An impairment and disability assessment and treatment protocol for community-living elderly persons.

Key Words: Elderly, Falls and immobility, Home care, Physical therapy,

Falls and immobility are common problems among community-living elderly persons. At least one third of community-living persons over age 65 years fall each year; 10% of these falls result in serious fractures and other injuries.[1-5] Almost one half of persons who experience a fall are unable to get up without help,[2,4,6] and 25% of fallers report avoiding previously performed activities because of fear of falling.[3,5] Similarly, the proportion of elderly persons reporting difficulty with mobility tasks increases with age. Among persons over age 75 years, for example, 10% have difficulty walking across a room, 25% are unable to walk half a mile, and 30% cannot climb stairs.[7,8] Until recently, falls and mobility problems were considered inevitable accompaniments of aging and thus received little remedial attention. Increasingly, however, it is being recognized that falls and immobility result not from increasing age but from the accumulated effect of multiple impairments and disabilities.[1-5,9,10] Many of these impairments are modifiable, suggesting that the rate of falls and immobility problems among elderly persons can be reduced. Physical therapists should play a central role in testing and implementing fall and immobility prevention strategies, as impairments in muscle strength, range of motion (ROM), gait, and balance as well as hazards in the environment are among the most modifiable risk factors.[4,5,10,11]

A successful fall and immobility prevention program requires assessment and intervention strategies that target the important modifiable impairments and that are feasible in a home setting. Assessments that were developed for specific diseases such as Parkinson's disease or stroke,[12-14] may not be appropriate if, as is usually the case, the immobility results not from a single disease process but from the accumulated effects of multiple chronic processes. Programs of general strengthening and conditioning exercises may be ineffective, however, if not sufficiently intense or not targeted to each person's unique combination of impairments.[15] Finally, programs that were developed for frail, supervised nursing home residents would not be appropriate for community-living persons who must carry out the exercise regimens unsupervised.[16]

Perhaps the optimal assessment and intervention strategy would be one that couples the identification of modifiable impairments with interventions that are safe and effective yet feasible in the home setting. If this assessment and intervention strategy were standardized, additional benefits would include enhanced ability to document progress for reimbursement purposes; communicate care plans efficiently among therapists; and train physical therapist assistants, physical therapy aides, and families to assist in implementing the intervention. A standardized approach would also reduce the likelihood of inadvertently neglecting a potentially effective treatment. Before this, or any, strategy can be recommended for clinical implementation, it must first be proven reliable, feasible, and effective at achieving its stated goals. This report addresses the first two characteristics.

In this article, we describe an assessment and intervention protocol that was developed to assess the effectiveness of home-based exercises, targeted at identified impairments, in reducing the risk of falls and increasing or maintaining mobility among community-living elderly persons.[17] The purposes of our study were (1) to describe the development of the assessment and intervention protocol and (2) to determine the interrater reliability of both the assessment and intervention components of the protocol,

Method

Assessment and Intervention Protocol Development

The physical therapy protocol was developed through a consensus approach. The consensus group included a geriatrician with 9 years of rehabilitation experience, three physical therapists with a combined total of over 40 years of experience in home-based physical therapy, and two nurses with a combined total of 30 years of experience in rehabilitation and home care. The group was instructed to use their clinical experience and judgment, backed by the rehabilitative and geriatric literature when available, to develop the baseline assessment and intervention strategies. The development of the assessment component included selecting (1) impairments to target, (2) the techniques for diagnosing each impairment, and (3) the findings on the assessment that constituted indications or contraindications for intervention. The group targeted neuromuscular impairments that were felt to be (1) associated with falling or activities-of-daily-living dysfunction; (2) "common" among community-living elderly persons; and (3) potentially amenable to "safe, feasible" home-based interventions. The assessment techniques were chosen based on present usage in clinical practice, on purported interrater reliability, and on feasibility of use in the home setting. Using these criteria, the consensus group developed the falls and immobility assessments over several weeks of discussion and testing. The components of the assessment, along with the finding indicating need for intervention, are described in Appendix 1.

Following the development of the assessment protocol, the consensus group selected appropriate interventions for each of the targeted impairments based on the following criteria: (1) "safe" in an unsupervised home setting (this was felt most important), (2) feasible in the home setting in terms of space and time available and in terms of likely patient adherence, and (3) clinical evidence or experience supporting effectiveness at ameliorating the targeted impairment. The group agreed that the exercise components of the intervention protocol should be (1) competency-based, involving a program of graded exercises based on initial severity of impairment and on progressive improvement over time; (2) standardized for each targeted impairment as much as possible; and (3) aimed at reaching the optimal level of improvement in the impairment feasible in an unsupervised, home-based exercise program. Because the eventual goal was the development of a program that could be incorporated into standard clinical practice, intensive exercise programs that required transporting patients to a center or that required close, ongoing professional supervision were considered infeasible. The interventions selected for each impairment are described in Appendixes 2 and 3.

Our protocol was developed to complement a medical-nursing protocol aimed at other modifiable risk factors for falls and immobility such as sedative use, multiple medication use, and postural hypotension.

The consensus group recognized that many, if not most, patients likely to benefit from this strategy would be suffering from multiple impairments and chronic diseases. Therefore, we did not want to overburden these patients with multiple simultaneous exercise programs. Too many exercises could result in confusion and in poor adherence, or in increased chance of injury. Thus, the consensus group prioritized the interventions based on the group's assessment of the likely contribution of the impairment to fall and immobility risk. The order of priority was (1) balance and transfer interventions, (2) lower-extremity strengthening exercises, (3) lower-extremity ROM exercises, (4) upper-extremity strengthening exercises, and (5) upper-extremity ROM exercises. The group agreed that the maximal number of exercise-based interventions that could be prescribed was three. Each exercise program was designed to be performed twice a day. The combination of three exercise programs would require about 15 to 25 minutes twice a day. More than 30 minutes per session was felt to be inappropriately burdensome. Therefore, for patients with impairments requiring no more than three exercise regimens, all exercises were recommended, whereas for patients with impairments requiring more than three exercise regimens, the three exercise programs with the highest priority were recommended. These priorities pertained only to the exercise programs. All patients meeting the criteria for environmental modification, gait and transfer training, and foot care received these components of the treatment strategy.

Reliability Testing of Assessment Protocol

A convenience sample of 11 residents (3 male, 8 female) of a senior housing complex was selected to test the interrater reliability of the assessment. The mean age of these residents was 82.1 years (range=69-91). All subjects were cognitively intact, as judged by the staff of the senior housing complex, and ambulatory. Four subjects used assistive devices, and 2 subjects were receiving home health aide services.

The two physical therapists with experience in home therapy and geriatrics helped develop the assessment protocol, thereby standardizing their approach to assessment. Each of the therapists assessed the subjects in a variable order, blinded to the results of the other therapist's assessment. The two assessments on each subject were performed within a 2-week period. The assessments took place in the subjects' homes and required approximately 45 minutes to complete. Assessments by both therapists were completed for all but 1 of the 11 subjects, who received only one assessment due to intervening illness.

Percentages of agreement and Kappa statistics were calculated for each item in the assessment to determine interrater reliability.[18] The Kappa statistic estimates the extent of agreement above that expected by chance alone. To simplify calculations and to correspond to therapeutic decision making--that is, that subjects scoring less than normal should receive intervention--items with more than two categories were dichotomized to normal or less than normal. In calculating percentages of agreement and Kappa statistics, left- and right-side results were treated separately so that 22 comparisons were available for muscle testing, ROM, and foot items. For purposes of calculating Kappas, we assumed that these measurements were independent of each other.

Reliability Testing of Intervention Recommendations

Subjects involved in reliability testing of the intervention recommendations were drawn from participants in a multiple risk-factor intervention trial for fall prevention. This trial is described in detail elsewhere.[17] In brief subjects were members of a participating health maintenance organization who were [greater than or equal to] 70 years of age, cognitively intact, not terminally ill, not very physically active, and possessed at least one of the following fall risk factors: postural hypotension; sedative use; use of at least four prescription medications; upper- or lower-extremity strength or ROM impairments; foot problems; and balance, gait, or transfer dysfunctions. A total of 153 subjects were randomly assigned to the intervention group. These subjects were assessed and treated by one of the two study physical therapists. Ten subjects treated by each of the therapists were randomly selected for interrater reliability testing of the intervention protocol. The therapist who had not treated the subject reviewed the baseline assessment. Then, using results of the assessment as well as the decision rules described earlier, she selected the intervention she would have recommended had she treated the subject. Because these therapists were involved in the development of the assessment protocol, their degree of agreement may have been greater than that expected for other similarly experienced therapists.

The intervention recommendations of the two physical therapists were then compared. As for the assessment protocol, percentages of agreement and Kappa statistics were calculated for each intervention component. Only interventions recommended in at least 3 subjects are reported. The mean age of these subjects was 76.4 years, with a range of 71 to 86 years; 14 subjects were women, and 6 subjects were men.

Results

The interobserver reliability data for the components of the assessment protocol are listed in Table 1. The percentages of agreement ranged from 73% to 100%, with 25 of the 32 items (78%) showing agreement over 90%. With a few exceptions, the Kappa values ranged from .60 to 1.00, representing good to perfect agreement between the two assessments. The Kappa values for elbow extension and flexion, sitting balance, decreased turning, and carrying objects were .00 or .10 because of the uneven distribution of results. Almost all subjects scored normally; thus, even two disagreements greatly affected the results.[19]

[TABULAR DATA 1 OMITTED]

Results of interrater reliability testing of the intervention recommendations are shown in Table 2. Percentages of agreement between the two physical therapists ranged from 75% for lower-extremity strengthening exercises to 95% for gait training, transfer training, and foot care recommendations. The agreement for lower-extremity strengthening exercises was likely lower because the therapists had to choose among ankle, knee, and hip strengthening exercises rather than to select a single intervention such as a balance exercise program, Agreement on the total package of intervention recommendations was also evaluated. The two therapists showed complete agreement on all components of intervention recommendations for 12 subjects (60%) and partial agreement for the remaining 8 subjects (40%).

[TABULAR DATA 2 OMITTED]

Discussion

This article describes the development and testing of a home-based assessment and intervention program for reducing the risk of falls and immobility among elderly patients. The protocol was developed by therapists, nurses, and physicians with over 80 years of cumulative experience in treating multiply and chronically impaired elderly patients, thus endowing the protocol with good face validity. The assessment and intervention protocol proved to be feasible--the assessment required only 45 minutes on average to complete--and reliable. The amount of agreement between the two physical therapists as to what impairments subjects possessed and how to intervene with individual subjects was good to excellent. The instances in which the Kappas were either .00 (eg, shoulder extension, elbow extension and flexion) or .10 (eg, carrying an object) occurred when both therapists rated most of the subjects as unimpaired in these areas, The lack of variability in results among subjects results in low Kappas even when there is good interrater agreement.[19]

Falling and immobility are problems resulting from the accumulated effects of multiple diseases, impairments, and disabilities. A successful prevention and intervention program should therefore be multidisciplinary, Our protocol was developed to be used in conjunction with nursing or physician assessment of other factors known to contribute to risk of falls or immobility. Postural hypotension, vision impairment, and use of medications such as sedative-hypnotics are examples of other common problems contributing to immobility that need to be addressed in concert with the impairments outlined here. Systematic assessment and removal of environmental hazards represents another integral part of a fall and immobility prevention program.[11,20] The assessment and intervention protocol outlined in this article is oriented toward identifying and ameliorating impairments and disabilities with the goal of optimizing function. This protocol was not intended as a diagnostic tool. The targeted patients for this program are multiply and chronically impaired elderly persons in the community, who constitute a majority of persons at risk for falls and immobility. It is not clear whether the strategy recommended here is appropriate for persons with a single, overwhelming disease contributing to fall and immobility risk such as severe Parkinson's syndrome or hemiplegia. The program also has not been tested as yet in cognitively impaired elderly persons.

The order of priority among exercise regimens was determined with fall prevention and mobility maintenance as the primary goals of treatment. Establishing the goals of treatment should be the first step in developing a treatment plan for multiply and chronically impaired elderly persons. Priorities will vary depending on the identified goals. For example, upper-extremity exercises would receive higher priority if increased independence in basic activities of daily living were the primary goal.

Although not addressed in this study in which subjects were evaluated only at baseline, in clinical practice it will be necessary to reevaluate patients to determine whether progress is occurring and to decide whether addressing additional problem areas is feasible. The optimal frequency of these reevaluations and the criteria for deciding whether to substitute or add exercise regimens remains to be determined.

To optimize the sensitivity and reliability of the assessment, the number of possible categories of results was fewer than typically used in clinical practice. For example, we used a three-category scoring system for gross muscle testing rather than the more typical five- or six-grade measure used in manual muscle testing. The purpose of the assessment was to determine who required an intervention. In clinical practice, a physical therapist would likely intervene with patients showing any decrement in strength. Therefore, we contend that our three-level categorization, which probably increased reliability, did not impede clinical judgment. An obvious limitation of our simple categorization system, however, is that it is not sufficiently sensitive to monitor small increments of progress or change with therapy. To accomplish this task, the assessment would need to be coupled with a potentially more sensitive measure such as one repetition maximum strength testing or possibly portable dynamometry.

Interrater reliability testing was conducted on only 11 subjects because of limited resources. This number, however, resulted in good to excellent Kappas for most measurements. Interrater reliability was tested only between two experienced physical therapists who were integrally involved in the protocol's development. Interrater reliability should also be determined among less experienced and new therapists, and among therapists who have not reviewed the protocol as intensively as the study physical therapists.

As noted earlier, a standardized assessment and intervention protocol such as the one presented here may offer several advantages to physical therapists treating multiply and chronically impaired patients at home. In addition to improved communication among care providers and more efficient documentation for reimbursers, the protocol provides a direct link between impairment assessment and intervention recommendations, This link lessens the inevitable problem of overlooking possibly effective treatment options in patients with multiple impairments. It is important to bear in mind that, although the assessment is standardized, the intervention link and decision rules incorporate clinical judgment and individualization of treatment. The protocol, because of its thoroughness and consistency, enhances rather than detracts from therapist-patient interaction and decision making. A potential future advantage of a protocol such as this one is that it is readily adaptable to laptop or notebook computers. A move toward computerization of assessment in rehabilitation is occurring because of the potential for improved efficiency, documentation, and communication.

Conclusion

In summary, falling and immobility are common problems among community-living elderly persons that result from the cumulative effects of multiple impairments. Although previously considered inevitable accompaniments of aging, these problems are increasingly recognized as preventable or treatable. Physical therapists play a pivotal role in assessing and treating falls and mobility problems. A systematic assessment of risk factors coupled with feasible and effective interventions should aid physical therapists in treating at-risk elderly patients, The ultimate test of this strategy--determining the effectiveness of this approach in reducing the risk of falls and immobility--is currently under investigation.

References

[1] Blake AJ, Morgan J, Bendall MJ, et al. Falls by elderly persons at home: prevalence and associated factors. Age Ageing, 1988;17:365-372. [2] Nevitt MC, Cummings SR, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls: a prospective study. Jama. 1989;261:2663-2668. [3] Nevitt MC, Cummings SR, Hudes ES, Risk factors for injurious falls: a prospective study. J Gerontol. 1991;46:M164-M170. [4] Campbell AJ, Borrie MJ, Spears GF, Risk factors for falls in a community-based prospective study of people 70 years and older. J Gerontol. 1989;44:M112-M117. [5] Tinetti ME, Speechley M, Ginter SF, Risk factors for falls among elderly persons living in the community. N EnglJ Med. 1988;319: 1701-1707. [6] Tinetti ME, Liu WL, Claus E. Predictors and prognosis of inability to get up after falls among elderly persons. JAMA. 1993;269:65-70. [7] Cornoni-Huntley J, Brock DB, Ostfeld AM, et al. The Established Populations for the Epidemiologic Study of The Elderly: Resource Data Book. Bethesda, Md: National Institutes of Health; 1986. NIH Publication No. 86-2443. [8] Jette AM, Branch LG. The Framingham Disability Study: physical disability among the aged. Am J Public Health. 1981;71:1211-1216. [9] Tinetti ME. Falls and immobility, In: Beck JC, ed. Geriatrics Review Syllabus. 2nd ed. New York, NY: American Geriatrics Society; 1991: 195-203. [10] Whipple RH, Wolfson LI, Amerman PM. The relationship of knee and ankle weakness to falls in nursing home residents: an isokinetic study. J Am Geriatr Soc. 1987;35:13-20. [11] Tideiksaar R. Preventing falls: home hazard check list to help older patients protect themselves. Geriatrics, 1986;41:26-28. [12] Webster DD. Clinical analysis of the disability in Parkinson's disease. Mod Treat. 1968; 5:257-282. [13] Canter CJ, de la Toree R, Mier M. A method of evaluating disability in patients with Parkinson's discase. J Nerv Ment Dis. 1961; 133: 143-147. [14] Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology. 1967; 17:427-442. [15] Reinsch S, MacRae P, Lachenbruch PA, Tobis JS. Attempts to prevent falls and injury: a prospective community study, Gerontologist. 1992;32:450-456. [16] O'Neil MB, Woodard M, Sosa V, et al. Physical therapy assessment and treatment protocol for nursing home residents, Phys Ther. 1992; 72:596-602. [17] Tinetti ME, Baker DI, Garrett PA, et al. Yale FICSIT: risk factor abatement strategy for fall prevention, J Am Geriatr Soc. 1993;41:315-320. [18] Fleiss JL. Statistical Methods for Rates and Proportions. 2nd ed. New York, NY: John Wiley & Sons Inc; 1981:212-236. [19] Feinstein AR, Ciccetti DV. High agreement but low Kappa, I: the problems of two paradoxes. J Clin Epidemiol. 1990;43:543-549. [20] Tinetti ME, Speechley M. Prevention of falls among the elderly. N Engl J Med 1989;320: 1055-1059,

Appendix 1. Description of Assessment Protocol

Strength Testing

General Procedures: All muscle groups are tested bilaterally initially to avoid substitution and t when a grade is less than normal or when bilateral testing is not feasible (eg, ankle or hip). For u stabilizes the patient just proximal to the joint being tested.

The therapist first demonstrates the motion, then assists the patient in assuming the correct test times. If the patient is able to hold the test position, the therapist first applies fingertip resis the position against fingertip resistance, the therapist applies resistance equal to about 4.5 kg (1

Muscle Groups Tested: The muscle groups selected for testing are those that we believe to be most that are highly correlated with the groups not tested. These muscle groups include shoulder extensor and extensors; hip extensors, flexors, and abductors; knee extensors and flexors; and ankle dorsifle

Patient Positioning: Seated in hard, armless chair for shoulder, elbow, and hip flexion and abduct Standing holding on to chair for hip extension and ankle plantar flexion. In general, testing is don tested at full range of motion (ROM).

Instructions for Physical Therapist: Therapist applies resistance as described at distal point of knee flexion).

Grading: Each muscle group is tested three times, and the best result is graded. In our study, gra resistance, some resistance, and no resistance or less than full active ROM. However, the assessment several widely accepted muscle grading systems.

Indications for Intervention: Patient unable to hold position against maximal resistance and no co long-standing plegia).

Range of Motion

General Procedures: Done after strength testing, which will result in optimal ROM as the joint cap testing. The therapist takes each joint to full passive ROM.

Joints Tested: Correspond to the muscle groups tested previously.

Patient Position: Supine for all testing.

Instructions for Physical Therapist: For each joint, the starting position, the method for stabili goniometer, the stationary arm of the goniometer, and the movable arm of the goniometer are specifie described by the American Academy of Orthopedic Surgeons. The patient is instructed not to assist th complete when no further movement is possible or when the adjacent limb moves.

Grading: As per goniometer.

Indications for intervention: If ROM is less than considered necessary for activities of daily liv abduction <90[degrees], flexion <150[degrees], extension <20[degrees]; elbow flexion <140[degrees], <90[degrees], extension not within 10[degrees] of full extension; ankle--unable to attain neutral do

Balance, Transfers, Position Changes

General Procedures: The patient is observed performing a series of balance maneuvers and position patient is asked to perform the maneuver or transfer as he or she normally does, as the purpose is t should be tested using the furniture he or she normally uses (eg, armchair, couch, bed). The patient assistive device normally used.

Maneuvers and Position Changes Tested: Stand to sit, sitting, sit to stand, standing, lie to sit,

Observations: As the patient performs these position changes, the therapist should observe and rec excessive leaning in any direction, loss of balance in any direction, need to hold on to object for forward in chair or too close to edge in bed).

Indications for Intervention: Any of the observations noted previously during any of the maneuvers indication for intervention.

Gait

General Procedures: The patient is asked to walk, wearing the shoes normally worn at home. The pat without any assistive device. Patients are asked to walk at their usual pace. The most challenging s ground better than thick carpet, thick carpet better than bare floor). The patient is asked to walk four times, The therapist observes only one aspect of gait during each walk to ensure reliable obser obstacle such as a shoe or book is placed in the path. After testing patient's "usual pace," the pat as a more rapid pace may bring out more subtle impairments.

Gait Components: Initiation, step height and length, arm movements, turning, trunk and lower-extre obstacles.

Observations: Therapist should observe and record the following: missed steps or stumbles; loss of forward, to the side, or backward; loss of balance or stumbles on turns; need to reach for objects s step length, resulting in consistently less stance time on one foot compared with other foot; decrea deviation of path, resulting in weaving or swaying. These impairments may be more obvious during a r

Indications for Intervention: Any of these observations during any of the walks constitutes an ind

Foot Problems

Calluses, bunions, hammertoes, or toenails causing pain or abnormal gait pattern.

Appendix 2. Description of Intervention Protocol

I. General Principles

Agreement Between Patient and Physical Therapist: Following the assessment and identification of i impairments with the patient, explaining the risk of falling and immobility and emphasizing the like therapy. Understanding the risk of these impairments and potential benefit of therapy is essential t exercise program.

Exercise Instructions: During the home visits, the therapist demonstrates, then observes the patie safe and appropriate performance. A single page with illustrations and simple instructions for each patient is instructed to perform the exercises twice a day. The number of repetitions varies for eac following sections.

Priority: Most patients will have multiple impairments. To decrease burden and increase compliance important to limit the total exercise time to 20 to 30 minutes per session. For patients with three three exercise programs can be instituted. For patients with more than three exercise-responsive imp exercises felt most likely to have an impact on fall risk or immobility are selected for implementat as follows: (1) balance, (2) lower-extremity strength, (3) lower-extremity range of motion (ROM), (4 upper-extremity ROM. For muscle groups in the same extremity (eg, hip and knee), the weakest muscle intervention, If the muscle weaknesses are roughly equivalent, the more proximal muscle group is sel improves, additional muscle strengthening or other exercises can be substituted. This priority syste that must be performed unsupervised by patients twice a day. All patients who meet the criteria rece and environmental interventions.

II. Specific Interventions

A. Progressive Resistance Strength Exercises

General Instructions: The therapist instructs the patient in correct positioning and setting up th perform each resistive exercise 10 times on the right and left.

Positioning: Supine for shoulder, elbow flexion, hip flexion and abduction, ankle plantar flexion knee extension and flexion.

Initiation: Patients with a decrease in ROM or who are unable to move against minimal resistance b resistance. Patients exhibiting full ROM and who are able to move against some resistance begin resi yellow Thera-Band[R].

Progression: Once patients are able to perform 10 repetitions through full ROM (unless passive ROM Those without resistance are progressed to yellow Thera-Band[R], and those who begin on yellow Thera Thera-Band[R]. Additional progression can be instituted if indicated.

Maintenance: Once patients reach their maximal resistance (eg, red Thera-Band[R]) and can perform passive ROM limited), they can be discontinued from active therapy. These patients are instructed to level. Patients must be informed that their strength gains will be maintained only if the exercises decreased from daily to three times a week.

Contraindications or Interruptions: For patients with severely decreased ROM (eg, secondary to con whose muscle strength cannot be tested secondary to severe deformity, these exercises should not be facilitatory therapy would be necessary. For patients who have intercurrent illnesses, resulting in perform the exercises without great discomfort, the exercises should be reinstituted at a lower leve

B. Range-of-Motion Exercises

The positioning and movements for ROM are the same as for strengthening exercises. The ROM exercis

C. Progressive Balance Exercises

General Instructions: Patients with any impairment in balance, gait, and transfer testing receive These exercises are prescribed in addition to specific gait and transfer training. As for the streng demonstrates, and then observes the patient performing, the balance exercises. Patients are given si instructions for the exercises, Each movement is repeated 5 or 10 times per session.

Levels: There are four levels of progressive balance exercises, corresponding to progressively dec The specific balance exercises are shown in Appendix 3.

Initiation and Progression: All patients begin at balance level 1. They progress to the next level performed correctly and safely. The patient should demonstrate correct form of the exercise for at l

Maintenance: Some patients will progress quickly through a level, whereas others will progress mor plateau at which there are no further improvements. This level is operationally defined as no progre assuming the exercise sessions have been completed as prescribed. At this time, the patient is moved maintenance phase, patients are instructed to continue performing the balance exercises at the level The patient may decrease from daily to three-times-a-week performance of the exercises. The therapis exercises are required to maintain the improvements.

It is important to note that the levels of balance exercises are limited in an at-home, unsupervis require hands-on supervision for safety.

D. Gait and Transfer Training

Patients will receive the strengtheneing, balance, and ROM exercise programs as indicated by their who exhibit any deficits during the gait and transfer assessments will also receive gait and transfe

Gait training consists of instructions in: (1) appropriate footwear--comfortable, well fitting, pr that are neither too slippery nor "sticky"; heels that are low, broad, or wedge-shaped; avoidance of walking--feet should be 10.2 to 15.2 cm (4-6 in) apart; with each step, heel should land on floor fi lands on floor; ensure steps are of equal length; let arms swing naturally at sides, with right arm versa; (3) turning--when making turns, avoid sharp pivots or twisting on either leg; make shorter st letting upper body get ahead of hips and legs; avoid rushing; (5) losing balance--if patient begins readjust footing, and continue or hold on to a movable object or sit down.

Transfer Training: Patient is instructed to perform transfer in a stepwise fashion. (For example, then rolls to side, then pushes to sitting with both arms, then swings legs off bed, then sits on ed a sitting to a standing position, the patient sits on bed or chair, then scoots to edge, then puts f leans slightly forward so body is over feet, then pushes up with arms and legs to stand, then ensure When transferring from a standing to a sitting position, patient touches back of chair with legs, th feet. then reaches back for arms or seat, then slowly sits, then sits on edge of chair, then scoots a lying position, patient turns with back to bed, then touches bed with back of legs, then leans sli hands, then slowly lowers self to sit, then sits on edge of bed, then scoots back, then lowers self on to bed, then rolls to back.

Appropriate assistive devices are procured as indicated by the assessment.

E. Foot Care

General Procedures: Patients noted to have bunions, hammertoes, or calluses resulting in pain or g footwear and provided appropriate foot care.

Footwear: Recommended footwear include the following: (1) soft uppers that allow foot to breathe, are caused by seams or rough edges, (3) adequate support to the arch of the foot and conformed to ho enough to allow a 2.5-cm (1 -in) width between longest toe and end of shoe, (5) wide enough to comfo metatarsal head, (6) rounded toe, (7) increased depth of toe box to allow for deformities and preven greater than 3.8 cm (1.5 in) to allow for proper weight distribution and ensure adequate base of sup (exception may be women who have worn high heels all their lives may have difficulty with changing t gradual decrease in height of heel as well as ankle ROM exercises).

Nails: Clip nails straight across end of nail bed, buff or file edge of nails.

Calluses: Shave with a #10 scalpel by layers, holding the scalpel parallel to the surface of the s surrounding tissue. Wipe the callus with alcohol intermittently to better visualize the margins of t

Overlapping Toes: Position lamb's wool between toes, ensure patient dries well between toes after breakdown.

Bunions: Recommend proper footwear; if painful, refer to podiatrist.

Hammertoes: Recommend proper footwear; if painful, refer to podiatrist. (*) The Hygenic Corp, 1245 Home Ave, Akron, OH 44310.

Appendix 3. Progressive Balance Exercises

Level I

1. Sink toe stand with both hands 2. One-leg sink stand with both hands 3. Sink hip circle 4, Sitting arms circles 5. Sitting knee lifts, arms to side

Level II

1. Sink toe stand with one hand 2. One-leg sink stand with one hand 3. Bed walk with arms out 4. Sink side step with both hands 5. Sitting march 6. Sitting knee lifts, arms across chest

Level III

1. Sink toe stand with no hands 2. One-leg sink stand with no hands 3. Bed walk with arms folded 4. Sink side step with one hand 5. Sink leg cross 6. Sink leg swing 7. Sink leg lift 8. Heel stand

Level IV

1. Standing arm/leg march 2. Crossover walk 3. Tandem walk 4. Heel-toe walk 5. One-leg sink toe stand Progression of exercises: All subjects begin at level I and progress to higher levels when all exerc

Invited Commentary

Recently, there has been growing recognition of falls and immobility as problems in community-living old persons. This has prompted the need for the development of methods for assessing and treating community-living old persons for falls and immobility, In their article, Koch and colleagues present such an assessment and intervention protocol and examine interrater reliability and the feasibility of both the assessment and intervention components of the protocol. The epidemiological comments in their introduction are selected to fit their goal: to emphasize the need for both assessment and intervention. Nothing is mentioned about the literature on which the assessments and the interventions are based. Thus, we consider their study preliminary based on several issues related to assessment and intervention protocol development and interrater reliability testing.

Assessment Protocol Development

When assessing falls and immobility in old individuals, several factors need to be considered: ecologic inputs or environmental factors that contribute to falls and immobility; biomedical inputs or the medical events (eg, acute or chronic disease, drug side effects, infections, electrolyte imbalances) that contribute to falls and immobility; physiologic inputs or sensory (visual, vestibular, and somatosensory systems), central processing, and effector (strength, range of motion [ROM], biomechanics, flexibility, endurance) components of the postural control system; and functional inputs or the identification of routine movements that old persons have difficulty performing.[1] In their assessment protocol, Koch and colleagues evaluate two effector components (strength, ROM) of the postural control system, functional inputs, and ecologic inputs. They state

The assessment techniques were chosen

based on present usage in clinical

practice, on purported interrater reliability,

and on feasibility of use in the

home setting. A discussion of the assessment techniques chosen by the authors is warranted.

Strength Testing

Manual muscle testing (MMT) is insensitive in quantifying the strength of muscles with higher grades.[2-5] Koch et al recognize this limitation in their discussion and suggest that additional strength testing such as one repetition maximum (1 RM) or hand-held dynamometry may be necessary. In addition, for strength increases to occur in muscle, the principles of overload and specificity must be applied during resistance testing and training. The optimal training load for an individual cannot be determined from a MMT. There is also a lack of specificity between the authors' assessment technique (MMT) and resistance training techniques (Thera-Band[R](*)). Although it is laudable to attempt to make strength testing simple, standardized, and feasible, application of correct exercise physiology principles must occur or the assessment is inappropriate.

There are additional limitations of MMT as it relates to the assessment of falls and immobility in old persons that suggest that this technique is questionable. First, for various MMT grading schemes, the amounts of resistance applied to individuals of different ages, sizes, gender, and exercise history have not been determined.[6] This lack of data could affect interrater reliability measurements. Second, in balance, stabilization as well as large displacements and fast, forceful movements are required of muscles. Thus, an isometric strength test or torque measurements at different speeds of movement may be more appropriate than MMT for determining muscle strength in an assessment of falling.

Range of Motion

Measurement of ROM using goniometric techniques is considered to be a factor that needs to be evaluated when assessing falls in old individuals.[1] It should be recognized, however, that the extent to which limited ROM in joints such as the elbow contribute to loss of balance and falling may be small. An old individual can have good balance despite having limited ROM of the elbow joint.

Balance, Transfers, Position Change

Functional (eg, position changes, gait) and environmental assessments are performed in this section of the protocol. Balance is "observed" by the examiner during these assessments. Specific clinical tests (Sensory Integration Test,[7] Postural Stress Test[8]) that assess the components of the postural control system are not performed during the assessment protocol. Thus, caution must be exercised so that conclusions about the sensory or central-processing components of the postural control system are not inferred from the assessment.

Intervention Protocol Development

Progressive Resistance Strength Exercises

To obtain a training effect in muscle, an exercise overload must be applied.[9] Significant gains in force-generating capacity occur when high-resistance loads (eg, 10 RM) are applied to old muscle.[10,11] To our knowledge, there is no information as to how Thera-Band[R] correlates to maximal muscle loading. In addition, heterogeneity in aging must be considered when designing strength training programs.

To produce significant increases in muscle strength in old individuals, training at a high intensity, individualized to the muscle being trained, must be performed and maximum training loads need to be reestablished on a regular basis. It has been shown that frail persons up to 90 years of age can tolerate this type of program.[12] Additionally, when old or frail individuals are performing any type of resistance exercise program, physiological variabless (heart rate, blood pressure) should also be monitored periodically.

Finally, the authors state, "If the muscle weaknesses are roughly equivalent, the more proximal muscle group is selected [to strengthen]." Recurrent falling in old persons, however, is associated with lower-extremity weakness, especially at the ankle and knee.[13,14] Again, although there is a desire to make strength training procedures simple, standardized, and feasible in community-living old persons, application of correct exercise physiology principles must occur or effectiveness of the intervention is compromised.

Progressive Balance Exercises

There is little evidence that balance training carries over to improved function. Balance appears to be task specific. one accepts these premises, the balance exercises performed should be task specific and based on deficits noted in the functional assessment.

Interrater Reliability Determination

As stated by the authors in their discussion of the results, interrater reliability testing should be performed on a greater number of subjects, using a diverse and larger number of therapists. Even if two experienced physical therapists come to a similar result, there is no guarantee that they are not biased. Also, if a group of people who need some form of intervention is selected, the probability that agreements on intervention strategy will be reached should be high. Finally, the agreements are dominated by simple function tests where the agreements are expected to be close. Disagreements concern very important activities of daily living crucial to the evaluation of the risk of falling.

Before a method can be evaluated, results of assessment and intervention should be available. Only when expanded interrater reliability and the effectiveness of the protocol, as demonstrated by a reduction in falls and an improvement in the mobility of old persons, has been shown can the potential of this tool be assessed.

Jane F Hopp, PhD, PT Assistant Professor Department of Physical Therapy and

Section of Geriatric Medicine,

Department of Medicine University of Illinois at Chicago 1919 W Taylor St, m/c 898 Chicago, IL 60612

Alvar Svanborg, MD, PhD Professor and Chief of

Geriatric Medicine Department of Medicine University of Illinois at Chicago 840 S Wood St, m/c 787 Chicago, IL 60612 (*) The Hygenic Corp, 1245 Home Ave, Akron, OH 44310.

References

[1] Studenski S. Falls. In: Calkins E, ed. The Practice of Geriatrics, 2nd ed. Philadelphia, Pa: WB Saunders Co; 1992. [2] Beasley WC. Influence of method on estimates of normal knee extensor force among normal and postpolio children. Phys Ther Rev. 1956;36:21-40, [3] Griffin JW, McClure MH, Bertorini TE. Sequential isokinetic and manual muscle testing in patients with neuromuscular disease. Phys Ther, 1986;66:32-35. [4] Saraniti AJ, Gleim GW, Melvin M, et al. The relationship between subjective measurements of strength. J Orthop Sports Phys Ther, 1980;2: 15-18. [5] Bohannon RW. Manual muscle test scores and dynamometer test scores of knee extension strength. Arch Phys Med Rehabil. 1986;67: 390-392, [6] Bohannon RW. Manual muscle testing of the limbs: considerations, limitations, and alternatives. Physical Therapy Practice. 1992;2 (1): 11-21. [7] Shumway-Cook A, Horak F. Assessing the influence of sensory interaction on balance. Phys Ther, 1986;66:1548-1550. [8] Wolfson LI, Whipple R, Amerman P, et al. Gait and balance in the elderly. In: Clinics in Geriattic Medicine. Philadelphia, Pa: WB Saunders Co; 1985:649-659. [9] Atha J. Strengthening muscle. Exerc Sports Sci Rev. 1981;9:173. [10] Kauffman TL, Strength training effect in young and aged women. Arch Phys Med Rehabil. 1985;65:223-226. [11] Perkins LC, Kaiser HL. Results of short-term isotonic and isometric exercise programs in persons over sixty. Phys Ther Rev. 1961;41:633-635. [12] Fiatarone MA, Marks EC, Ryan ND, et al. High-intensity strength training in nonagenarians: effects on skeletal muscle. JAMA. 1990;263: 3029-3034. [13] Studenski SA, Duncan PW, Chandler JM. Postural responses and effector factors in persons with unexplained falls: results and methodologic issues. J Am Geriatr Soc. 1991;39: 229-234. [14] Whipple RH, Wolfson LI, Amerman P. The relationship of knee and ankle weakness to falls in nursing home residents: an isokinetic study. J Am Geriatr Soc. 1987;35:13-20.

Author Response

As the purposes of our study were to describe the development of the assessment and intervention protocol and to determine the interrater reliability of the assessment and intervention components, we did not discuss the literature on which the assessments and interventions were based. For readers who are interested, however, this literature is cited in an earlier manuscript.[1]

In discussing strength testing and training, Dr Hopp and Dr Svanborg seem to imply that not only are there limitations to the manual muscle testing and Thera-Band[R](*) techniques--which we acknowledged in the manuscript--but that these techniques are inappropriate. Manual muscle testing and the Thera-Band[R] exercise programs, however, are the standard of practice in home-based physical therapy, where issues of timing, feasibility, portability, and safety are paramount. Dr Hopp and Dr Svanborg state that we should have taken isometric strength test or torque measurements at different speeds of movement. Although we certainly agree that from a pure exercise physiology standpoint this would have been ideal, there is no portable equipment currently available for measuring torque at different speeds. Dr Hopp and Dr Svanborg misunderstood that we were using manual muscle testing results to determine optimal training load. Rather, all subjects who scored less than full range of motion with full resistance began exercising without resistance. The subjects were then progressed based on their ability to complete 10 repetitions, not on the results of manual muscle testing. Dr Hopp and Dr Svanborg further state that in order to obtain a training effect, an "exercise overload" must be applied. More recent data, specifically in elderly persons, challenge this principle. Strength gains have been seen with submaximal effort.[2] Again, our protocol was designed for use in home-based rehabilitation programs in which frail elderly persons, after being taught the technique, must perform the exercise programs unsupervised. The high-intensity, maximal training loads that Dr Hopp and Dr Svanborg recommend would be extremely hazardous if unsupervised, whereas a continuously supervised program in home-based clinical practice is obviously infeasible. The study that Dr Hopp and Dr Svanborg cite was performed on a very small number of elderly people in a highly supervised laboratory environment. Hopefully, researchers such as Dr Hopp and Dr Svanborg will soon translate the principles of exercise physiology into a program that can be implemented in home-based clinical practice. We chose Thera-Band[R], not only because of the frequency of use in clinical practice, but because our preliminary efforts at using ankle weights in frail elderly persons showed us that many have great difficulty applying the weights and several could not learn correct, safe technique. The effectiveness of our strength training program is suggested by the fact that the vast majority of subjects did progress to higher levels of resistance with the Thera-Band[R].

Dr Hopp and Dr Svanborg correctly point out that some studies have shown that recurrent falling is associated with lower-extremity weakness, especially at the ankle and knee. They therefore question the priority given to proximal muscle groups. First, other studies have shown that hip weakness is the best predictor of falls.[3] Second, as Dr Hopp and Dr Svanborg themselves point out, "... in balance, stabilization as well as large displacements and fast, forceful movements are required of muscles." This requirement for stabilization would argue for strengthening proximal muscle groups. Third, it is a well-accepted premise of facilitation techniques that hip flexion strengthening facilitates dorsiflexion.[4] Fourth, because our priority system favored the lower extremity over the upper extremity and because we targeted weaker muscle groups, essentially all persons with ankle and knee weakness received these strengthening programs. Dr Hopp and Dr Svanborg question the extent to which limited range of motion of the elbow contributes to loss of balance and falling. Upper-extremity functioning, however, has been shown to be an important predictor of fall injury, perhaps because of the contribution of upper-extremity movement to balance as well as protective responses.[5,6] Indeed, one of the postulated explanations for the increased frequency of hip fractures seen in the late seventies and eighties is the loss of the protective response of putting out the arm once a fall becomes inevitable.[6]

Dr Hopp and Dr Svanborg state that there is little evidence that balance training carries over to improved function. Indeed, the lack of such evidence was an important reason for conducting our trial. They note that balance appears to be "task specific." Our balance exercise program was selected to train maintenance of position and postural adjustment as needed during daily activities. We did not attempt to train response to perturbations as we felt this would be unsafe in a home-based program. Although we agree that it is a well-accepted--albeit little-studied--principle that rehabilitation is "task specific," the question is: How specific is task specific? There obviously must be some generalization of training effect to related movements. Otherwise, training would be infeasible, as it is impossible to train--or even anticipate--every possible position or adjustment an individual will need during his or her daily activities.

Dr Hopp and Dr Svanborg restate our caveat that interrater reliability needs to be tested on a greater number of subjects by a large number of therapists. Indeed, one of the reasons for publishing the assessment and intervention protocol is to encourage just such investigations.

Dr Hopp and Dr Svanborg appropriately point out that the real test of this protocol is a reduction in falls and improvement in mobility. Our preliminary results suggest that, indeed, our assessment and intervention protocol results in a reduction in falls and improvement in balance.[7]

Marie Koch, PT Margaret Gottschalk, PT Sally Palumbo, PT Dorothy I Baker, PhD, RNC Mary E Tinetti, MD (*) The Hygenic Corp, 1245 Home Ave, Akron, OH 44310.

References

[1] Tinetti ME, Baker DI, Garrett PA, et al, Yale FICSIT: risk factor abatement strategy for fall prevention, J Am Geriatr Soc, 1993;41.315-320. [2] Agre JC, Pierce LE, Raab DM, et al. Light resistance and stretching exercise in elderly women: effect upon strength. Arch Phys Med Rehabil. 1988;69:273-276. [3] Robbins AS, Rubenstein LZ, Josephson KR, et al. Predictors of falls among elderly people. Arch Intern Med 1989;149:1628-1633. [4] Voss D, Ionta M, Myers B. Proptioceptive Neuromuscular Facilitation: Patterns and Techniques. 3rd ed. Philadelphia, Pa: Harper & Row; 1985. [5] Nevitt MC, Cummings SR, Hudes ES. Risk factors for injurious falls: a prospective study. J Gerontol. 1991;46:M164-M170. [6] Cummings SR, Nevitt MC. A hypothesis: the causes of hip fractures, J Gerontol. 1989;44: M107-M111. [7] Tinetti M, Baker DI, Koch MC, et al. Multiple risk factor fall prevention trial: one-year results. Clin Res. 1993;41:190A. Abstract.

M Koch, PT, is Assistant Professor and Academic Coordinator of Clinical Education, Department of Physical Therapy, Quinnipiac College, Hamden, CT 06518.

M Gottschalk, PT, is Staff Physical Therapist, Department of Rehabilitatibe Services, Yale-New Hav Hospital, New Haven, Ct 06504.

DI Baker, PhD, RNC, is Research Scientist, Yale University School of Nursing, New Haven, CT 06510.

S Palumbo, PT, is a Clinical Education Coordinator-Staff Physical Therapist, Department of Rehabil Services, Yale-New Haven Hospital.

ME Tinetti, MD, is Associate Professor of Medicine, Department of Internal Medicine, Yale Universi School of Medicine, 333 Cedar St, PO Box 3333, New Haven, CT 06510-8056 (USA). Address all correspondence to Dr Tinetti.

This study was approved by the Human Investigation Committee of Yale University School of Medicine.

This study was supported by grant #UO1 AG09087 from the National Institute on Aging.

This article was submitted February 19, 1993, and was accepted October 27, 1993.