Background and Purpose. Most patients with hip fracture do not return to prefracture functional status 1 year after surgery. The literature describing interventions, however, does not use classic overload and specificity principles. The purpose of this case report is to describe the use of resistance training to improve functional outcomes in a patient following hip fracture. Case Description. The patient was a 68-year-old woman who had a comminuted intertrochanteric fracture of the left hip 3 months previously. She used a cane for ambulation, and her walking was limited. The patient received 16 sessions of lower-extremity strengthening exercises, aerobic training on a stationary bicycle, functional training supervised by a physical therapist, and a home stretching program. Outcome. The patient's isometric muscle force for involved hip extension, hip abduction, and knee extension improved by 86%, 138%, and 33%, respectively; walking endurance increased by 22.5%; balance improved by 400%; balance confidence increased by 41%; and self-reported ability to perform lower-extremity functional activities increased by 20%. Discussion. The authors believe that some patients can perform comprehensive exercise programs after hip fracture and that properly designed programs can affect patient outcomes beyond observed impairments.

Hip fracture is a common medical problem that can reduce the quality of life for older adults. More than 300,000 people over the age of 50 years are expected to fracture a hip each year, at an estimated cost of $5 billion.1,2 The costs associated with care are high because hip fracture results in long-term disability for most people.3 Although the incidence rate for hip fracture has stabilized, the total number of hip fractures is predicted to increase because of growth in the older population.4

The improvements in medical management of patients with hip fracture (eg, postoperative mortality and infection rates) have not been matched by improvements in function.5 In a sample of 306 community-dwelling elderly people (aged 65 years and older), 68% had difficulty rising from a chair and 66% had unequal step length and step time 2 months after hip fracture.6 Another study of 120 people aged 65 years and older showed that, 6 months after hip fracture, only 8% could climb a flight of stairs, 15% could walk across a room independently, and 6% could walk a half mile.7 Tolo et al8 reported that, 8 months after fracture, 42% of a sample of elderly people 65 years to 100 years of age required a cane when they previously walked independently, the sample's walker use tripled, and 56% could not walk as well as they could prior to the fracture. Research has established that the majority of patients with hip fracture do not return to prefracture functional status 1 year after surgery.7,9 The ability to regain the prefracture level of function following hip fracture has been characterized as a function of the patient's prefracture functional status,10 factors such as comorbidities and mental status,1011 the type of medical setting,12 and the type of intervention.13

Therapeutic exercise is the least-examined factor affecting outcome in patients with hip fracture.14 General categories of physical therapy interventions include active-assistive, active, and resistance exercises; transfer and ambulation training; instructions on weight-bearing limitations and precautions; and use of moist heat.15,16 However, complete exercise prescriptions—including descriptions of the frequency, intensity, and duration of exercise programs—are sparse. Binder and colleagues17 have recently reported the results of an exercise trial in patients after hip fracture in which frequency, intensity, and durations were reported. Binder et al followed the classical principle of overload. The overload principle suggests that exercise should be performed at an intensity level higher than “normal” to facilitate physiological adaptations (neural recruitment or hypertrophy) that result in a training response.18,19 The other principle of exercise is specificity, in which adaptations in the metabolic and physiologic systems depend on the type of overload imposed on the system (ie, the predominant energy system or the movement pattern and specific muscle groups exercised).18 We could not find descriptions of interventions that indicated that classic overload and specificity principles were used for patients following hip fracture. Some of the residual disability after hip fracture, however, might be addressed by applying standard principles of therapeutic exercise. The purpose of this case report is to describe the application of therapeutic exercise principles with a woman following hip fracture.

Case Description

Patient Description

The patient was a 68-year old-woman who sustained a comminuted intertrochanteric fracture of the left hip when she slipped and fell on an icy sidewalk. Her past medical history included a fall 10 years previously that resulted in a right intertrochanteric fracture, which was reduced with a compression screw. Physical therapist management following her first hip fracture consisted of intervention in the acute care setting for bed transfers and gait and stair training with a walker. She then had physical therapy in her home twice a week for 8 weeks. This intervention consisted of quadriceps femoris muscle sets; heel-slides; supine hip abduction; leg massage; long-arc quadriceps femoris muscle exercises with weights; standing hip abduction, extension, and flexion; hamstring muscle curls; and heel-raises. After 4 weeks, the patient progressed to the use of axillary crutches on level surfaces and stairs. She walked outdoors using axillary crutches during her last week of intervention in the home. Other past medical history included an episode of supraventricular tachycardia. The event, which occurred 1 year before the second hip fracture, was medically evaluated and found to be self-limited.

She reported taking no medications and needing no assistance in any basic activities of daily living (ADL) or instrumental activities of daily living (IADL). The patient lived alone in an apartment. She worked part-time at a university in the billing department. Before the most recent hip fracture, the patient was active; she enjoyed walking and jogging and assisted in the care of her young grandchildren.

The second hip fracture was surgically repaired with open reduction and internal fixation using an intramedullary rod and a gamma nail. Physical therapy interventions included 4 days of physical therapy in the acute care setting, 9 days of inpatient rehabilitation, and 4 home care visits over 4 weeks. Physical therapy in the acute care setting focused on bed and toilet transfers and gait training with a rolling walker. The inpatient rehabilitation consisted of hip active range of motion exercise, hamstring muscle curls, and heel-raises when standing in the parallel bars; supine hip abduction; isometric quadriceps femoris muscle sets; heel-slides; sitting long-arc quadriceps femoris muscle exercises with weights; sitting balance on a therapy ball; stair training; and outdoor ambulation. The 4 home care physical therapy visits consisted of reinforcement of the previously assigned standing and supine exercises (30 repetitions of each exercise were performed) and progression to the use of a single-point cane. On the final home care visit, the patient practiced supervised outdoor ambulation with the single-point cane. At the end of her 4 weeks of home care, the patient was instructed to perform a home exercise program that consisted of the following supine exercises: isometric quadriceps femoris and gluteus maximus muscle sets, heel-slides, hip abduction, and straight leg raises. The exercise program also included the following standing exercises: hip abduction, hip extension, marching, mini-squats, and heel-raises. She reported performing 30 repetitions of all exercises daily.

The patient was discharged from physical therapy 6 weeks after the fracture. She returned to her part-time job 2 months after the fracture and resumed driving. She used a cane, except in her home. The patient approached the physical therapy faculty at the university at which she worked approximately 2½ months after the fracture. She said that she was not satisfied with her recovery and wanted to know whether she could obtain additional physical therapy. She reported that that she was unable to walk long distances; that she was unable to perform leisure activities such as shopping because of lower-extremity and body fatigue; that she was fearful of long-distance outdoor ambulation and floor-to-stand transfers; and that her left lower extremity was shorter than her right lower extremity, which impaired her balance and walking ability. Her impaired walking ability and her fear of engaging in activity limited her return to her previous functional level. The patient provided informed consent for this case report. We obtained a copy of her medical record and a referral for physical therapy from her orthopedic surgeon and agreed to provide pro bono physical therapy services at the university.

Although the patient was younger than most people who sustain a hip fracture, people between the ages of 65 and 74 years comprise 20% of the patients with hip fractures.3 Her case, therefore, is not rare. We thought that her past history of a hip fracture, her current fracture, and the fact that she had not returned to her previous level of function made her more “typical” despite the age discrepancy. We thought she would be a good candidate to determine whether additional intervention could improve her functional status and activity level and reduce her impairments and disabilities. Despite her fears and limitations, she said she was motivated to improve her current level of function.

Examination, Evaluation, Diagnosis, and Prognosis

A physical therapist examination was conducted after taking the patient's medical history. We selected tests and measures based on expected problems associated with hip surgery and her existing problems (inability to walk long distances, lower-extremity and body fatigue, fear of performing high-level activities, balance problems, and leg-length discrepancy). The same physical therapist, who had more than 7 years of experience working with older adults, performed all tests and measures with the exception of isometric muscle force production. Isometric muscle force production was tested by a different physical therapist with more than 18 years of experience using handheld dynamometry and electromechanical dynamometry. We did not estimate the reliability of our measurements. These 2 physical therapists performed all interventions for the patient.

Walking endurance was assessed in response to the patient's reports of inability to walk long distances as well as to the patient's fear, which limited engagement in activities that required good aerobic condition. The patient's heart rate and blood pressure were measured before and after testing to ensure appropriate response to activity. Walking endurance was assessed using the 6-minute walk test. The patient walked along a 30.48-m, linoleum floor and was instructed to cover as much distance as possible in 6 minutes. Standardized, strong, verbal encouragement was given every minute (“You're doing great. Keep going.”). The patient performed 2 trials of the test, with a 5-minute rest period between trials, and the second test result was used. The distance of the second walk was 348.39 m. Intraclass correlation coefficients (ICCs) for intrarater, interrater, and test-retest reliability of data obtained with this test are reported to be .90.20 Concurrent validity has been established with measures of aerobic capacity (peak oxygen consumption and maximal metabolic equivalents).21

Gait was assessed visually for deviations and with the GaitMat II* system for gait speed, single-leg support time, and symmetry. The GaitMat II consists of pressure-sensitive switches in a 4-m walkway and a computer system that runs the program and analyzes the data. The switches open and close when the patient walks on the walkway. Gait speed is calculated by the computer, which divides the time between the first and last switch closures by the distance traveled. The ICCs for reliability of data obtained with the GaitMat II have been reported to range from .90 to .99 for older women walking at a variety of speeds.22 Gait speed measurements have been shown to have predictive validity for decline in ability to perform basic ADL and IADLs.23

The patient completed walking trials at 2 different speeds. A trial consisted of walking over the mat in one direction. The first speed tested was 2 trials at free speed, in which the patient was instructed to “walk at your normal or comfortable pace.” The last test was 2 trials of fast speed, in which she was asked to “walk as quickly as possible without running.” Fast gait speed may be an important measure of a person's ability to walk quickly to cross the street within the time frame of a traffic light24 and has been used to identify characteristics of frailty in elderly people.25 The patient's average free gait speed was 0.84 m/s with the cane and 0.66 m/s without the cane. These values are below the normal values for elderly people of 1.0 to 1.2 m/s.22 Her fast speed was 1.19 m/s. A left Trendelenburg gait pattern26 was apparent during visual gait analysis. Additional gait speed values are shown in Table 1.

Table 1.

Gait Speed and Balance Performance in a 68-Year-Old Woman After Hip Fracture

Fall risk and static and dynamic balance were assessed. The Functional Reach Test was used as a measure of fall risk.27 The patient's functional reach was 28 cm, which suggested that she was not at a high risk for falls (<15 cm=fall risk).28 The ICCs for interrater reliability for the Functional Reach Test have been estimated to be .79 to .98 and for test-retest reliability to be .92.27 The Functional Reach Test has been shown to have predictive validity for falls.28 The Timed “Up & Go” (TUG) Test time was 16.3 seconds, with less than 20 seconds suggesting independent mobility.29 Intrarater and interrater reliability (r) were estimated to be greater than .93,29 and the ICC for test-retest reliability is reported to be .56.30 The TUG Test has good criterion validity with the Berg Balance Scale, gait speed, and the Barthel Index of Activities of Daily Living.29 Semi-tandem and tandem standing balance were assessed. Initial times for semi-tandem and tandem stands were 60.0 seconds and 12.0 seconds, respectively. The Women's Health and Aging Study showed that, of 388 women aged 65 to 74 years, 76.1% were able to sustain a semi-tandem standing position for 10 seconds and 39.7% were able to sustain a tandem standing position for 10 seconds.31 Interrater and test-retest reliability (r) of data for these tests were estimated to be .99 and .97, respectively.32

Range of motion of the lower extremities was measured using a universal goniometer, as described by Kendall et al.33 Limitations in the bilateral straight leg raise were noted. Left straight leg raise was measured as 0 to 40 degrees, and right straight leg raise was 0 to 65 degrees. No other limitations in range of motion were found. Intrarater reliability (r) for universal goniometry of the lower extremity has been reported to be .80.34

We screened lower-extremity muscle performance using manual muscle tests (MMT), and then we measured isometric muscle force production with the Chattilon handheld dynamometer (CSD 500) and the KinCom II dynamometer.§ The handheld dynamometer was used to quantify hip muscle force scores. Because of the difficulty stabilizing the patient for measuring knee extension force, we used the KinCom II dynamometer. Manual muscle tests of the lower extremities were performed using the technique and positions described by Kendall et al.33 Deficits (grades <4/5) were found in bilateral hip extension and left knee extension. Interrater reliability (r) of MMT grades for the hip and knee flexors was estimated to be .74 and .63, respectively.35 Kappa values for interrater reliability for the gluteus medius muscle measurement were estimated to be .30 to .42.36 The patient was positioned supine for testing of the hip extensors and hip abductors with the Chattilon handheld dynamometer.37 She was asked to push as hard as possible while the physical therapist matched the resistance (make test). The patient performed 2 practice trials. After a 1-minute rest, she performed 2 maximal-effort trials for 5 seconds each, with a 1-minute rest between trials. The peak force was recorded for each of the 2 trials, and the mean of the trials was reported.

To measure the force of the hip extensors, her hip was passively flexed to 90 degrees with the knee relaxed. The trunk was stabilized to the plinth with a belt across the pelvis. Resistance was applied just proximal to the knee on the posterior surface of the leg.37 The hip abductors were tested with the knee straight and the hip in a neutral position. The stabilization was the same as for the hip extensors, and resistance was applied at the lateral femoral condyles.37 For knee extension, the patient was seated on the KinCom II dynamometer with the hip flexed to 90 degrees and the knee joint stabilized at 75 degrees of flexion.38 The axis of the lever arm of the dynamometer was aligned with the patient's lateral femoral condyle. A shin pad attached to the lever arm was secured to the patient approximately 2 cm proximal to the lateral malleolus on the anterior aspect of the leg. Straps across the thigh and trunk also were used for stabilizing the patient. Handheld dynamometry and KinCom results are shown in Table 2. Test-retest reliability (ICC) for handheld dynamometry has been estimated to be .93 to .98,37 and intrarater reliability (r) has been estimated to be .69 to .90.35 Test-retest reliability (ICC) for peak knee extensor torque was estimated to be .89 to .98.39

Table 2.

Muscle Performance (in Newtons) for a 68-Year-Old Woman After Hip Fracturea

Leg length was assessed because the patient reported that her left lower extremity was shorter than her right lower extremity. Leg-length measurement revealed that the left lower extremity was 2.4 cm shorter than the right lower extremity. We used the indirect method of leg-length measurement as described by Hanada et al.40 The patient's iliac crests were palpated and the left lower extremity was elevated by placing a number of books of measurable width under the left foot. Books were placed under the foot until the pelvis was level. Hanada et al40 reported ICCs for intrarater and interrater reliability greater than .90. Concurrent validity (ICC) with radiographic imaging was reported to be .76.40

The patient's straight cane was inspected to ensure that the cane tip was in good condition. Proper cane height was assessed using the right greater trochanter as the bony landmark for cane height. The cane was both in good condition and the proper height.

Pain was assessed using a visual analog scale (VAS). The VAS was a 10-cm horizontal line with a hash mark at 1-cm intervals (0 cm indicates “no pain” and 10 cm indicates “pain as bad as it could possibly be”). The patient was instructed to place a hash mark that corresponded to her present level of pain, and the distance (in centimeters) from “no pain” was measured and corresponded to the patient's pain level. Test-retest reliability has been reported to be (κ) .66 to .9341 and (r) .82.42 The patient rated her pain rated as 3.5/10 in the left lower extremity.

The Activities-specific Balance Confidence (ABC) Scale was used to assess the patient's reported fear of falling.43 The ABC Scale is a 16-item assessment tool that asks the patient to write a percentage indicating confidence performing a specific activity. Higher percentages are associated with greater balance confidence. The patient scored 33.75% at initial testing. Test-retest reliability (r) for the ABC Scale was estimated to be .92.43 Internal consistency of this scale (Cronbach α) is reported to be .96.43 The ABC Scale has concurrent validity with the Physical Self-Efficacy Scale and its physical abilities subscale (r=.49 and r=.63, respectively) and concurrent validity with scores on the Falls Efficacy Scale (r=−.84).43,44 The ABC Scale can discriminate between people with high and low mobility.44

Global outcome measures used were the Lower Extremity Functional Scale (LEFS) and the Medical Outcomes Study 36-Item Short-Form Health Survey questionnaire (SF-36). The LEFS assesses difficulty with functional activities that may affect basic ADL and IADL ability. The LEFS is a 20-item assessment tool, with a maximum score of 80 points. Higher scores are associated with better lower-extremity function. The patient scored 41 points. Test-retest reliability (R) was estimated to be .86 to .94.45 The LEFS has demonstrated construct validity and is correlated with the SF-36 physical function subscale and physical component summary scores (r=.80 and r=.64, respectively) and with the SF-36 mental component summary score (r=.30).45 The SF-36 was used to assess 3 health concepts comprising physical health status: physical functioning, role–physical, and bodily pain. Each subscale is scored on a 0-to-100 scale, with 100 representing excellent health status.46 The patient scored a score of 65 on the physical function subscale, a score of 25 on the rol–physical subscale, and a score of 84 on the bodily pain subscale. The physical health subscales of the SF-36 have reported internal consistency (Cronbach α=.79–.93) and test-retest reliability (r=.79–.91) on a wide range of patient and nonpatient samples.47

Based on the results of the examination, the diagnostic classification based on the Guide to Physical Therapist Practice48 was “Impaired Joint Mobility, Motor Function, Muscle Performance, and Range of Motion Associated With Joint Arthroplasty.” The impairments that we found that were consistent with this diagnostic classification were impaired muscle performance, range of motion, walking endurance, gait, and balance. There is evidence that these impairments can be improved with exercise training.4951 We believed that the patient would improve more completely and more quickly with an exercise program that adhered to the principles of overload and specificity. Because she was independent and active before this fracture, because she had an unremarkable medical history, because she had apparent motivation for recovery, and because her impairments were expected to be improved by exercise training, we expected that the patient had a good prognosis to return to her prior level of exercise and activity level in 2 months.


Physical therapy interventions were targeted toward improving the identified impairments. Therapeutic exercise was the major focus of the intervention and targeted lower-extremity muscle force, walking endurance, functional training, and flexibility training. The intervention was conducted twice a week for 8 weeks. The patient also was instructed to perform daily hamstring muscle stretching independently to address her range-of-motion limitation during the straight leg raise.

Muscle strengthening.

The 8-repetition maximum (8-RM) was used as the initial point of overload for the strengthening intervention. The 8-repetition maximum was defined as the amount of resistance a person can push against for a maximum of 8 repetitions in good form.18 The patient performed the following 3 strengthening exercises: combined hip and knee extension in a supine position (leg press), hip abduction in a supine position, and standing hip extension.

The 3 exercises were performed using a portable progressive resistive exercise machine, the Shuttle MiniClinic.# The MiniClinic consists of a metal frame; 6 resistive latex bands, spanning the length of the machine; and a footplate attached to a movable carriage that the patient can push against. Engaging the latex bands provides resistance; each band that is engaged increases the resistance of the machine. The bands are precalibrated to ensure that an increase in length of the band is associated with an increase in tension (in pounds). Each black latex band's load ranges from approximately 31 N (7 lb) to approximately 89 N (20 lb) of force; each red latex band's load ranges from approximately 9 N (2 lb) to approximately 40 N (9 lb) of force. The metal frame has a numbered progress monitor strip along its length. The numbers indicate the force being applied as the carriage is moved.

Each session consisted of the patient performing 3 sets of 8 repetitions of each exercise. For the supine leg press, the MiniClinic was placed on a mat table so that her foot rested on the footplate, which was positioned at a 60-degree angle. The patient's hip was flexed to 90 degrees, and she pushed her leg into full hip and knee extension against the predetermined resistance. This activity mimics lifting body weight from a lower surface (getting out of bed, rising from an armless chair, getting off the toilet) in which the lower extremity starts in a position of hip and knee flexion and then moves into hip and knee extension.52

For the hip abductors, the patient was positioned supine with a pillow under her buttocks to ensure that her hip was level with her foot. The MiniClinic was positioned at her feet, perpendicular to her legs. The footplate was flattened, and the patient's foot was strapped to the footplate, enabling the footplate to move with the patient's leg into abduction. The patient started in 5 degrees of adduction and moved 10 degrees into abduction. Fifteen degrees of movement was chosen because it approximates the 8 degrees of motion associated with gait and takes into account the variations in hip positions in a standing posture.53 Hip extensors also were trained with the patient standing. The MiniClinic was positioned behind the patient, with her foot resting on the footplate (knee flexed approximately 60°). The hip was flexed approximately 35 degrees, and she extended to neutral position. This range of motion approximates the time in the gait cycle when the gluteus maximus muscle shows the highest electromyographic activity (ie, from heel contact through 20% of stride).54 For hip extension, she held on to a walker for balance.

The 8-RM for each exercise was reassessed every 2 weeks at sessions 5, 9, and 13, and the number of cords was increased as appropriate. The initial assessment and subsequent reassessments of 8-RM were performed to ascertain a starting point and an appropriate progression for the strengthening intervention. Adjustments to the resistance level also were made in response to the patient's reports of pain. During the first session, she was strong enough to push against 4 black cords for the leg press; however, this caused hip and thigh pain during the exercise. Consequently, the next lowest resistance, 3 black cords, were used as a starting point for 8-RM. During session 8, the patient reported pain (rated as 8/10) after the previous session in the right posterior knee with use. She said she had no pain with rest. We suspected muscular pain, and we decreased the resistance level from 6 black cords to 5 black cords for the leg press during session 8. A scheduling conflict and a flat tire prevented the patient from participating during week 8, and she completed sessions 15 and 16 during week 9. Figures 1 through 3 show the approximate increase in resistance for each exercise throughout the 8 weeks of intervention.

Figure 1.

Changes in leg press 8-repetition maximum (8-RM), measured in approximate newtons of force for both the involved and uninvolved lower extremities (LEs). Four assessments of 8-RM were performed: at the initial session and at sessions 5, 9, and 13.

Figure 2.

Changes in hip abduction 8-repetition maximum (8-RM), measured in approximate newtons of force for both the involved and uninvolved lower extremities (LEs). Four assessments of 8-RM were performed: at the initial session and at sessions 5, 9, and 13.

Figure 3.

Changes in hip extension 8-repetition maximum (8-RM), measured in approximate newtons of force for both the involved and uninvolved lower extremities (LEs). Four assessments of 8-RM were performed: at the initial session and at sessions 5, 9, and 13.

Endurance retraining.

The patient performed aerobic training on a Monark 817 stationary bicycle** after the strength training exercises. The bicycle was chosen as the mode for exercise because we believed it would be difficult and perhaps painful for her to walk at a speed that would induce cardiovascular changes due to her gait deviations. Because the patient did not have any overt cardiac disease and had clearance from her physician to return to exercise, we determined her training heart rate to be 70% to 80% of her age-predicted maximum heart rate (106–122 bpm).18 The cycling sessions began at the end of the second week of the strengthening intervention (session 4). The patient wore a Polar heart rate monitor†† during cycling. Blood pressure and pulse were monitored for the first few sessions, to ensure appropriate response to aerobic exercise. The load was originally set at a resistance of 3, and the patient cycled at a rate of 90 rpm. After the fourth session, she reported generalized hip and thigh pain. We lowered the seat one notch to a seat height of 4 and increased the load on the cycle to a resistance of 5, and she pedaled at rate of 70 to 80 rpm for 20 minutes for each of the remaining sessions. A record of her cycling sessions was kept, including heart rate, perceived exertion, pedaling rate, and resistance level, until the end of the strengthening intervention.

Functional training.

During the sixth session, the patient was taught a floor-to-standing transfer. She was shown how to move from a supine to a side-lying position to a quadruped position. The patient then was instructed to come into a full-kneeling position using a chair for balance, followed by moving to a half-kneeling position, and finally to standing using the arms with the planted foot for weight bearing. The patient was able to successfully perform this transfer after instruction that day.

Home program.

During session 3, the patient was instructed in a seated hamstring muscle stretch to improve flexibility of this muscle group. She was instructed to sit on her bed with one lower extremity straight out in front of her, with her knee maintained in extension and her hip in neutral rotation. Her other lower extremity was off the bed, with the foot supported by the floor. The patient was instructed to bend forward at the waist and to lead with her chest until she felt a stretch behind the knee. She was instructed to perform 3 to 5 repetitions and to hold each stretch of 60 seconds.49 She was provided with daily stretching logs to record the frequency of stretching. She required review during sessions 5 and 7 to reinforce proper stretching technique as well as to reinforce that these stretches should be performed daily.

Shoe lift.

At approximately the midpoint of the intervention, the patient had a 12.7-mm external shoe lift placed on the soles of 2 pairs of everyday shoes.


The patient was retested using the same method as the initial examination for tests and measures that showed deficits initially. These tests and measures included 6-minute walk distance, gait speed, tandem standing balance, hamstring muscle flexibility, isometric muscle force, pain, balance confidence (ABC Scale), lower-extremity disability (LEFS), and SF-36. The patient noted no changes in her health history since the initial examination. The patient noted that she was still using the cane when she ambulated in the community, but no longer needed the cane in the house. The patient stated that she was continuing the hamstring muscle stretching exercises daily.

Walking endurance was retested using the 6-minute walk test. She walked 426.72 m during the second of 2 trials. This measurement represents a change in distance of 78.33 m and a gain of 22.5%. A change greater than 54 m has been found to be the minimal clinically important difference in patients with pulmonary disease.55

Gait speed was tested with the patient wearing her modified shoes. For the trials with the cane, her free gait speed increased to 0.86 m/s. For the trials without the cane, her free speed increased from 0.66 m/s to 0.94 m/s. A summary of data from the gait speed trials is shown in Table 1. The patient continued to demonstrate a Trendelenburg gait pattern. Tandem standing balance time increased to 60.0 seconds, which was a 400% increase from initial testing.

At the completion of the intervention phase, the patient demonstrated proper stretching technique and reported that she was performing the stretches. Goniometric reassessment of bilateral straight leg raising, however, showed no change in range of motion. The patient did not return the stretching log, and we do not know how frequently she performed the exercise.

Isometric muscle force increased for bilateral hip extensors, hip abductors, and knee extensors. Increases ranged from 61 to 82 N on the left and 24 to 75 N on the right (Tab. 2). The left lower extremity remained weaker than the right lower extremity except for hip abduction. Pain in the left lower extremity decreased from 3.5 to 2.5/10.

The patient's ABC Scale score increased to 75%, which was an increase of 41.25% from initial testing. The patient scored 53/80 on the LEFS, an increase of 12 points, and her SF-36 scores improved for physical function and role–physical subscales. The patient reported that she was able to walk long distances for shopping and was able to get up and down from the floor with ease when playing with her grandchildren. Table 3 provides a summary of outcomes scores.

Table 3.

Summary of Outcomes Assessments for a 68-Year-Old Woman After Hip Fracturea


This case report describes what we believe is a comprehensive examination and exercise program for a patient after hip fracture who had been discharged from physical therapy. Despite being discharged from routine care, the tests and measures provided useful information that indicated that she still had residual deficits. The Functional Reach Test, TUG Test, and semi-tandem balance test did not indicate that the patient was at risk for falls or that she had “poor balance.” Although they typically are used with older adults, these tools did not appear sensitive enough to identify her balance problems. Free and fast gait speeds, ABC Scale scores, LEFS scores, and SF-36 scores, however, all suggested that the patient had significant disability (limitation or inability in performing tasks, activities, and roles to levels with physical and social contexts).56 Although we used a technical piece of equipment (GaitMat II), gait speed also can be determined using a stopwatch and a measured distance. This method also has been estimated to yield valid and reliable measurements.57 By choosing tests and measures that measure a wide range of function, we believe we were able to identify problems that may have been overlooked.

We have not seen descriptions of these exercises being performed with patients following hip fracture, and we believe that the overload principles that we applied are not being used with other patients following hip fracture. Our clinical experience suggests that the most frequently performed exercises are those that the patient stated she was performing since discharge from home care: isometric quadriceps femoris and gluteus maximus muscle sets, heel-slides, hip abduction, straight leg raises, hip extension, marching, mini-squats, and heel-raises. Individualized exercise prescription has not been reported for patients after a fracture; therefore, exercise may not provide sufficient overload, or it may not be comprehensive (ie, involves muscle force and endurance training for all major groups of lower-extremity muscles). The literature supports this contention. Binder et al17 described a facility-based weight training program for patients 6 months following hip fracture. Although this exercise produced sufficient overload for some of the lower-extremity muscles, this type of training is unlikely to address the possible impairments in patients after a fracture. In contrast, Tinetti and colleagues58 described a comprehensive exercise approach for flexibility, “muscle conditioning,” balance retraining, and gait and transfer training, but the exercise, as described, did not appear to provide sufficient overload, nor did the program include aerobic training. We could not find reports that address aerobic training in patients following hip fracture.

Aerobic training was indicated for this patient who reported fatigue and an inability walk long distances as needed for shopping. Anecdotal experience suggests that, during acute and subacute rehabilitation, patients most likely engage in short-distance gait training (up to a maximum of 45 m [150 ft]) and active exercise a half hour to 2 hours per day. The rest of the day, patients are in bed or in a wheelchair. This amount of activity most likely contrasts sharply to the activity performed before the fracture when the patient was independent in the home and community. The activity level may increase during home care, but compared with the prior level of function, we believe that a net deconditioning effect occurs during the recuperative period. The duration of this deconditioning may be one of the factors that contribute to the poor functional outcomes of patients after hip fracture.59 Whether aerobic training is safe for all patients after hip fracture and the effect of training on functional outcomes are necessary areas for research.

The changes in lower-extremity force production, measured both isometrically and during training with the 8-RM method, suggest that the patient improved. There is strong evidence that high-intensity training is effective in increasing force production in elderly people. A recent systematic review of progressive resistive training in elderly people has shown a strong positive effect on leg extensor muscle force with moderate- to high-intensity training.50 The mechanism underlying the increase in force production for this patient is not known, but we speculate that it could be due to a combination of muscular and neural adaptations.

In a review by Barry and Carson,60 the effects of muscle force training in elderly people were reviewed at the level of the whole muscle and at the level of single muscle fibers and for the myriad of neural adaptations that can occur at supraspinal levels (motor learning), at the spinal cord, and at the muscle. For example, training has been shown to cause muscle hypertrophy and increased tendon stiffness. At the level of the single muscle fiber, increases in contractile velocity and power and muscle fiber composition changes have been noted in both type I and type II fibers, with more pronounced changes in type I fibers. Neural adaptations include increased motor unit recruitment, increased motor unit firing rate, improved coordination between agonist and synergist muscles, and decreased coactivation of agonist and antagonist muscles.60 Although we are not sure what mechanism was responsible for improved force production, we believe that, by adhering to the overload principle and having the patient exercise in movement patterns used during function, we addressed the muscular and neural components of force production. We believe the “plasticity” of the neuromuscular system to adapt to resistance training strongly argues that resistance training can and should be done with elderly people.

It is interesting to note the amount of force this patient was producing at the initial examination. The patient began the sessions performing a unilateral leg press against a resistance of 267 N (60 lb). At the end of 6 weeks, she was pressing against a resistance of 445 N (100 lb) with the fractured leg. Similarly, she initially moved 173 N (39 lb) with the hip extensors, which increased to 218 N (49 lb) at 6 weeks. The intensity of these exercises is in sharp contrast to the exercises she was performing (quadriceps femoris and gluteal muscle sets in the supine position and standing active range of motion). Although gains in muscle force can be produced from free weights and elastic bands,58 little evidence supports one type of device over another. The ability of patients after fracture to work at higher intensities of resistance to return muscle to prefracture levels of force production should be investigated.

For this patient, gait speed was initially enhanced with the use of a cane, but after intervention, she could walk faster without the cane. We assumed a cane or assistive device would improve ambulation, but, in her case (and maybe others), cane use was associated with slower gait speed. To our knowledge, no one has examined the effect of the use of assistive devices on gait speed in elderly people. A recent report61 suggested that elderly people who use canes may develop additional walking difficulties at a 2-year follow-up. The relationship between gait speed and assistive device use should be explored in future research.

The use of outcome measures gave us insight into other areas of function that otherwise might have been difficult to quantify. Balance confidence is a construct that attempts to define how assured a person is that he or she can avoid a fall in a given situation.43 Programs that attempt to build physical skills and confidence have been recommended for elderly people with fear of falling.51 By including the simple, yet necessary, skill of moving from the floor to standing, we believe we addressed physical skill and confidence. The patient reported that she was amazed at how simple the technique was to perform and that the ability to perform the task allowed her to play on the floor with her grandchildren. The LEFS score also improved. By the end of the program, the patient reported that she had little or no difficulty doing her usual hobbies and recreation, lifting objects such as a bag of groceries from the floor, performing heavy activities around the home, getting in and out of a car, walking 2 blocks, and rolling in bed. The improvement of 12 points suggests that the change was clinically meaningful.45 We advocate the use of disability scales with patients after hip fracture and believe that floor-to-stand transfers should be addressed with these patients.

Despite the improvements noted, the patient did not recover to her prefracture level of function. We can speculate that there are several reasons for this different level of recovery. One factor may have been the patient's age. Speed of healing decreases with age.62 Another factor that may have contributed was her fear of falling. Despite high levels of motivation, she also reported great fear of falling. People with a fear of falling have been reported to limit activities themselves.63 The exercise program may not have been of sufficient duration. Although gains in force production have been noted after 8 weeks of exercise, other programs last 16 weeks and report gains throughout the entire 16 weeks.50 The relative contribution of each of these factors also is unknown.

This report has several limitations. We selected a person who had completed routine care for 2 reasons: (1) to demonstrate the presence of a residual disability at the completion of care and (2) to determine whether applying the classical principles of overload and specificity would affect function. The patient was younger than the average patient with a hip fracture and was highly motivated. This program may not be tolerated as well by someone who is 20 years older or someone who is less motivated for recovery. It also is not clear whether this program could have been introduced earlier after fracture. We attempted to provide one example of a comprehensive physical therapy program suitable for patients following hip fracture. The program was not meant to be the definitive exercise prescription for all patients with hip fractures. We believe that research trials should be conducted to address comprehensive care and timing of exercise programs.

Although we do not suggest that this therapeutic exercise program caused the changes in muscle force, gait speed, balance, balance confidence, and endurance, we believe that our program addressed the specific needs of the patient and helped her achieve her goals. The majority of time spent with the patient was with muscle force training and aerobic training on a cycle ergometer, yet some of the changes the patient experienced were in gait speed, walking endurance, balance, lower-extremity function, and balance confidence. We believe that comprehensive exercise programs can be performed with some patients after hip fracture and that properly designed programs can affect patient outcomes beyond observed impairments.


  • Dr Mangione provided concept/idea/research design, patient, and facilities/equipment. Both authors provided writing, data collection and analysis, and project management. The authors acknowledge Rebecca Craik, PT, PhD, FAPTA, for providing consultation (including review of manuscript before submission).

    This study was approved by the Committee on the Protection of Research Subjects at Arcadia University.

    This work was funded by a Research Grant from the Foundation for Physical Therapy.

  • * EQ Inc, PO Box 16, Chalfont, PA 18914-0016.

  • Sammons Preston Roylan, 4 Sammons Ct, Bolingbrook, IL 60440-4989.

  • Chatillon Force Measurement Systems, AMETEK TCI Division, 8600 Somerset Dr, Largo, FL 33773.

  • § Chattecx Corp, 101 Memorial Dr, PO Box 4287, Chattanooga, TN 37405.

  • # Contemporary Design Co, PO Box 5089, Glacier, WA 98244.

  • ** Monark Exercise AB, Kroonsväg 1, 780 50 Vansbro, Sweden.

  • †† Polar Electro Oy, Professorintie 5, 90440 Kempele, Finland.

  • Received July 14, 2004.
  • Accepted December 13, 2004.


View Abstract