies of Adams and deVries [21] and Seals et al [22] were relatively young. Obviously, a need exists for studies of subjects older than the average ages of 65.9 and 62 years of the studies by Adams and deVries [21] and Seals et al, [22] respectively. Other studies of older subjects have not demonstrated a training effect for the cardiopulmonary system, [23-25] have not tested for a cardiopulmonary training effect, [26,27] or have used training methods that require unreasonably large gymnasiums or expensive equipment (bicycle ergometers or treadmills). [28,29]

Even though existing scientific evidence, general reviews, and feasibility studies indicate that physical training is likely to be beneficial for elderly subjects, the effectiveness of specific exercise programs still needs to be demonstrated for specific populations of elderly subjects. A large number of people currently live in high-rise apartments reserved for the elderly. These residents have unique needs for physical training, and these buildings almost always have space that can easily be used for group exercise. This space has usually been set aside for meetings, dining, recreational activities, or parties. Exercise rooms equipped with stationary bicycles, bicycle ergometers, treadmills, and other exercise equipment are not unusual. Even well-equipped exercise rooms, however, almost never entice or allow large numbers of residents to train at appropriate intensities with optimal routines. The ability of the people living in these buildings to travel to off-site exercise programs or even to walk outside is often limited by subclinical cardiopulmonary disease, orthopedic problems, obesity, chronic pain, metabolic disorders, vision and hearing impairments, and environmental hazards. Environmental hazards could include snow and ice, cold or hot weather, busy street crossings, or high-crime areas. The advantages of group exercise offered at the high-rise apartment appear obvious.

Ideally, all exercise programs and recommendations should be evaluated prior to routine or general implementation. Studies of the effectiveness of specific convenient and inexpensive group exercise programs for residents of apartments reserved for the elderly are needed.

Method

Subjects

Volunteers from a high-rise apartment complex supervised by the Minneapolis Housing Authority served as subjects for the study. The protocol for testing and training was presented to each subject in written format as a component of the consent form and verbally by one of the investigators (CTE). The project was approved by the University of Minnesota Committee on the Use of Human Subjects. All subjects completed a preliminary screening that consisted of the review of a medical history form, a resting single-lead electrocardiogram, and resting blood pressure and HR. An electrocardiograph, (*2) consisting of an ECG monitor, oscilloscope, and HR meter, and a Gould strip chart recorder (*3) were used to obtain the ECGS. Subjects were not tested if the resting ECG showed abnormalities not previously documented and cleared by the attending physician. The ECG was considered abnormal if any of the following were present: premature ventricular contractions exceeding six per minute, ST-segment depression or elevation greater than 1 mm 0.08 second after the J point, ventricular or atrial arrhythmias, or bundle branch blocks. Subjects were not tested if the resting HR exceeded 100 beats per minute (bpm), the SBP exceeded 200 mm Hg, or the diastolic blood pressure exceeded 110 mm Hg.

Procedure

Both the preliminary screening and the pretest were scheduled for one visit. Testing or training was terminated or delayed pending a physician's approval if the preliminary screening or further testing appeared to be abnormal. Measurements of height and weight were obtained for all subjects.

Following these measurements, cardiopulmonary fitness was assessed by a submaximal GXT test. All subjects were then invited to participate in an eight-week group exercise program. After the training period, all subjects who completed the pretest were invited to participate in the posttest. Subjects who completed the pretest and the posttest, but were unable to attend exercise, classes because of conflicts in their schedules, served as the control group.

Graded Exercise Tolerance Test 

The submaximal GXT test used in this study involved a stepping ergometer designed for testing elderly or sedentary subjects. [30-34] The step heights and cadences are described in Table 1. Heart rate and ECG readings were constantly displayed during the GXT test. Electrocardiographic tracings were recorded routinely during the last 20 seconds of each exercise stage. Blood pressure was measured as soon after each exercise stage as possible. Delay time for BP measurement was estimated to be 15 to 20 seconds.

The GXT test consisted of stepping in the pattern shown in Figure 1 at a constant freqency of 20 mounts per minute and 80 counts per minute for three minutes at each stage of the test and progressively increasing the height of stepping for additional test stages. The following test protocol was established: 1) Stage 1--level stepping, 2) Stage 2--stepping up and down on a 4-in (*4) platform, 3) Stage 3--stepping up and down on an 8-in platform, and 4) Stage 4--stepping up and down on a 12-in platform. The test was terminated if the HR exceeded 75% of its range, symptoms or signs of exercise intolerance were observed, or special precautions needed to be followed because of a medical history of arthritis or severe cardiopulmonary disease. For convenience, target HR ranges for the GXT test were selected from Table 2. [8,9] The GXT test was terminated at a target HR of at least 60% but not greater than 75% of the age-predicted HR range. The test was terminated earlier if the following objective signs of exercise intolerance were observed or reported: 1) the ECG results were abnormal, 2) the SBP exceeded 225 mm Hg, 3) the diastolic blood pressure exceeded 130 mm Hg, or 4) the SBP decreased more than 20 mm Hg. The GXT test was also terminated if the subject asked to stop or reported or demonstrated shortness of breath; chest, neck, jaw, or arm pain; back, hip, knee, or ankle pain; dizziness, nausea, or confusion; general fatigue or fatigue of the lower extremities; or sudden pallor or inappropriate sweating. [30,35] Tests were not progressed to target HRs if the medical history included reports of significant joint pain (especially in the hips or knees), a diagnosis of significant cardiopulmonary disease, severe osteoporosis, or a rate of increase in the HR that indicated the target HR would be significantly exceeded on the next stage of the test.

Each three-minute exercise stage was followed by a one-minute pause to allow the measurement of BP without extraneous noise from stepping. The one-minute pause also allowed time to position the appropriate stepping platform. Standing BP was measured immediately after the subject stopped stepping. Blood pressure was measured noninvasively using a standard stethoscope and aneroid gauge sphygmomanometer. One investigator (LRA) measured all BPs.

Training Regimen

The physical training consisted of group exercise supervised by a licensed physical therapist (JMD). Formal exercise sessions were held twice per week, and subjects were encouraged to exercise with a partner one additional session per week. Subjects were taught to monitor their own pulse rate and to exercise within a target HR zone. The target HR zone was established on the basis of the pretraining GXT test. Because of the submaximal design of the GXT test, the training HR zone was set at a range starting 10 bpm below the highest asymptomatic HR and progressing only as high as 75% of the age-predicated maximal HR range. The target HR range, however, was never set higher than the highest asymptomatic HR observed on the pretraining GXT test. Name tags labeled with the target HR were used during all group exercise sessions to facilitate the monitoring of exercise intensity.

A calisthenics program was designed to exercise all major muscle groups of the body (Figs. 2, 3). Exercises 1 to 5 were designed to allow a gradual warm-up and to improve local muscle endurance (Fig. 2). Exercises 6 to 8 were designed primarily to promote cardiopulmonary endurance and secondarily to improve lower extremity local muscle endurance (Fig. 3). Exercises 9 to 12 were designed primarily to promote flexibility and to provide a supervised "cool-down" perod.

During each session, subjects monitored and reported HR prior to starting exercise 1, before starting exercise 6, after completing exercise 8, and at the end of the session. The exercises were progressive as shown in Table 3. This progression was designed to accommodate the most adaptive or the most fit subjects. Subjects were instructed to add rest periods or to bend their knees less for exercises 6 to 8, when HR exceeded the target range, or when any pain or unusual fatigue was experienced. Exercise 1 and exercises 3 to 5 were regulated primarily by the resistance given by the partner, rather than by intermittent rest periods. All exercises were performed to music, which assisted in pacing and added variety to the program.

Data Analysis

All statistical analyses were performed on a personal computer with a statistical analysis package. (*5) Means and standard deviations were calculated for all subjects for physiological measurements. We performed t tests for related measures to test for changes in submaximal GXT HR, SBP, and RPP over the training period for the exercise and control groups.

Identical work rates for submaximal GXT test responses were compared for pretest versus posttest results. For example, if the highest work rate achieved on the pretest was 3.4 METs, comparisons were calculated for responses to 3.4 METs on the posttest, even if the subject was capable of progressing to the 4.7-MET level. In addition, predicted MAPs recorded before and after training were compared using the t test for related measures. Predicted MAP was calculated using the following formula [15,36]:

Predicted Map = GXT MET / 141 (GXT HR/age-predicted HR) - 42

where GXT MET is the highest MET level performed on the GXT test, GXT HR is the HR recorded at GXT MET, and age-predicted HR is 220 minus the subject's age (in years).

The highest achieved MET levels on the pretest and posttest were used to calculate the predicted MAP. Percentages of change were also calculated to allow a convenient quantification of direction and magnitude of change over the training period for the exercise and control groups. Measurements of secondary interest (age, height, and weight) were reported only as descriptive statistics.

Results

Initially, 36 elderly residents from a high-rise apartments supervised by the Minneapolis Housing Authority volunteered for the exercise program. No subjects were eliminated because of resting ECG, HR, or blood pressure abnormalities during the preliminary screening. The medical history screening procedure eliminated two subjects from the study because of their inability to perform the step test. Five female subjects who participated in both the pretest and the posttest, but were unable to attend exercise sessions because of other commitments during the time scheduled for group sessions, served as the control group. Fifteen subjects were eliminated from the study because they attended fewer than half of the group sessions. Fourteen female subjects constituted the exercise group. These subjects rarely performed the recommended third session per week. Demographic data for all exercise group and control group subjects are shown in Table 4.

The GXT tests were stopped at Stage 1 for six subjects at Stage 2 for two subjects because they achieved their target HR. The GXT test was stopped at Stage 2 for one subject because she complained of shortness of breath. Extra precautions necessitated by a history of painful hips or knees secondary to arthritis, heart disease, pulmonary disease, implantation of a fixed-rate pacemaker, or severe osteoporosis caused us to terminate the GXT test of six subjects after Stage 1 and three subjects after Stage 2. One subject's GXT test was stopped after Stage 2 because her HR rate of increase between Stage 1 and Stage 2 was considered to be excessive.

Statistically significant decreases (p [is less than] .05) were observed after training for resting SBP and RPP and for submaximal GXT HR, SBP, and RPP for the exercise group (Tab. 5). Predicted MAP increased 12.4% after training for the exercise group, but this increase was not statistically significant. Resting HR and RPP did not change significantly for the control group; however, the resting SBP did decrease significantly on the posttest for the control group. Submaximal GXT HR did not decrease over the training period for the control group; however, submaximal GXT SBP did decrease significantly on the posttest for the control group.

Discussion

For the experimental group, all physiological variables changed in the direction that indicated a training effect. Physical training is expected to improve the efficiency of the cardiovascular and respiratory systems. The observed decreases in submaximal GXT HR, SBP, and RPP are indicative of improved efficiency of cardiovascular adjustments required for oxygen delivery and utilization. Heart rate, SBP, and RPP responses to known submaximal work rates are readily-available objective quantitative measures. When these tests are performed as a component of pretraining and posttraining assessments, the effectiveness of the treatment for an individual patient or group of patients can readily be determined. Given that all of these measures decreased significantly in our study, we would logically conclude that training was effective for our exercise group. Other studies of the effectiveness of physical training on the elderly have apparently not reported the statistical significance of the changes. Indeed, it appears that only Adams and deVries even mentioned these changes at repeated submaximal levels. [21] The relative magnitude of the change in RPP (-13.9%) in our study is in agreement with the changes reported by May and Nagle (-14.87%) for patients with coronary artery disease. [11]

The control group was analyzed to evaluate the effects of familiarization with the testing procedure. Any novel experience is likely to cause slight increases in HR and SBP. It is also possible that the practice will result in greater mechanical efficiency in the performance of the step test. The significant decrease in GXT SBP in the control group could be due to either of these reasons. Novelty or anxiety, however, will increase HR as much as or more than SBP and cause relatively greater increases during the resting measurements, which are taken during the peak of anticipation, than during exercise. Because the HR responses of the control group did not decrease on the posttraining step test, an increase in mechanical efficiency or psychological adaptation to circumstances perceived to be novel, exciting, or threatening is not likely to have caused the observed decreases in SBP. Other explanations for the decreases in SBP are possible. A high percentage of the subjects were taking medication to control high blood pressure. Random or systematic changes in the length of time between ingestion of medication and testing could have occurred for control subjects between the pretest and the posttest. The relatively small number of subjects in the control group would make it more sensitive to these possible changes. No subjects, however, reported changes in dosage or types of medication at the posttest review of the medical history.

The method used for predicting MAP needs to be discussed. Obviously, the prediction of hypothetical maximal value from different submaximal variables is likely to introduce systematic or random error. An inspection of the formula used for predicting MAP indicates that two submaximal measured values and one value estimated by age are required. The actual measurement of HR from ECG strips obtained during the step test is not likely to introduce significant error. Heart rate, which was constantly displayed on the HR meter, appeared to be quite stable during the last 20 seconds of each exercise stage; aging has been shown to decrease the variability of HR during the respiratory cycle. [37] The MET levels for this step test were established from data obtained on young adults rather than older adults; however, oxygen cost is not likely to change with age unless the mechanical work of stepping is altered by factors indirectly related to aging. The factors that could alter the pattern of stepping include osteoarthritis, mal-alignment related to healed hip fractures, cerebrovascular insufficiency, and peripheral nerve involvement. Obvious failure to follow the prescribed stepping cadence and pattern was not present in either the experimental group or the control group.

Estimates of MAP should be used primarily for comparing pretraining and posttraining or treatment performance for the same subject or for gross fitness classifications. Ideally, predicted MAP values should be used for inter-group comparisons only if end points and GXT test protocols are identical. Comparisons between values obtained from different protocols should be made with caution, because each method of predicting MAP is likely to contain specific characteristics that introduce subtle differences into the calculated MAP. For example, differences between MAPs measured with different modes of exercise have been documented, but these differences have been determined to have minimal practical implications. The results of tests conducted with a bicycle ergometer, treadmill, or step test yield comparable results for young healthy subjects and for patients with cardiac disease unless the subject has obvious problems with the performance of the test. [16] Additionally, the advantages of predicting MAP from submaximal values rather than establishing MAP from volitional maximal work rates must be noted. Volitional maximal values are likely to introduce error unless stringent criteria are used to document that MAP was actually measured. The MAP can be documented objectively, for example, if the following criteria are met: 1) [Vo.sub.2] no longer increases when work rate is increased, 2) exercise HR equals or exceeds the age-predicted maximal HR, 3) the respiratory exchange ratio ([Vco.sub.2/Vo.sub.2]) exceeds 1.15, and 4) blood lactate four minutes after exercise is at least 8 mmol per liter of blood. [22] The difficulties and dangers of routinely following these guidelines when testing the elderly appear obvious.

Our MAP values can be compared with values obtained in other studies. Morse and Smith reported MAP values of 5 to 7 METs for 65- to 75-year-olds and 2 to 4 METs for ambulatory nursing home residents. [38] They determined volitional maximal values with treadmill tests. Even though their previous work had demonstrated the feasibility of maximal tests for the elderly, Sidney and Shephard considered it prudent not to attempt maximal tests on elderly, sedentary volunteers for a study of the effects of physical training. [28,39] The average MAP they predicted from the submaximal bicycle ergometer tests of their 60- to 83-year-old subjects was 6.12 METs. The average age was not reported, but this preretirement fitness training group is likely to approximate an average age of 65 years. Thompson et al performed symptom-limited tests for MAP using the Balke treadmill protocol on a sample of 58 people, 64 to 83 years of age. [24] Four subjects had positive test results (eg, ST-segment depression) and were eliminated from the study, 3 were unable to walk on the treadmill, and 10 exceeded the arbitrary upper limit of 8.1 METs. The average MAP (5.5 METs) was reported only for the subjects who completed the study (n = 21). Thus value is similar to our exercise group's mean pretest and posttest predicted MAP values of 4.5 and 5.1 METs, respectively. Our subjects could be classified as having average to low fitness for their respective age group.

Even though elderly subjects have demonstrated a training effect from intensities as low as 40% to 50% of the HR range, we used a training intensity goal of 60% to 75% and modified this target HR range according to the results of the pretest GXT. The average percentages of the HR range achieved on the GXT test were 51.8% for the exercise group and 57.2% for the control group. When expressed simply as a percentage of age-predicted maximal HR, these values were 72.0% for the exercise group and 74.4% for the control group. Each subject was assigned a minimum and a maximum target HR for training. The minimum and maximum values averaged 94.6 and 101.5 bpm, respectively. The average training HR was 41.4% to 50.1% of the available HR range. When expressed as a simple percentage of the age-predicted maximal HR, the percentages were 66.0% to 71.0%

Even with these apparently conservative target HRs, we were more likely to have difficulty reaching than exceeding the target HR. Amundsen et al, however, demonstrated that exercise HRs reported by the subjects tend to be lower than HRs measured by the exercise leaders. [30] Subjects tend to count accurately but to be slower to locate the pulse and to begin counting after exercise has ceased. Nevertheless, further study of similar subjects using higher target HRs appears to be advisable to explore the possibility of obtaining greater and statistically significant gains in predicted MAP. Despite the absence of statistically significant gains and because of the relatively small number of subjects in our study, our observed percentage of increase in MAP (12.4%) after eight weeks of training compares favorably with the gain observed (12%) for the six months of low-intensity training used by Seals et al. [22]

Conclusion

The results suggest that this particular training regimen, which consists of calisthenics that require very little space and no equipment other than common straight-back chairs and possibly background music, can be applied to groups of elderly subjects safely and productively. The decrease in submaximal HR, SBP, and RPP, and the increase in predicted MAP indicate that a training effect can be expected from this or similar physical exercise training protocols performed by elderly subjects.

(*1) MET (metabolic equivalent) = 3.5 mL [O.sub.2.kg.sup.-1.min.sup.-1]

(*2) Model PM 2A, Electronics for Medicine, 30 Virginia Rd, White Plains, NY 10603.

(*3) Model 220, Gould Inc, 3631 Perkins Ave, Cleveland, OH 44114.

(*4) 1 in = 2.54 cm.

(*5) NH Analytical Software, 1958 Eldridge Ave, Roseville, MN 55113.

References

[1] Stamford BA: Exercise and the elderly. In Pandolf KB (ed): Exercise and Sports Reviews. New York, NY, Macmillan Publishing Co, 1988, vol 16, pp 341-379

[2] Dehn MM, Bruce RA: Longitudinal variations in maximal oxygen intake with age and activity. J Appl Physiol 33:805-807, 1972

[3] Drinkwater BL, Horvath SM Wells CL: Aerobic power of females, ages 10 to 68. J Gerontol 30:385-394, 1975

[4] Aniansson A, Sperling L, Rundgren A, et al: Muscle function in 75-year-old men and women: A longitudinal study. Scand J Rehabil Med [Suppl] 9:92-102, 1983

[5] Fries JF, Crapo LM: Vitality and Aging: Implications of the Rectangular Curve. San Francisco, CA, W H Freeman & Co Publishers, 1981, pp 1-9, 25-40

[6] Grimby G: Physical activity and muscle training in the elderly. Acta Med Scand [Suppl] 711: 233-237, 1986

[7] Shephard RJ: Gross changes in form and function. In Shephard RJ (ed): Physical Activity and Aging. Chicago, IL, Year Book Medical Publishers Inc, 1978, pp. 52-145

[8] Karvonen MJ, Kentala E, Mustala O: The effects of training on heart rate. Annales Medicinae Experimentatus et Biologie Fenniae 35:307-315, 1958

[9] Amundsen LR: Establishing activity and training levels for patients with ischemic heart disease. Phys Ther 59:754-758, 1979

[10] Carter CL: Cardiac rehabilitation of outpatients during the recovery phase following myocardial infarction. In Amundsen LR (ed): Cardiac Rehabilitation: Clinics in Physical Therapy. New York, NY, Churchill Livingstone Inc, 1981, pp 111-119

[11] May GA, Nagle FJ: Changes in rate-pressure product with physical training of individuals with coronary artery disease. Phys Ther 64: 1361-1366, 1984

[12] Stippig J, Berg A, Keul J: Medical and physiological aspects. In Amundsen LR (ed): Cardiac Rehabilitation: Outpatient Physical Training Methods. Rockville, MD, Aspen Publishers Inc, 1988, pp 3-6

[13] Graf RS: Rehabilitation during the acute and convalescent stages following myocardial infarction. In Amundsen LR (ed): Cardiac Rehabilitation: Clinics in Physical Therapy. New York, NY, Churchill Livingstone Inc, 1981, pp 99-109

[14] Hagberg JM, Ehsani AA, Holloszy HO: Effect of 12 months of intense exercise training on strok volume in patients with coronary heart disease. Circulation 67:1194-1199, 1983

[15] Hellerstein HK, Hirsch EL, Adler R, et al: Principles of exercise prescription for normals and cardiac subjects. In Naughton J, et al (eds): Exercise Testing and Exercise Training in Coronary Heart Disease New York, NY, Academic Press Inc, 1973, pp 129-167

[16] Shephard RJ, Allen C, Benade AJ, et al: The maximum oxygen intake. Bull WHO 38:757-764, 1968

[17] Blomquist CG, Saltin B: Cardiovascular adaptations to physical training. Annu Rev Physiol 45:169-189, 1983

[18] Vallbona C, Baker SB: Physical fitness prospects in the elderly. Arch Phys Med Rehabil 65: 194-200, 1984

[19] Payton OD, Poland JL: Aging process: Implications for clinical practice. Phys Ther 63: 41-48, 1983

[20] Frontera WR, Evans WJ: Exercise performance and endurance training in the elderly. Topics in Geriatric Rehabilitation 2:17-32, 1986

[21] Adams GM, deVries HA: Physiological effects of an exercise training regimen upon women aged 52 to 79. J Gerontol 28:50-55, 1973

[22] Seals DR, Hagberg JM, Hurley BF, et al: Endurance training in older men and women: I. Cardiovascular responses to exercise. J. Appl Physiol: Respirant Environ Exercise Physiol 57: 1024-1029, 1984

[23] Benestad AM: Trainability of old men. Acta Med Scand 178:321-327, 1965

[24] Thompson RF, Crist DM, Marsh M, et al: Effects of physical exercise for elderly patients with physical impairments. J Am Geriatr Soc 36:130-135, 1988

[25] Gutman GM, Herbert CP, Brown SR: Feldenkrais versus conventional exercises for elderly. J. Geontol 32:532-572, 1977

[26] Smith EL, Reddan W, Smith PE: Physical activity and calcium modalities for bone mineral increase in aged women. Med Sci Sports Exerc 13:60-64, 1981

[27' Frekany GA, Leslie DK: Effects of exercise program on selected flexibility measurements of senior citizens. Gerontologist 15:182-183, 1975

[28] Sidney KH, Shephard RJ: Frequency and intensity of exercise training for elderly subjects. Med Sci Sports 10:125-131, 1978

[29] Barry AJ, Daly JW, Pruett ED, et al: The effects of physical conditioning on older individuals. I. Work capacity, circulatory-respiratory function, and work electrocardiogram. J. Gerontol 21:182-191, 1966

[30] Amundsen LR, Ellingham CT, Eliason ML, et al: Cardiac considerations and physical training. In Jackson O (ed): Physical Therapy of the Geriatric Patient: Clinics in Physical Therapy. New York, NY, Churchill Livingstone Inc. 1983, pp 181-201

[31] Amundsen LR, Ellingham CT: Cardiac considerations and physical training. In Jackson O (ed): Physical Therapy of the Geriatric Patient: Clinics in Physical Therapy, ed 2. New York, NY, Churchill Livingstone Inc, to be published

[32] Nielsen DH, Amundsen LR: Exercise physiology: An overview with emphasis on aerobic capacity and energy cost. In Amundsen LR (ed): Cardiac Rehabilitation: Clinics in Physical Therapy. New York, NY, Churchill Livingstone Inc, 1981, pp 11-28

[33] Amundsen LR, Ellingham CT, Shore S: Cardiopulmonary Fitness Test. Abstract. Cardiopulmonary Record 2(3):13, 1987

[3] Amundsen LR, Ellingham CT: Safe and effective exercise for the elderly. In: First World Congress on Allied Health, Abstracts--Access to Quality Health Care for all: The Provision of Human Resources. Elsinore, Denmark, June 14-17, 1988

[35] Amundsen LR: Assessing exercise tolerance: A review. Phys Ther 59:534-437, 1979

[36] Patterson RP: Home computer based performance evaluation. In: Proceedings of the Twentieth Annual Rocky Mountain Bioengineering Symposium and the Twentieth International ISA Biomedical Sciences Instrumentation Symposium. Rochester, MN, April 18-19, 1983, pp 77-81

[37] Hrushesky WJM, Fader D, Schmitt O, et al: The respiratory sinus arrhythmia: A measure of cardiac age. Science 224:1001-1004, 1984

[38] Morse CE, Smith EL: Physical activity programming for the aged. In Smith EL, Serfass RC (eds): Aging and Exercise. Hillsdale, NJ, Enslow Publishers, 1981, pp 109-129

[39] Sidney KH, Shephard RJ: Maximum and submaximum exercise tests in men and women in the seventh, eighth, and ninth decades of life. J Appl Physiol 43:280-287, 1977

L Amundsen, PhD, PT, is Director and Assistant Professor, Graduate Studies in Physical Therapy, Department of Medicine and Rehabilitation, University of Minnesota, PO Box 388 UMHC, Minneapolis, MN 55455 (USA).

J DeVahl, MS, PT, is Marketing Clinical Therapist, Medtronic, Incorporated, 6951 Central Ave NE, PO Box 1250, Minneapolis, MN 55440. She was a student in the master's degree program in physical therapy, University of Minnesota, when this study was conducted.

C Ellingham, MS, PT, is Assistant Professor, Physical Therapy Program, Department of Physical Medicine and Rehabilitation, University of Minnesota.

This article submitted June 15, 1988; was with the authors for revision for 12 weeks; and was accepted January 27, 1989.