By Mary Murphy, PhD, PT, NCS
Falls are a leading cause of injury, immobility, disability, nursing home placement, and premature death in the elderly.1 Approximately 30% of the community-based population over the age of 65 and 40% of the community-based population over the age of 80 will fall at least one time per year.2 Instability and falling are reported as predominate contributing factors in 40% of nursing home admissions3 and the rate of falling in institutional settings is 45% to 70%.4 Approximately 95% of injuries in long-term care facilities were due to falls.5 Falls that do not result in injury may lead to decreased mobility due to fear of falling, as 50% of repeated fallers have reported restricting their activities to avoid falls.6 Falls are associated with more than 300,000 hip fractures annually resulting in a 20% mortality rate after 1 year.7 Of the elderly fallers who require hospitalization, only 50% will be alive 1 year later.8 The direct care costs for elderly fall-related injuries were $20.2 billion in 1994.9 By 2020, the cost of fall injuries is projected to reach $32.4 billion.10 It is clear that fall prevention is imperative.
Recently, recommendations for routine care of older persons were outlined by the American Geriatrics Society Panel on Falls in Older Persons.6 They recommended asking about falls and if the person reported a single fall, a quick screen for difficulty with a simple functional task should be completed. Those who demonstrate difficulty or unsteadiness with this task or report recurrent falls should receive a detailed evaluation including assessment of fall circumstances, medications, risk factors and environmental risks, medical and mobility problems, neurologic function, and cardiovascular status.
Falls may occasionally result from a single factor but are likely due to the accumulated effects of multiple risk factors.11 Risk factors that have been associated with fall incidence include older age, cognitive impairment, visual impairment, disease, postural hypertension, medications, neurological or musculoskeletal disabilities, impaired balance and gait, and environmental hazards. A problem may not arise until several factors are involved. Tinetti et al reported that the chance of falling in the elderly increases linearly from 8% with no risk factors or disabilities to 78% with four or more risk factors or disabilities, which indicates a relationship between level of disability and fall frequency.12
PHYSICAL PERFORMANCE MEASURES AND IMPAIRMENT MEASURES
Physical performance measures are objective assessments of impairment or functional limitation, defined as restrictions of physical activity, rather than assessments at either the organ or disability level. Fried et al13 hypothesized the existence of a preclinical disability stage, which they described as a decrease in function and may be identified with functional physical performance tests. Researchers have described numerous valid objective performance measures consisting of either a battery of tests or a single measure, which identify fall risk or predict increased dependence and frailty.14-16 Guralnik et al developed a brief physical performance battery that can be administered in the home and strongly correlates with self-reported disability.15 The tests consist of side by side, semitandem, and tandem balance tests; chair stands; and a timed eight-foot walk. A relatively new single-item test of functional performance is a timed floor transfer during which a subject comes to a sitting position on the floor from standing, and then rises to standing as quickly as possible.17 Elders tend to overestimate their ability to perform this task. Tinetti et al4,11 documented that 47% of elders who did not sustain injury during a fall still could not get up off the floor and were more likely to suffer functional decline or die within the following year than the nonfaller group or those who could get up independently.18 Pilot studies have been conducted to develop and test a short screen protocol originally derived from numerous measures.17 Cut-off scores that had the highest combination of specificity and sensitivity were also derived for each individual performance variable to determine thresholds for fall risk.
In addition to high-level functional performance tasks, investigators have documented diminished lower extremity strength as well as balance and flexibility deficits in older adults and the elderly who have fallen.19,20 Impairment measures are objective assessments primarily focused at the organ level but may include functional components. Once impairment and physical performance deficits are identified in the elderly, appropriate interventions need to be designed and implemented.
HUMAN PERFORMANCE ENGINEERING MODEL
Previously, relationships among measures of impairment, functional performance, and disability in the elderly have been based on nonexperimental, predictive models such as logistic regression and multiple regression.12 The information from these studies is predicated on a group model and does not help the clinician in determining the patient’s specific limitations that impair normal functional performance.
Human performance engineering can play an important role by introducing the concepts of human resources and task demands as well as the interaction between the two.21 The General Systems Performance Theory (GSPT) framework is based on an engineering model that examines the architecture of the system providing the resources relative to the demands of the tasks in a top-down developmental method, whereas the medical community traditionally takes the bottom-up approach, beginning with organs and other elemental levels.
The Elemental Resource Model (ERM) has been developed from GSPT and applied to the human system. ERM characterizes specific human resources as they relate to high-level performance tasks and identifies the resources limiting performance for each subject. ERM has been designed to address the functional performance demands on any individual. It can be an efficient way to define and quantify those resources that appear to limit the ability of each subject to complete a high-level task. This concept is similar to functional capacity evaluations used in work injury management.
GSPT and the ERM examine the minimal levels of basic elemental resources (ER) required to perform a task as contrasted with the medical model of simply measuring the individual’s performance capacities, some of which may well exceed the requirements for task performance. Therefore, the most important feature of ERM is the threshold principle. The ability to complete a high-level performance task (such as a floor transfer) is dependent on the minimal level of the ER (strength, balance, and flexibility) required for the task. If a person has twice as much strength in the hip, knee, or ankle as that which is necessary to get up from the floor but does not have the hip, knee, or ankle flexibility to place the leg to get up from the floor, that high-level task cannot be performed due to the flexibility ER limitation. In order to perform such a high-level task, a set of unique demands is placed on several combined ER. Therefore, it is the amount of each ER necessary to perform a task that is the essential measurement, not the total amount that is available. Previously, studies of relationships between impairment measures and function or disability have focused on the availability of ER such as hip strength and not on whether the person has the minimal ER requirements to achieve the high-level task.
Nonlinear Causal Resource Analysis (NCRA) is the ERM’s quantitative methodology for task analysis and prediction of human performance in high-level tasks based on the threshold amounts of necessary ER. NCRA generates a set of curves called resource demand functions (RDF) that can present the least amount of a given ER needed for each level of execution of a high-level task. Limiting elemental resources (LER) are the ER that do not achieve a threshold level sufficient for the performance of the functional tasks. NCRA is employed to develop RDF curves for different high-level tasks and to determine the specific LER in which the individual subjects exhibit subthreshold levels. Subject-specific LER for each high-level task (timed floor transfer, timed 50-foot walk, and timed tandem stance) are identified from a group of measures consisting of strength, flexibility, and balance tests by identifying which resources are below the RDF value at the cut-off score for the high-level task. The identified LER are the focus of the individualiz-ed treatment.
In order to use the NCRA analytical model, each performance score must be adjusted so that the higher numeric score represents a higher level of performance. Therefore, on tests such as the timed floor transfer where less time to complete the task infers better function, the inverse of the score must be used in the analysis. Strength measures are normalized by body mass index (BMI). For each graph, the high-level performance task makes up the independent (x-axis) variable and the ER provides the dependent (y-axis) variable. For resources that have two end points, such as hip flexion and extension range of motion, each is graphed as a separate ER. Since each of the three high-level tasks (timed floor transfer, timed 50-foot walk, and timed tandem) is plotted against 43 potential limiting resources (ROM, strength, and balance measures), a total of 129 graphs are used to draw the RDF.
To target subject-specific deficiencies in these ER, an optimal combination of rehabilitative exercises are developed for each individual. One to three specific LER are identified for each of the high-level tasks, resulting in directed exercises for nine or fewer LER. It is possible that one LER may limit performance on more than one high-level task. Changes in performance following the individualized resource-specific exercise intervention are compared to determine the effect of the individualized exercise program.
By developing interventions directed at the LER, physical performance should improve and the number of future falls decreases. A series of different studies have been conducted to test this model on athletes22 and significant group improvements in both targeted LER and physical performance supported the use of individualized exercise intervention based on NCRA with community-dwelling elderly fallers.23 Trends may emerge in terms of ER identification for specific functional tasks, allowing us to suggest threshold levels of ER necessary for execution of the performance tasks at speeds that are indicative of high functional levels, such as the amount of hip extensor strength necessary to perform a floor transfer in under 13 seconds.
Clinicians may develop individualized interventions based on NCRA, targeting LER to improve physical performance, maximize functional independence, and prevent future falls. This model needs to be tested and validated with larger sample sizes, different ER and functional tasks, and elderly persons living in different settings. Future research efforts can be directed toward the establishment of absolute thresholds of ER that are essential for performance of functional activities required for independent living and interventions could then be targeted to each individual’s LER to maximize functional independence.
Mary Murphy, PhD, PT, NCS, is assistant professor, Division of Physical Therapy, Department of Rehabilitation Medicine, Emory University School of Medicine, Atlanta.
