Can elderly patients who have had a hip fracture perform moderate- to high-intensity exercise at home?

Background and Purpose. The majority of patients after a hip fracture do not return to prefracture functional status. Depression has been shown to affect recovery. Although exercise can reduce impairments, access issues limit elderly people from participating in facility-based programs. The primary purpose of this study was to determine the effects and feasibility of a home exercise program of moderate- or high-intensity exercise. A secondary purpose was to explore the relationship of depression and physical recovery. Subjects. Thirty-three elderly people (24 women, 9 men; [bar.X] = 78.6 years of age, SD = 6.8, range = 64-89) who had completed a regimen of physical therapy following hip fracture participated in the study. Subjects were randomly assigned to a resistance training group, an aerobic training group, or a control group. Methods. Subjects were tested before and upon completion of the exercise trial. Isometric lower-extremity force, 6-minute-walk distance, free gait speed, mental status, and physical function were measured. Each exercise session was supervised by a physical therapist, and subjects received 20 visits over 12 weeks. The control group received biweekly mailings. The resistance training group performed 3 sets of 8 repetitions at the 8-repetition maximum intensity using a portable progressive resistance exercise machine. The aerobic training group performed activities that increased heart rate 65% to 75% of their age-predicted maximum for 20 continuous minutes. Results. Resistance and aerobic training were performed without apparent adverse effects, and adherence was 98%. All groups improved in distance walked, force produced, gait speed, and physical function. Isometric force improved to a greater extent in the intervention groups than in the control group. Depressive symptoms interacted with treatment group in explaining the outcomes of 6-minute-walk distance and gait speed. Discussion and Conclusion. High-intensity exercise performed in the home is feasible for people with hip fracture. Larger sample sizes may be necessary to determine whether the exercise regimen is effective in reducing impairments and improving function. Depression may play a role in the level of improvement attained. [Mangione KK, Craik RL, Tomlinson SS, Palombaro KM. Can elderly patients who have had a hip fracture perform moderate- to high-intensity exercise at home? Phys Ther. 2005;85:727-739.]

Key Words: Exercise, aerobic performance; Exercise, force production; Hip fractures; Home care services.

Hip fracture is a common medical problem that can drastically change the quality of life for the elderly person. More than 300,000 older people are expected to fracture a hip each year (1) at an estimated cost of $5 billion. (2) It has been well established that the majority of patients with hip fracture do not return to prefracture functional status at 1 year after surgery. (3-7) In a study of 120 people, Marottoli et al (5) showed that, 6 months after hip fracture, only 8% climbed a flight of stairs, 15% walked across a room independently, and 6% walked a half mile. Tolo et al (8) reported results from a sample that required no assistive devices for walking before a hip fracture; however, 8 months after a fracture, 42% of the sample required a cane, their walker use tripled, and 56% of the sample reported not walking "as well" as they could before the fracture. Even at a 2-year follow-up, patients with a hip fracture are reported to be 4 times more likely to be homebound, 3 times more likely to be dependent in basic activities of daily living (ADL), and more likely to spend less time on their feet when compared with control subjects. (9)

To varying degrees, exercise and physical activity have been shown to be effective in reducing impairments, functional limitations, and disability in elderly people who are healthy. (10-12) Initiating exercise programs for elderly people with disabilities, however, is reported to be difficult because of problems with access to exercise facilities. (13) Home-based exercise is an approach to address the problem of environmental access for patients after hip fracture.

Only 2 trials in the literature have examined the effectiveness of home-based exercise for patients after hip fracture. Sherrington and Lord (14) examined the effectiveness of performing unsupervised, daily "step-up" exercises in patients 7 months after a hip fracture. A physical therapist determined the number of repetitions and the height of the step that the patients used. After 1 month, isometric force production of the quadriceps femoris muscle improved in the exercise group approximately 3.0 kg and gait speed increased 0.05 m/s.

Tinetti and colleagues (15,16) compared outcomes of "usual care" rehabilitation and "systematic multicomponent rehabilitation (SMR)" at 6 and 12 months in 304 patients receiving rehabilitation in the home after hip fracture. "Usual care" was not defined, but SMR included daily performance of 3 sets of 8 repetitions for seated hip flexion, hip abduction, knee extension, and ankle dorsiflexion. All patients exercised with elastic bands and began with the band of least resistance. Patients also performed transfer, balance and gait training, and range-of-motion (ROM) exercises. There were no differences between groups at either time period when examining recovery of prefracture basic ADL and instrumental activities of daily living (IADL) or in measures of physical performance (gait, muscle force, and balance). Changes in outcomes from the beginning of the intervention to the 6- or 12-month follow-up were not reported.

The studies of home-based exercises reported to date do not provide sufficient detail to determine the intensity of exercise and do not provide guidance to establish exercise protocols for patients after hip fracture. The program described by Sherrington and Lord (14) did not report training intensity, but the results suggest that exercise in people who have had a hip fracture may have positive effects on impairments and functional limitations. Tinetti and colleagues' (15,16) program was comprehensive in scope, but it is not clear whether the training intensity was adequate to improve impairments or functional limitations.

Because there are limited data to guide exercise prescription needed to remediate impairments in patients after hip fracture, applying the evidence from other groups of elderly people appears to be warranted. Highintensity muscle force training improves muscle force production, gait speed, and balance in elderly people who are healthy and those who are frail. (11,17) Training typically involves 3 sets of 8 repetitions at 80% of the one-repetition maximum for 8 to 16 weeks. (18,19) Training has been directed to single muscle groups, such as the knee extensors, and groups of functionally related muscles, such as the hip and knee extensors together. Likewise, endurance training has been examined in elderly people who performed 20 to 40 minutes of exercise for 12 to 16 weeks ranging from low to high intensity (40% to 70% of heart rate reserve). (12,20) The types of training performed by elderly people who are healthy and those who are frail include walking, cycling, and water and land aerobics. Training has been shown to improve maximal oxygen consumption and 6-minute-walk distance as well as measures of health status. Because the fracture site is healed, it appears safe to apply the principles emerging from the rapidly growing body of knowledge regarding exercise prescription in elderly people who are healthy or frail to patients with hip fracture.

Many factors, including depression, have been cited as determinants of recovery after a hip fracture. Depression in people with hip fracture has been associated with greater physical disability and the need for longer rehabilitation. (21,22) For example, in one study that followed 196 Caucasian women 1 year after fracture, women with persistently elevated depressive symptoms were 3 times less likely to achieve independence in walking and 9 times less likely to return to prefracture function than women reporting few depressive symptoms. This study controlled for age, prefracture physical function, and cognitive status. (22) A recent study by Scaf-Klomp et al (23) examined the effect of incomplete recovery and depression 1 year after fall-related injury. Unlike previous studies, this investigation tracked changes in depression during recovery and showed that depressive reactions occurred only when physical function appeared to stagnate. (23) These findings emphasize an important link between depression and physical recovery after hip fracture and have led investigators to suggest that clinicians should screen for depression and initiate appropriate medical intervention. (24) It is also important to further explore the role that depression plays in measures of physical performance after hip fracture.

The primary purpose of this study was to determine the effects of a 12-week program of high-intensity, supervised resistance training or moderate-intensity aerobic training on specific impairments, functional limitations, and disabilities in people after hip fracture. The primary outcome variables were maximum voluntary isometric lower-extremity force, 6-minute walk distance, free gait speed, and self-reported physical function. A secondary purpose was to determine the feasibility of such an exercise program. Feasibility, or ability to perform the exercises at home, was defined as adherence to scheduled appointments, number of sessions the subject was able to achieve the target intensity without reports of muscle soreness or shortness of breath, and number of sessions that the routine was not altered because of reports of pain. Finally, because depression is another factor reported to affect hip fracture recovery, an additional purpose of this investigation was to explore the relationship of depression and physical recovery.

Method

Subjects

The principal investigator (KKM) contacted physical therapists in a 10-mile radius of the research center to identify potential subjects for the study. One hundred three subjects were contacted by the physical therapists from the various health care settings. The physical therapists described a broad overview of the study and asked patients to sign a written permission for contact. The principal investigator called each prospective subject and performed a cursory telephone screening. Inclusion criteria included successful fixation (partial or total hip replacement or open reduction internal fixation [ORIF]) of a hip fracture, 65 years of age or older, living at home, willing to come to Arcadia University for testing before and after the intervention, and discharged from physical therapy. Exclusion criteria included a medical history of unstable angina or uncompensated congestive heart failure, metabolic conditions (such as renal dialysis) that preclude resistance training or aerobic training, history of stroke with residual hemiplegia, Parkinson disease, life expectancy of less than 6 months, Folstein Mini-Mental Status Exam scores less than 20, (25) and living in a nursing home.

Sixty subjects agreed to be interviewed (Fig. 1). If subjects met the initial criteria, the principal investigator visited them in their home, described the study in greater detail, and administered the Folstein Mini-Mental Status Exam. Subjects were informed that assignment to a group was determined by referring to a list of computer-generated random numbers and that if assigned to the wait-list control group, they would be eligible to receive either intervention at the end of the study. Participants then signed the informed consent and medical history release form and completed a form listing demographic characteristics, medical conditions, surgical procedure, date of surgery, other comorbidities, and medication usage. The principal investigator sent each subject's orthopedic surgeon a brief description of the study, a copy of the signed consent form, and a request that the surgeon prescribe 12 weeks of physical therapy for the patient.

[FIGURE 1 OMITTED]

Design

This randomized clinical trial included a baseline examination after patients completed the physical therapy regimen they received following their hip fracture and another examination after completion of the intervention. The patients were randomly assigned after baseline testing to aerobic training, muscle force training, or control groups. The physical therapist examiner was masked to group assignment and performed all testing at baseline and after treatment. Different physical therapists provided the interventions and were masked to outcome testing results. The patients who were assigned to the control group were able to receive the intervention of choice after completion of the study, but outcomes data following this intervention were not collected.

Demographic information was collected prior to the physical examination. The subjects' height and weight were measured on a calibrated physician scale (Health-O-Meter). * Resting heart rate was measured by palpation of the radial artery, and blood pressure was auscultated at the brachial artery and measured with a mercury sphygmomanometer. The Barthel Index of Activities of Daily Living and the Lawton Instrumental Activities of Daily Living Index also were administered to the subjects. (26,27) The Barthel Index scores range from 0 to 100, with 100 being complete independence. (26) The IADL scale scores range from 0 to 8, with 8 representing complete independence in IADL. (27) Each participant completed the Geriatric Depression Scale (GDS) (28) because depression is a known comorbidity that affects functional outcome in subjects with hip fracture. (22,99) The GDS is a 30-item yes/no questionnaire in which scores greater than 9 suggest the presence of depressive symptoms. (30) Upon completion of all testing, the subject was randomly assigned to the resistance training group, the aerobic training group, or the waitlist control group.

Testing Procedures

All testing occurred at the Arcadia University research facilities. Each subject was tested in this order: walking endurance, lower-extremity force production, and gait speed. Endurance was assessed using the 6-minute walk test, (31) in which the patient was instructed to "cover as much distance as possible in the next 6 minutes." Because of concerns about patient fatigue for the remaining tests, only one trial was performed. This test has reported reliability and validity and differentiates between elderly people living in retirement homes and those living in the community. (32) The subjects walked over a 100-foot-long (30 m) linoleum floor corridor and were provided with standardized verbal encouragement once per minute. Distance was recorded to the nearest foot and converted to meters.

Isometric force was measured with a handheld digital strain-gauge dynamometer (Chatillon Model 500) [dagger] that displays force measurements to the nearest 0.1 lb (0.045 kg) to a maximum of 500 lb (225 kg). The machine was factory calibrated before data collection and has reported accuracy of [+ or -] 1 lb ([+ or -] 0.45 kg). Maximal isometric force tests were obtained for hip extensors, hip abductors, knee extensors, and ankle plantar flexors, bilaterally. To measure the force of the hip extensors, the subject was placed in a supine position and the hip was passively flexed to 90 degrees and the knee was relaxed and supported on the examiner's shoulder. (33) The trunk was stabilized to the table using a mobilization belt around the subject's pelvis and another belt around the contralateral lower extremity. Resistance was applied just proximal to the knee on the posterior surface of the leg. (33,34)

The hip abductors were tested with the knee extended and the hip in a neutral position. Stabilization was the same as for the hip extensors. Resistance was applied just above the lateral femoral condyles. (33,34) The plantar flexors were tested with the hip and knee extended, the trunk and legs were stabilized as described earlier, and resistance was applied to the plantar surface of the foot at the metatarsal heads. The knee extensors were tested with the subject seated on a raised chair fixed to the ground and the hip and knee flexed to 90 degrees. The trunk was stabilized with a strap across the pelvis, the thigh was stabilized with a strap across the upper thigh, and the examiner was stabilized by a belt around the waist that was then attached to the leg of the chair. Resistance was applied just proximal to the ankle on the anterior surface of the leg.

For each test, the subject was asked to push as hard as possible as the physical therapist examiner matched the resistance the subject produced (make test). The subject performed one 5-second submaximal trial, followed by one maximal practice trial. After a 1-minute rest period, the subject performed 2 maximal effort trials, with a minute rest between trials. The mean peak force of the 2 trials was recorded. The training we prescribed did not target one specific muscle group, but rather the lower extremity as a whole; therefore, the isometric force values for each muscle group were summed to form a unilateral lower-extremity force score for each limb. Summed force scores have been reported in the geriatric training literature because individual lower-extremity muscle force scores are highly correlated. (35)

Temporal and spatial characteristics of gait were measured with the GaitMat II. [double dagger] The mat consists of an instrumented walkway 3.87 m long and 0.81 m wide. The walkway is divided into 38 rows and 256 columns of pressure-sensitive switches that are 15 [mm.sup.2]. The switches and circuitry are covered with black rubber. The switches are open until the subject's foot contacts them; the switches are reopened after the foot is removed. The time required to scan the entire array is 10 milliseconds. A computer constantly monitors the state of the switches. Data collected included step length, step time, swing time, double and single support times, base of support, and average walking speed. To become familiar with the walking surface, each subject was permitted several practice trials walking across the mat. A trial consisted of walking over the mat in one direction. The subject completed 3 trials at free gait speed in which he or she was instructed to "walk at your normal or comfortable pace," and we used the mean of the 3 trials for data analysis. Each subject determined whether a rest period was needed between the trials. Intraclass correlation coefficients (3,1) for the reliability of gait speed measurements have been reported to range from .90 to .99 for older women walking at a variety of speeds. (36)

We used the Medical Outcomes Study 36-Item ShortForm Health Survey (SF-36) physical function subscale to measure disability. Physical function is scored on a 0-to-100 scale, with 100 representing excellent health status. (37) The physical function subscale of the SF-36 has demonstrated internal consistency, test-retest reliability, and substantial validity on a wide range of patient and nonpatient samples. (38,39)

Prior to data collection, intrarater reliability of the examiner was established for all the tests included in the assessment. The physical therapist examiner, who had 21 years of experience, was trained in each of the procedures. Eight elderly people (1 man and 7 women) without hip fracture participated in the reliability testing. The mean age ([+ or -]SD) of the reliability sample was 79.2 [+ or -] 6.7 years, and these subjects took an average of 4.3 medications. The demographic characteristics were similar to people who sustain hip fractures. (4,16) Each subject was tested twice with at least 1 week separating trials. A second person recorded the data so that the examiner was masked to all data obtained. The reliability coefficients for this sample were found overall to be good to excellent (ICC [3,k] = .87 - .99 for lower-extremity isometric force, ICC [3,1] = .99 for 6-minute-walk distance, ICC [3,k] = .99 for gait speed, and ICC [3,1] = .86 for SF-36 physical function).

Infervention

Six physical therapists were trained to provide resistance training or aerobic exercise in subjects' homes and to complete daily forms documenting precise exercise prescription. The experience level of the physical therapists ranged from 13 years to 25 years. These daily documentation forms provided the data used for the feasibility assessment. Target intensity values and actual intensity values were recorded for each session. Reports of pain and treatment alteration because of pain also were noted daily in the comment section of the exercise form. The total duration for each session was 30 to 40 minutes. Exercise training consisted of 2 phases: (1) the "overload phase" of the interventions was 2 times per week on nonconsecutive days for the first 2 months, and (2) the "maintenance phase" was 1 time per week for the third month. There were a total of 20 visits. This 2-phase approach with decreasing frequency was used to mimic what might happen in home care with a more frequent visit schedule (2 times per week) decreasing to a maintenance phase of 1 time per week at the same intensity (Tab. 1).

Resistance training. Exercises were selected to target bilateral hip extensors, hip abductors, knee extensors, and plantar flexors. These muscles were chosen because of their role in function, specifically gait and transfer activities. (40,41) The intervention was performed using a portable progressive-resistive exercise machine (Shuttle MiniClinic)

[section] and body weight. The machine has 6 latex bands, each with a starting load equal to approximately 6 lb (2.7 kg). The bands provide resistance as the subject moves. At full excursion, one band can provide approximately 20 lb (9 kg) of resistance. The latex bands are attached to the machine by a slotted bar on the frame. Inserting more bands into the slotted bar increases the resistive load for the subject. A progress monitor strip is located on the top of the frame of the machine. The strip indicates the resistance by showing the distance that the load is moved. The numbers indicate the resistance for one band as the carriage is moved. When more than one band is used, the values are added.

The physical therapist determined the amount of resistance the subject could push against in order to complete a maximum of 8 repetitions (8-RM). Studies have suggested that the 8-RM is more effective than training at 10-RM or 2-RM, (42) but it is not so aggressive that it is associated with injuries. (43) The 8-RM also is strongly

related to the 1-RM, (44) and determining the 8-RM allowed the physical therapist to know the training intensity without further subcalculations (eg, 80% of the 1-RM). The subjects performed 3 sets of 8 repetitions at the 8-RM intensity. Intensity was reassessed every 2 weeks. This training routine has been demonstrated to produce muscle force gains in elderly people. (11,40,45) The physical therapists recorded the amount of resistance (number of cords and band length) for the hip and knee exercises. For the plantar flexors (unilateral or bilateral), the number of sets and the number of repetitions per exercise were recorded.

The subjects were positioned supine for the combination exercise of hip and knee extension as well as for the hip abduction exercise. For the combination hip and knee extension, each subject positioned his or her foot on the footplate and extended the lower extremity from approximately 90 degrees of hip flexion into full hip and knee extension against the predetermined resistance (Fig. 2A). For the hip abductors, the footplate was flattened, pillows were placed under the subject's buttocks, and the subject moved from 5 degrees of adduction to 10 degrees of abduction (Fig. 2B). Fifteen degrees of movement was chosen because it approximates the 8 degrees of motion associated with the stance phase of gait and takes into account the variations in hip positions in standing posture. (46)

[FIGURE 2 OMITTED]

Hip extensors were trained while standing. The initial position of the exercising leg was approximately 35 degrees of hip flexion, and the subject extended the hip to neutral position (Fig. 2C). This ROM approximates the phase in the gait cycle when the gluteus maximus muscle shows the highest electromyographic activity (ie, from heel contact through foot-flat phase). (47) Plantar flexors were strengthened in standing. The subject was asked to perform a unilateral heel-rise through full ROM (by visual inspection). If the subject was unable to complete the full ROM in good form (eg, flexing the knee or lurching the body in an attempt to use momentum to raise the heel), the subject performed bilateral heel-raises. Participants were able to hold onto an assistive device for balance or support.

Aerobic training. Subjects receiving aerobic training had their blood pressure and resting heart rate measured at the beginning of each session. Heart rate was measured with a Polar heart rate monitor [paragraph] worn during the treatment session or, if the subject had cardiac arrhythmia, by palpation of the radial artery. The physical therapist calculated the intensity of training based on the prediction equation of (maximum heart rate = 220 - age). (48) The value was then multiplied by both 65% and 75% to obtain the target heart rate range for training. This intensity has been shown to increase aerobic capacity in elderly people. (12,20) If the person took medications that altered heart rate response (eg, [beta]-blockers), the Borg Rating of Perceived Exertion Scale was used. (49) Perception of exertion has been shown to be strongly related to the physiological indicators of work (oxygen consumption and heart rate). (49,50) The training intensity using the Borg Rating of Perceived Exertion Scale was "moderate" to "strong" work as consistent with a rating of 3 to 5 on the 0-to-10 scale.

The aerobic intervention began with 2 to 3 minutes of warm-up active ROM exercise. The subject then walked on level surfaces and on stairs, if he or she was able, to keep the heart rate within the training zone for 20 minutes. If the participant was unable to walk for 20 continuous minutes of exercise, the physical therapist had the subject perform additional exercises such upper-and lower-extremity active ROM exercises to keep the heart rate elevated. The basic guidelines to which the physical therapist adhered were that the intensity and duration of the training were within the 65% to 75% of age-predicted maximal heart rate and that the duration was 20 continuous minutes. (48) The physical therapist recorded the specific activity, time spent performing each activity, and the average heart rate or perceived exertion per activity.

Control group. Subjects who were assigned to the control group received biweekly mailings of the National Institutes of Health "Age Pages" on a variety of nonexercise topics. Subjects in the control group were asked not to begin any new exercise programs until the study was completed. Control subjects were retested after 8 weeks. Eight weeks was the length of the first "overload phase" of the resistance and aerobic training and was chosen to minimize the likelihood that the subject would not drop out during the waiting period.

Data Analysis

The data were analyzed with SPSS software. (#) Descriptive statistics were used to describe the sample and feasibility. One-way analyses of variance (ANOVAs) were used to compare baseline demographic and outcome values for the 3 groups. The dependent variables were isometric force (summed hip abduction, knee extension, and plantar flexion) for the involved lower extremity, 6-minute-walk distance, free gait speed, and SF-36 physical function. Isometric hip extension force was not included in the lower-extremity summed force score because the tester was unable to stabilize several subjects adequately. Force production was examined for the involved lower extremity and the noninvolved lower extremity and by normalizing the values to body weight. Because the results were the same for each limb, only the summed isometric force values for the involved LE are presented. A one-way ANOVA was used to compare baseline measurements for all variables, and then the outcomes were assessed with a 2 x 3 repeated-measures ANOVA.

Although there were no statistical differences in baseline characteristics, there were apparent clinical differences in time after fracture and depression scores among the intervention groups and the control group. Additional analyses, therefore, were performed using general linear modeling to explore the relationship of depression and recovery. Four separate analyses were performed, one for each main outcome: summed isometric force, 6-minute-walk distance, free gait speed, and SF-36 physical function scores. Time after fracture and depression scores were used as covariates in the analyses. In each analysis, the dependent variable was the final outcome measure, the independent variable was treatment group, and the covariates were baseline values for the outcome, time after fracture, and depression score. In addition to the main effects, the role of depression was further explored by examining the interaction of depression scores and group. A significance level of .05 was used for all statistical analyses.

Results

Of the 103 potential subjects, 60 elderly people agreed to be interviewed. Forty-one of the 60 people were eligible, agreed to participate, and were randomly assigned to groups (Fig. 1). Reasons for nonparticipation included people who were lost to follow-up (n = 4), who refused to participate (n = 7), who did not meet the eligibility criteria (n = 6), or whose physician would not provide a prescription for participation (n = 2). Twelve people completed aerobic training, 11 people completed the resistance training, and 10 people were in the control group. Eight subjects dropped out of the study. One subject in the aerobic training group was unable to perform the exercise at the intensity level recommended, and, one subject in the control group dropped out because she thought the testing was "too much for her." The remaining subjects who dropped out were in the resistance training group: 4 were hospitalized, 1 was diagnosed with a progressive neuromuscular disorder midway through the training, and the final subject preferred "traditional, individually tailored physical therapy." Of those hospitalized, 1 was placed in long-term care, 2 died, and 1 had multiple intervening surgeries. None of the hospitalizations were related to the training. Comparisons of baseline values of the subjects who dropped out with those of the subjects who completed the intervention using t tests showed that there were no differences in performance variables of lower-extremity isometric force, 6-minute-walk distance, or free gait speed. The subjects who dropped out, however, had lower reported physical function as measured by the SF-36 (t = -2.02, P<.05).

The demographic characteristics of the patients who completed the study are reported in Table 2. The sample included 24 women and 9 men whose mean age was 78.6 years. The participants had a variety of comorbidities, including hypertension, hypercholesteremia, coronary artery disease, osteoarthritis, osteoporosis, diabetes, cancer, congestive heart failure, and depression. They were relatively independent in basic ADL and required some assistance for IADL. They received a variety of fixation techniques, including hemiarthoplasty (n = 9), plate-and-screw fixation (n = 18), and nails, pins, or rods (n = 6). There were no differences among the 3 groups on any of the demographic characteristics.

Outcomes

Repeated-measures ANOVAs showed that isometric lower-extremity force, 6-minute-walk distance, free gait speed, and self-reported physical function improved with time (Tab. 3). There was an interaction effect for isometric lower-extremity force that suggested that the 2 exercise groups improved isometric force production more than the control group. There was no group effect for any of the other dependent variables.

General linear modeling showed that the final outcome of 6-minute-walk distance (adjusted [R.sup.2] = 0.88, P = .000) was explained by baseline 6-minute-walk distance (P = -.000) and the interaction of treatment group and depressive scores (P = .011). Isometric lower-extremity force (adjusted [R.sup.2] = 0.87, P = .000) was explained by baseline force values (P = .000). Gait speed (adjusted [R.sup.2] = 0.82, P = .000) was explained by baseline speed (P = .000), the interaction of depression and treatment group (P = .012). For both interactions, people in the control group had more depressive symptoms than those in the other groups. No individual factors were related to physical function (adjusted [R.sup.2] = 0.28, P = .029).

Feasibility

Adherence to exercise training was determined by number of sessions attended divided by total number of possible sessions (n = 20). Adherence was 98% and did not differ in the resistance training or aerobic training groups. The percentage of sessions that subjects were able to achieve target intensity without reporting muscle soreness or shortness of breath was determined as the number of sessions at the target intensity divided by the total number of sessions. Ninety-five percent of the sessions were conducted at the target intensity. Ninety-six percent of treatments were provided routinely and were not altered because of non-muscular-type pains. In the 4% of the sessions that were altered because of pain, the reasons provided were hernia pressure, back pain, and knee joint pain.

One subject fell during the post-training examination. The subject did not require medical attention and was able to continue with testing without ill effects. Several subjects in the resistance training group reported muscle soreness or "fatigue" after exercise, but more frequently subjects in that group reported that their muscles "felt alive again."

Table 1 shows the training loads for the intervention groups. At the end of the "overload phase," subjects in the resistance training group were performing unilateral leg press exercises against 96 lb (43.2 kg) with the fractured leg. The hip abductors of the fractured side were contracting against loads of approximately 12 lb (5.4 kg) and the hip extensors against loads greater than 50 lb (22.5 kg). The aerobic training group was able to achieve and maintain 20 minutes of continuous aerobic exercise at an intensity of 65% to 75% of their age-predicted maximal heart rate through a combination of indoor and outdoor walking (100% of subjects) and, less frequently, stair climbing. Active ROM exercises were used with 50% of the subjects, but not as the sole mode of aerobic training (ie, walking was the primary form of exercise each session).

Discussion

This study demonstrated that it is feasible to complete a supervised 12-week program of moderate- or high-intensity exercise in the home. Training appeared to be safe; that is, it was not associated with adverse effects. Adherence rates were very high.

We found that isometric force of the involved limb improved over time in all groups but to a greater extent in the intervention groups than in the control group. Improvement in isometric force appeared to occur with both resistance and aerobic training regimens. The increases in isometric force production and gait speed reported by Sherrington and Lord (14) after a program of step-up exercises are consistent with the increases demonstrated in our study. In contrast, Tinetti and colleagues (15) did not report improvements in force production after 6 months of training. Training intensity, in the study by Tinetti et al, was not individualized; all patients began with the elastic band of least resistance, used the same band for ankle muscles and knee muscles, were encouraged to train daily, and could progress no further than a band that provided 1.5 to 12 lb (0.675-5.4 kg) of resistance (depending on the length of the band). This intensity does not suggest that the patients achieved overload with maximal or near-maximal resistance.

The sample of patients in this study deserves attention. Although the sample appears similar to that in other studies reporting on home-dwelling elderly people after hip fracture, (4,16) the means of both demographic and performance data are somewhat misleading in our study because of the heterogeneity of the sample. There were large ranges and standard deviations in both demographic characteristics and performance variables. Ages ranged from 64 to 93 years; number of medications ranged from 1 to 11, depressive symptoms ranged from none to severe, mental status scores ranged from 20 to 30, and IADL scores ranged from 0 to 8. Before intervention, 6-minute-walk distances ranged from 63 to 472 m, isometric lower-extremity force ranged from 25 to 128 kg, free gait speed ranged from 0.13 to 1.34 m/s, and SF-36 physical function subscale scores ranged from 0 to 90. With such extensive ranges in a relatively small sample, it is evident that differences between the groups are difficult to detect without very large samples. Indeed, this preliminary study was not powered to detect between-group differences. Thus, although the heterogeneity of the sample suggests that the methods could be used with a large range of elderly people, the within-group differences may have contributed to the lack of statistical differences between groups.

Because all of our subjects increased their walking speed, the intervention cannot be considered the single factor that contributed to the improvement in this sample. Natural recovery may be a possible explanation for changes in the control group. Natural recovery of gait shows a rapid improvement in the first 6 months and a gradual, but insignificant, change between 6 and 12 months. (4,50) The control subjects were measured, on average, at 3 months and 5 months after fracture, during the time when the greatest change in gait speed is expected. The intervention groups, however, were measured, on average, at 5 and 8 months after fracture, at times when improvement is not as dramatic. Recording gait speed during the natural recovery phase complicates the interpretation of changes because there is no literature to indicate the contribution of exercise versus natural recovery on the restoration of walking ability. Gait speed, however, is considered to be an essential component for community participation. Crossing a street within the time frame of a traffic light or getting to the bathroom in a timely fashion have obvious meaningfulness to most elderly people. The subjects in Sherrington and Lord's study (14) (8 months after fracture) walked at a speed of 0.51 m/s (an increase of 0.05 m/s) after 1 month of training. Tinetti and colleagues' subjects who received physical therapy walked at 0.44 m/s 6 months after fracture and did not improve at 12 months. (16) We know of no home-based exercise programs for elderly people who are frail or disabled that reported an increase in gait speed following intervention.

The interaction of depressive symptoms and treatment group explained part of the variability for 6-minute-walk distance and gait speed outcomes. Interestingly, isometric force production was not explained by depressive symptoms. The strong verbal encouragement provided during each muscle contraction may have had some positive effect. In contrast, gait speed was measured without verbal encouragement, and the once-a-minute standardized encouragement provided during the 6-minute-walk test is not comparable to the maximal verbal encouragement during force testing. Whether feedback affected performance is not known. The mean GDS scores reported by Binder et al (51) and Hauer et al (52) suggest that their samples had fewer depressive symptoms than our sample. However, none of the exercise trials of patients with hip fractures have examined the influence of depression on exercise outcomes. (4,16,51,52) The results of this study suggest that depressive symptoms should be considered when examining walking as an outcome in future exercise trials.

There are several limitations to this study. One limitation is the possibility of selection bias. Patients who volunteer for exercise studies may believe that exercise will help them. This potential bias may have been further complicated by the nature of the testing. All subjects went through 2 hours of physical performance tests in which they were encouraged by a physical therapist to "push as hard as you can," "walk for 6 minutes and cover as much distance as you can," and so forth. Interestingly, in the intervention study with the largest number of subjects to date, the fractured leg was not tested for force production, and there were no measures of walking endurance or stair climbing. (20,21) Our subjects were given positive verbal feedback after completing a task. Many of the subjects commented at the end of the test that they "never knew they could do all that work." Others thanked the tester for "helping so much." The effect of the testing session on performance is not known. Although the subjects' statements are anecdotal, if they believed that exercise could help and demonstrated under supervision of a physical therapist that they were capable of performing more than they thought they could do, it is not unreasonable to assume that this "education" persuaded them to attempt exercise or to increase activity. We asked subjects to refrain from starting new exercise programs until the retest, but we could not control this variable.

Other limitations were the different levels of attention among groups and the time between assessments for the intervention groups versus the control group. The control group did not receive biweekly visits like the intervention groups. The intervention groups had 4 additional weeks of intervention compared with the control group. Despite these disparities, all groups improved.

We have demonstrated that a frail, home-dwelling sample of elderly people who have had a fracture can tolerate a moderate- to high-intensity home exercise program with appropriate supervision. Future studies with larger samples and more strict exclusion criteria are needed.

Conclusion

This study describes how high- and moderate-intensity exercise can be performed at home for elderly people with a hip fracture. The interventions provided applied knowledge regarding exercise prescription to patients with hip fracture in the home setting and were designed to improve muscle force, endurance, and gait and to reduce disability. The exercise did not appear to produce adverse events, and adherence to training was excellent. The study did not have sufficient power to draw conclusions about the effectiveness of the intervention. All groups improved in distance walked, force production, and free gait speed, although the improvement in force production was greater for the intervention groups than for the control group. Our data also suggest that depressive symptoms interacted with treatment group in explaining improvements in 6-minute-walk distance and gait speed.

This study was funded by a Foundation for Physical Therapy Research Grant, 2000.

This article was submitted January 30, 2004, and was accepted December 23, 2004.

* Health-O-Meter Professional Products, Pelstar LLC, 7400 W 100th Pl, Bridgeview, IL 60455.

([dagger]) Chatillon Force Measurement Systems, Ametek TCI Division, 8600 Somerset Dr, Largo, FL 33773.

([double dagger]) EQ Inc, PO Box 16, Chalfont, PA 18914-0016.

([section]) Contemporary Design Co, PO Box 5089, Glacier, WA 98244.

([paragraph]) Polar Electro Oy, Professorintie 5, 90440 Kempele, Finland.

(#) SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.

References

(1) National Center for Health Statistics. Health, United States, 1999: With Health and Aging Chartbook. Hyattsville, Md: National Center for Health Statistics, 1999.

(2) US Congress, Office of Technology Assessment. Hip Fracture Outcomes in People Age Fifty and Over [background paper]. Washington, DC: Government Printing Office; 1994. OTA-BP-H-120.

(3) Mossey JM, Mutran E, Knott K, Craik R. Determinants of recovery 12 months after hip fracture: the importance of psychosocial factors. Am J Public Health. 1989;72:279-286.

(4) Magaziner J, Hawkes W, Hebel JR, et al. Recovery from hip fracture in eight areas of function. J Gerontol A Biol Sci Med Sci. 2000;55: M498-M507.

(5) Marottoli RA, Berkman LF, Cooney LM Jr. Decline in physical function following hip fracture. J Am Geriatr Soc. 1992;40:861-866.

(6) Zuckerman JD, Koval KJ, Aharonoff GB, Skovron ML. A functional recovery score for elderly hip fracture patients, II: validity and reliability. J Orthop Trauma. 2000;14:26-30.

(7) Jacobsen SJ, Goldberg J, Miles TP, et al. Hip fracture incidence among the old and very old: a population-based study of 745,435 cases. Am J Public Health. 1990;80:871-873.

(8) Tolo ET, Bostrom MPG, Simic PM, et al. The short term outcome of elderly patients with hip fractures. Int Orthop. 1999;23:279-282.

(9) Norton R, Butler M, Robinson E, et al. Declines in physical functioning attributable to hip fracture among older people: a follow-up study of case-control participants. Disabil Rehabil. 2000;22:345-351.

(10) Buchner DM, Cress ME, de Lateur BJ, et al. The effect of strength and endurance training on gait, balance, fall risk, and health services use in community-living older adults. J Gerontol A Biol Sci Med Sci. 1997;52:M218-M224.

(11) Fiatarone MA, O'Neill EF, Ryan ND, et al. Exercise training and nutritional supplementation for physical frailty in very elderly people. N Engl J Med. 1994;330:1769-1775.

(12) Foster VL, Hume GJ, Byrnes WC, et al. Endurance training for elderly women: moderate vs low intensity. J Gerontol A Biol Sci Med Sci. 1989;44:M184-M188.

(13) US Dept of Health and Human Services. Healthy People 2010: Understanding and Improving Health. 2nd ed. Vol 1. Available at: http://www.healthypeople.gov/ document/pdf/Volume1/06Disability.pdf. Accessed February 14, 2005.

(14) Sherrington C, Lord SR. Home exercise to improve strength and walking velocity after hip fracture: a randomized controlled trial. Arch Phys Med Rehabil. 1997;78:208-212.

(15) Tinetti ME, Baker DI, Gottschalk M, et al. Systematic home-based physical and functional therapy for older persons after hip fracture. Arch Phys Med Rehabil. 1997;78:1237-1247.

(16) Tinetti ME, Baker DI, Gottschalk M, et al. Home-based multicomponent rehabilitation program for older persons after hip fracture: a randomized trial. Arch Phys Med Rehabil. 1999;80:916-922.

(17) Nichols JF, Hitzelberger LM, Sherman JG, Patterson P. Effects of resistance training on muscular strength and functional abilities of community-dwelling older adults. Journal of Aging and Physical Activity. 1995;3:238-250.

(18) Latham NK, Anderson CS, Lee A, et al. A randomized, controlled trial of quadriceps resistance exercise and vitamin D in frail older people: the Frailty Interventions Trial in Elderly Subjects (FITNESS). JAm Geriatr Soc. 2003;51:291-299.

(19) Singh NA, Clements KM, Fiatarone MA. A randomized controlled trial of progressive resistance training in depressed elders. J Gerontol A Biol Sci Med Sci. 1997;52:M27-M35.

(20) Posner JD, Gorman KM, Windsor-Landsberg L, et al. Low to moderate intensity endurance training in healthy older adults: physiological responses after four months. JAm Geriatr Soc. 1992;40:1-7.

(21) Ford AB, Folmar SJ, Salmon RB, et al. Health and function in the old and very old. J Am Geriatr Soc. 1988;36:187-197.

(22) Mossey JM, Knott K, Craik R. The effects of persistent depressive symptoms on hip fracture recovery. J Gerontol. 1990;45:M163-M168.

(23) Scaf-Klomp W, van Sonderen E, Sanderman R, et al. Recovery of physical function after limb injuries in independent older people living at home. Age Ageing. 2001;30:213-219.

(24) Wells JL, Seabrook JA, Stolee P, et al. State of the art in geriatric rehabilitation, part II: clinical challenges. Arch Phys Med Rehabil. 2003;84:898-903.

(25) Folstein MF, Folstein SE, McHugh PR. "Mini-mental state": a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.

(26) Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J. 1965;14:61-65.

(27) Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9: 179-186.

(28) Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. 1982-1983;17:37-49.

(29) Fox KM, Hawkes WG, Magaziner J, et al. Markers of failure to thrive among older hip fracture patients. JAm Geriatr Soc. 1996;44:371-376.

(30) Brink TL, Yesavage JA, Lum D. Screening tests for geriatric depression. Clin Gerontol. 1982;1:37-42.

(31) Guyatt GH, Sullivan MJ, Thompson PJ, et al. The 6-minute walk: a new measure of exercise capacity in patients with chronic heart failure. Can Med Assoc J. 1985;132:919-922.

(32) Harada ND, Chiu V, Stewart AL. Mobility-related function in older adults: assessment with a 6-minute walk test. Arch Phys Med Rehabil. 1999;80:837-841.

(33) Andrews AW, Thomas M, Bohannon RW. Normative values for isometric muscle force measurements obtained with hand-held dynamometers. Phys Ther. 1996;76:248-259.

(34) Bohannon RW. Reference values for extremity muscle strength obtained by hand-held dynamometry from adults ages 20 to 79 years. Arch Phys Med Rehabil. 1997;78:26-32.

(35) Buchner DM, Cress ME, de Lateur BJ, et al. A comparison of the effects of three types of endurance training on balance and other fall risk factors in older adults. Aging (Milano). 1997;9:112-119.

(36) Leiper CI, Craik RL. Relationships between physical activity and temporal-distance characteristics of walking in elderly women. Phys Ther. 1991;71:791-803.

(37) Ware JE Jr, Sherbourne CD. The MOS 36-Item Short-Form Health Survey (SF-36), I: conceptual framework and item selection. Med Care. 1992;30:473-483.

(38) Ware JE Jr, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey: Manual and Interpretation Guide. Boston, Mass: Nimrod Press: 1993.

(39) McHorney CA, Ware JE Jr, Lu JF, Sherbourne CD. The MOS 36-Item Short-Form Health Survey (SF-36), III: tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care. 1994;32:40-66.

(40) Chandler JM, Duncan PW, Kochersberger G, Studenski S. Is lower extremity strength gain associated with improvement in physical performance and disability in frail, community-dwelling elders? Arch Phys Med Rehabil. 1998;79:24-30.

(41) Knutson LM, Soderberg GL. EMG: use and interpretation in gait. In: Craik RL, Oatis CA, eds. Gait Analysis: Theory and Application. St Louis, Mo: Mosby; 1995:307-325.

(42) Berger RA. Optimum repetitions for the development of strength. Res Q. 1962;33:334-338.

(43) Pollock ML, Carroll JF, Graves JE, et al. Injuries and adherence to walk/jog and resistance training programs in the elderly. Med Sci Sports Exerc. 1991;23:1194-2000.

(44) Braith RW, Graves JE, Leggett SH, Pollock ML. Effect of training on the relationship between maximal and submaximal strength. Med Sci Sports Exerc. 1993;25:132-138.

(45) Fiatarone MA, Marks EC, Ryan ND, et al. High-intensity strength training in nonagenarians: effects on skeletal muscle. JAMA. 1990;263: 3029-3034.

(46) Smidt GL. Hip motion and related factors in walking. Phys Ther. 1971;51:9-22.

(47) Winter DA, Eng JJ, Ishac MG. A review of kinetic parameters in human walking. In: Craik RL, Oatis CR, eds. Gait Analysis: Theory and Application. St Louis, Mo: Mosby; 1995:252-270.

(48) Roitman JL, senior ed; American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription. 6th ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2000.

(49) Borg G. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982;14:377-381.

(50) Koval KJ, Skovron ML, Aharonoff GB, et al. Ambulatory ability after hip fracture: a prospective study in geriatric patients. Clin Orthop. 1995;310:150-159.

(51) Binder EF, Brown M, Sinacore DR, et al. Effects of extended outpatient rehabilitation after hip fracture. JAMA. 2004;292:837-846.

(52) Hauer K, Specht N, Schuler M, et al. Intensive physical training in geriatric patients after severe falls and hip surgery. Age Aging. 2002;31: 49-57.

KK Mangione, PT, PhD, GCS, is Associate Professor, Arcadia University, Department of Physical Therapy, Health Sciences Center, 450 S Easton Ave, Glenside, PA 19038 (USA) ([e-mail]). Please address all correspondence to Dr Mangione.

RL Craik, PT, PhD, FAPTA, is Professor and Chair, Department of Physical Therapy, Arcadia University.

SS Tomlinson, PT, DPT, is Assistant Professor and Academic Coordinator of Clinical Education, Department of Physical Therapy, Arcadia University.

KM Palombaro, PT, MS, is Research Associate, Department of Physical Therapy, Arcadia University, and a doctoral student at Temple University, Philadelphia, Pa.

Dr Mangione and Dr Craik provided concept/idea/research design, writing, data analysis, fund procurement, and facilities/equipment. Dr Mangione, Dr Tomlinson, and Ms Palombaro provided data collection and subjects. Dr Mangione and Dr Tomlinson provided project management.

The Committee on the Protection of Research Subjects at Arcadia University approved this study.

This research was presented as a poster presentation at the 56th Annual Scientific Meeting of the Gerontological Society of America; November 21-25, 2003; San Diego, Calif, and as a platform presentation at the Combined Sections Meeting of the American Physical Therapy Association; February 20-24, 2002; Boston, Mass.

Table 1.

Summary of Interventions (a)

                                         Resistance (lb)

                                         Session 1

Group                                    X       SD

Resistance training (b)

Supine hip and knee extension with PRE   74.0    22.0
  machine
Supine hip abduction with PRE machine    6.9     2.0
Standing hip extension with PRE          34.8    11.3
  machine

                                         Percentage of Subjects
                                         Performing Task
Aerobic training (c)

  Indoor walking                         100
  Outdoor walking                         83
  Stair climbing                          33
Control

                                         Session 16

Group                                    X       SD

Resistance training (b)

Supine hip and knee extension with PRE   96.4    19.7
  machine
Supine hip abduction with PRE machine    11.9     4.3
Standing hip extension with PRE          51.6    22.5
  machine

Aerobic training (c)

  Indoor walking
  Outdoor walking
  Stair climbing
Control

                                         Session 20

Group                                    X       SD

Resistance training (b)

Supine hip and knee extension with PRE   103.9   14.1
  machine
Supine hip abduction with PRE machine     11.6    1.1
Standing hip extension with PRE           57.0   22.2
  machine

Aerobic training (c)

  Indoor walking
  Outdoor walking
  Stair climbing
Control

Group                                    Frequency and Duration

Resistance training (b)                  2 sessions per week for 8 wk
                                           followed by 1 session per
                                           week for 4 wk
Supine hip and knee extension with PRE
  machine
Supine hip abduction with PRE machine
Standing hip extension with PRE
  machine

Aerobic training (c)                     22-23 min per session, 2
                                          sessions per week for 8 wk
                                          followed by 1 session
                                          per week for 4 wk
  Indoor walking
  Outdoor walking
  Stair climbing
Control                                  8 wk

(a) PRE-progressive resistive exercise.

(b) Intensity was set at 3 sets of 8 repetitions at the 8-repetition
maximum intensity. Resistance resistance.

(c) Intensity was set at 65%, to 75% of age-predicted maximum heart
rate or a rating of 3 to 5 for the aerobic training group was 91 to 105
bpin; actual heart rate range for the group was range of the sample.
The youngest person would have the highest target heart rate, and the
aerobic training task, the percentage of subjects who used that task as
a means of increasing training also included standing plantar flexion
with body weight as on the Borg Rating of Perceived Exertion Scale.
The target heart rate 95 to 104.5 bpm. The heart rate range provided
is based on age oldest person would have the lowest target heart rate.
For each heart rate is shown.

Table 2.

Demographic Characteristics of the Sample

                                          Aerobic Training
                                          Group     (n=12)

Sex (% women)                             75%
Race (% Caucasian)                        100%
Side of fracture (% right side)           50%
                                          [bar.X]   SD       Range

Age (y)                                      79.8    5.6     71-88
Body mass index (kg/m2)                      25.1    3.9   20.2-28.4
Weeks after surgery to start of study        19.7    8.4   10.5-40.5
Average no. of medications upon               4.1    2.5     1-10
  discharge from hospital
Barthel Index of Activities of Daily         96.8    9.0    70-100
Living Score (0-100)
Lawton Instrumental Activities of Daily       5.5    2.4      0-8
  Living Index score (0-8)
Folstein Mini-Mental Status Exam             27.8    2.8     20-30
  score (0-30)
Geriatric Depression Scale score              5.7    6.4     1-20
  (0-30)
                                          Resistance Training
                                          Group (n=11)

Sex (% women)                             64%
Race (% Caucasian)                        100%
Side of fracture (% right side)           64%

                                          [bar.X]   SD       Range

Age (y)                                      77.9    7.9     64-89
Body mass index (kg/m2)                      25.0    4.0   19.4-31.6
Weeks after surgery to start of study        19.4   11.7   10.0-50.5
Average no. of medications upon               4.7    2.8      1-9
  discharge from hospital
Barthel Index of Activities of Daily         95.0    5.9    85-100
Living Score (0-100)
Lawton Instrumental Activities of Daily       3.9    2.5      0-8
  Living Index score (0-8)
Folstein Mini-Mental Status Exam             27.3    3.2     21-30
  score (0-30)
Geriatric Depression Scale score              5.5    8.1     0-24
  (0-30)

                                          Control Group
                                          (n=10)

Sex (% women)                             80%
Race (% Caucasian)                        100%
Side of fracture (% right side)           50%

                                          [bar.X]   SD       Range

Age (y)                                      77.8    7.3     69-89
Body mass index (kg/m2)                      26.0    5.0   17.1-35.3
Weeks after surgery to start of study        12.6    2.3    7.0-16.0
Average no. of medications upon               4.9    3.1     2-11
  discharge from hospital
Barthel Index of Activities of Daily         92.0   10.3    70-100
Living Score (0-100)
Lawton Instrumental Activities of Daily       4.2    2.3      1-8
  Living Index score (0-8)
Folstein Mini-Mental Status Exam             28.0    2.0     25-30
  score (0-30)
Geriatric Depression Scale score             9.50   6.10     1-19
  (0-30)

Table 3.

Results of Repeated-Measures Analysis of Variance for Primary Outcome
Measures (a)

                                           Baseline

                                       [bar.X]    SD
6-minute-walk distance (m)
  Aerobic exercise group (n=12)         232.4    122.0
  Resistance exercise group (n=11)      197.1    104.2
  Control group (n=10)                  180.6    104.3
Maximal voluntary isometric force of
  the lower extremity (kg)
  Aerobic exercise group (n=12)         55.6     17.4
  Resistance exercise group (n=11)      48.5     12.6
  Control group (n=10)                  64.1     24.6
Free gait speed (m/s)
  Aerobic exercise group (n=12)          0.7      0.3
  Resistance exercise group (n =11)      0.5      0.3
  Control group (n=10)                   0.5      0.2
SF-36 physical function
  Aerobic exercise group (n=12)         51.7     20.8
  Resistance exercise group (n=11)      45.5     21.6
  Control group (n=10)                  44.1     24.9

                                        Posttraining

                                       [bar.X]   SD

6-minute-walk distance (m)
  Aerobic exercise group (n=12)         321.1    101.7
  Resistance exercise group (n=11)      278.9    114.6
  Control group (n=10)                  266.2     82.4
Maximal voluntary isometric force of
  the lower extremity (kg)
  Aerobic exercise group (n=12)          67.1     22.3
  Resistance exercise group (n=11)       59.6     18.2
  Control group (n=10)                   67.7     22.2
Free gait speed (m/s)
  Aerobic exercise group (n=12)           0.8      0.3
  Resistance exercise group (n =11)       0.7      0.3
  Control group (n=10)                    0.7      0.2
SF-36 physical function
  Aerobic exercise group (n=12)          57.5     24.3
  Resistance exercise group (n=11)       57.7     21.1
  Control group (n=10)                   48.0     18.9

                                        ANOVA(Time)

                                       F         P

6-minute-walk distance (m)             107.88    .000
  Aerobic exercise group (n=12)
  Resistance exercise group (n=11)
  Control group (n=10)
Maximal voluntary isometric force of    48.13    .000
  the lower extremity (kg)
  Aerobic exercise group (n=12)
  Resistance exercise group (n=11)
  Control group (n=10)
Free gait speed (m/s)                   56.52    .000
  Aerobic exercise group (n=12)
  Resistance exercise group (n =11)
  Control group (n=10)
SF-36 physical function                 4.17     .050
  Aerobic exercise group (n=12)
  Resistance exercise group (n=11)
  Control group (n=10)

                                       Interaction

                                       (Time x Group)

                                       F         P

6-minute-walk distance (m)               NS
  Aerobic exercise group (n=12)
  Resistance exercise group (n=11)
  Control group (n=10)
Maximal voluntary isometric force of    3.46      .04
  the lower extremity (kg)
  Aerobic exercise group (n=12)
  Resistance exercise group (n=11)
  Control group (n=10)
Free gait speed (m/s)                    NS
  Aerobic exercise group (n=12)
  Resistance exercise group (n =11)
  Control group (n=10)
SF-36 physical function                  NS
  Aerobic exercise group (n=12)
  Resistance exercise group (n=11)
  Control group (n=10)

(a) SF-36=Medical Outcomes Study 36-Item Short-Form Health Survey,
ANOVA=analysis of variance, NS=not significant.