A fall is one of the most common events threatening the independence of older persons. A fall is considered to have occurred when a person comes to rest inadvertently on the ground or a lower level. Most discussions in the literature of falls by older persons do not include falls associated with loss of consciousness (eg, syncope, seizure) or with overwhelming trauma. The majority of falls are not associated with syncope.
The incidence of falls increases with age and varies according to living status. Each year, between 30% and 40% of community-dwelling persons aged 65 years and older fall. Among those with a history of a fall in the previous year, the annual incidence of falls is close to 60%. In the long-term-care setting each year, about half of all persons fall.
Complications resulting from falls are the leading cause of death from injury in men and women aged 65 and older. The death rate attributable to falls increases with age, with white men aged 85 years and older having the highest death rate (> 180 deaths per 100,000 population). Most falls result in an injury of some type, usually minor soft-tissue injuries, such as bruises and scrapes; however, 10% to 15% result in fracture or other serious injury. In general, falls are associated with subsequent declines in functional status, greater likelihood of nursing-home placement, increased use of medical services, and the development of a fear of falling. Of those elderly persons who fall, only half are able to get up without help, thus experiencing the “long lie.” Long lies are associated with lasting declines in functional status.
The true cost of falls in health care dollars is difficult to ascertain. It has been estimated that in the United States the lifetime costs of fall-related injuries for persons aged 65 and older is $12.6 billion. Since many falls result in injury, there is a significant use of emergency department facilities among fallers. Studies from the early 1990s indicate that each year almost 8% of persons aged 70 and older go to emergency departments because of a fall-related injury, and close to a third of these people are admitted to the hospital for a median length of stay of 8 days.
Falls, incontinence, delirium, and other geriatric syndromes result from the accumulated effects of impairments in multiple domains. The falls of older people are rarely due to a single cause. Rather, there is often a complex interaction among factors intrinsic to the individual (age-related declines, chronic disease, acute illness, medications), challenges to postural control (environment, changing positions, normal activities), and mediating factors (risk-taking behaviors, underlying mobility level).
Multiple prospective cohort studies of risk factors for falls have been published over the past decade. Across these studies, many risk factors were found to be consistently associated with falls. These included age, cognitive impairment, female gender, past history of a fall, lower extremity weakness, gait problems, foot disorders, balance problems, hypovitaminosis D, psychotropic drug use, arthritis, and Parkinson’s disease. These studies differed significantly in the types of risk factors evaluated, the types of population studied (eg, past fall history was sometimes an entry criterion), and the outcome (one fall, two or more falls, injurious falls). The multiple risk factors found across the studies highlights the multifactorial nature of falls and suggests that there may also be unique circumstances surrounding falls that are not accounted for in the studies. In general, the risk of falling increases with the number of risk factors, although some persons with no risk factors experience falls.
Successful prevention of falls begins with a knowledge of the age-related changes that increase the risk of falls. Thus, with aging, there are declines in the visual, proprioceptive, and vestibular systems. For example, the visual system demonstrates reductions in visual acuity, depth perception, contrast sensitivity, and dark adaptation. The proprioceptive system loses sensitivity in the lower extremities. The vestibular system demonstrates a loss of labyrinthine hair cells, vestibular ganglion cells, and nerve fibers.
Despite these age-related changes in sensory systems, it has been difficult to quantify the age-related changes in postural control that are independent of disease. In general, when postural stability is tested in young and old persons with no apparent musculoskeletal or neurologic impairment, age-related differences in measured sway are found to be most pronounced when moderately severe perturbations of stance are administered, such as changing the support surface, changing body position, altering the visual input, or moving the support surface horizontally or rotationally. This occurs because these perturbations stress the redundancy of the sensory systems in their ability to maintain postural stability. In addition, there may be other age-related changes in the central nervous system that affect postural control, including the loss of neurons and dendrites, and the depletion of neurotransmitters, such as dopamine, within the basal ganglia.
Since it is difficult to find elderly persons without at least subtle neurologic findings, studies have been unable to determine whether some of the young-old differences may be due to these factors. Some of the most striking postural control differences between young and old persons relate to the order or grouping of muscle activation patterns. Thus, in response to perturbations of the support surface, older persons tend to activate the proximal muscles, such as the quadriceps, before the more distal muscles, such as the tibialis anterior. This strategy may not be an efficient way to maintain postural stability. Similarly, in the elderly person there may be greater co-contraction of antagonistic muscles, and the onset of the muscle activation and associated joint torque may be delayed. Finally, the ability to recover balance upon a postural disturbance may be compromised by an age-related decline in the ability to rapidly develop joint torque by using muscles of the lower extremity. All of these strategies potentially impair maintenance of upright posture.
Another important physiologic contributor to the successful maintenance of upright posture is the regulation of systemic blood pressure. The failure to perfuse the brain, which accompanies hypotension, increases the risk of a fall, usually in association with syncope. In addition to the age-related declines in baroreflex sensitivity to hypotensive stimuli manifested as a failure to cardio-accelerate, everyday stresses such as changing posture, eating a meal, or suffering an acute illness may result in hypotension. Since many elderly persons have a resting cerebral perfusion that is compromised by vascular disease, even slight reductions in blood pressure may produce cerebral ischemic symptoms, such as falls. Finally, with aging, there is a reduction in total body water, which places older persons at increased risk of dehydration with acute illness, diuretic use, or hot weather. Since there is a progressive decrease in basal and stimulated renin levels, as well as a decrease in aldosterone production with aging, dehydrating stresses may lead to orthostatic hypotension and a fall.
A number of age-related chronic conditions deserve special mention because of their association with fall risk. Parkinson’s disease, in particular, increases the risk of falls through several mechanisms, including the rigidity of lower extremity musculature, the inability to correct sway trajectory because of the slowness in initiating movement, hypotensive drug effects, and, in some cases, cognitive impairment. Another common disease contributing to falls is osteoarthritis. When present in the knee, osteoarthritis may affect mobility, the ability to step over objects and maneuver, and the tendency to avoid complete weight bearing on a painful joint.
One of the most easily modified risk factors for falls that has been repeatedly demonstrated in observational studies is medication use. Individual classes of medications, such as the benzodiazepines, antidepressants (including selective serotonin-reuptake inhibitors), and antipsychotic drugs, have been associated with an increased risk of hip fracture. An increased risk of falling has also been found to be associated with recent changes in the dose of a medication and the total number of prescriptions.
The relative importance of environmental factors to the risk of falling appears to be much less than intrinsic factors in the individual; however, the interaction between environmental factors and intrinsic factors has not been well quantified. Well-designed intervention studies have focused on improving the risk-factor profile of the person or have combined individual interventions with environmental manipulation, making it difficult to isolate the contributions of the environmental factors. Nevertheless, attention to safety hazards in the home environment would appear to be worthwhile, and one intervention study targeting environmental factors was successful in reducing falls.
The evaluation and management of falls in the geriatric patient may differ according to the clinical setting (home, hospital, nursing home). Table 28.1 highlights some of the differences that might be considered according to setting.
Many falls never come to clinical attention for a variety of reasons: The patient may never mention the event, there is no injury at the time of the fall, the clinician may fail to ask the patient about a history of falls, or the patient or the clinician may make the invalid assumption that falls are an inevitable part of the aging process. In institutional settings, despite attention to falls and even incident reporting, not all falls come to the attention of the nursing staff. The treatment of injuries resulting from falls commonly fails to include an investigation of the cause of the fall.
In the clinical evaluation of the noninstitutionalized geriatric patient who is not specifically being seen for a problem with falling, it is still important that an assessment of fall risk be integrated into the history and physical examination. (See Figure 28.1 for an overview of falls assessment and management in all older persons.) The most important point in the history is the previous history of a fall, since this is a strong risk factor for future falls. For patients presenting with a fall, important components of the history include the activity of the faller at the time of the incident, the occurrence of prodromal symptoms (lightheadedness, imbalance, dizziness), the location of the fall, and the time of the fall. Loss of consciousness is associated with injurious falls and should raise important considerations, such as orthostatic hypotension or cardiac or neurologic disease. (See Syncope.) Information on previous falls should be collected to identify patterns that may help target strategies to reduce risk factors. A complete medication history should focus specifically on the use of vasodilators, diuretics, and sedative hypnotics because these agents have been associated with increased risk of falls. In addition to inquiring about the circumstances surrounding the fall, the clinician taking the history should attempt to identify environmental factors that may have contributed. Thus, information on lighting, floor covering, door thresholds, railings, and furniture may add important clues.
The physical examination of the person who has fallen should focus on risk factors. Much of the examination duplicates that of a gait assessment (see Gait Impairment.) Footwear may also be an important factor to consider. In one small study to test the effect of various shoe types on balance in older men, shoes with thin, hard soles were found to produce the best results, even though they were perceived as less comfortable than thick, soft, mid-soled shoes, such as running shoes.
Probably the most important part of the physical examination is an assessment of integrated musculoskeletal function, which can be accomplished by performing one or more of the following tests of postural stability.
A simple maneuver called the functional reach test is a practical way to test the integrated neuromuscular base of support and has predictive validity for falls by elderly men. This test is performed with a leveled yardstick secured to a wall at the height of the acromion. The person being tested assumes a comfortable stance without shoes or socks and stands so that his or her shoulders are perpendicular to the yardstick. He or she makes a fist and extends the arm forward as far as possible along the wall without taking a step or losing balance. The total reach is measured along the yardstick and recorded. Inability to reach 6 inches or more is cause for concern and merits further evaluation. In its initial description, the functional reach correlated with other physical performance measures, such as walking speed (r = 0.71), tandem walk using an ordinal scale (r = 0.67), and standing on one foot measured as number of seconds that a one-footed stance could be maintained (r = 0.64).
Another useful test of integrated strength and balance is the Get Up and Go test, which can be performed with or without timing. It consists of observation of an individual standing up from a chair without using the arms to push against the chair, walking across a room, turning around, walking back, and sitting down without using the arms. This test can demonstrate muscle weakness, balance problems, and gait abnormalities.
There is no standard diagnostic evaluation of a person with a history of falls or with a high risk of falling. Obviously, laboratory tests for hemoglobin, serum urea nitrogen, creatinine, or glucose levels can help to exclude such causes of falling as anemia, dehydration, or hyperglycemia with hyperosmolar dehydration. There is no proven value of routinely performing Holter monitoring of persons who have fallen. Because data demonstrate that carotid sinus hypersensitivity contributes to falls and even hip fracture, some have advocated performing carotid sinus massage with continuous heart rate and phasic blood-pressure measurement in persons with unexplained falls. Similarly, the decision to perform echocardiography, brain imaging, or radiographic studies of the spine should be driven by the findings of the history and physical examination. Echocardiography should be reserved for those with cardiac conditions believed to contribute to the maintenance of blood flow to the brain. Spine radiographs or magnetic resonance imaging may be useful in patients with gait disorders, abnormalities on neurologic examination, lower extremity spasticity, or hyperreflexia to exclude cervical spondylosis or lumbar stenosis as a cause of falls.
Multiple studies of preventive interventions have been conducted over the past decade, including programs to improve strength or balance, educational programs, optimization of medications, and environmental modifications in homes or institutions. Some interventions have targeted single risk factors; others have attempted to address multiple factors.
A Cochrane Collaboration systematic review of interventions to reduce the incidence of falling in elderly persons was performed. This review considered only studies that included elderly persons of either sex who were living in the community or in institutional care, randomized to an intervention versus control or into one of two interventions. As of the February 2005 update of this systematic review, 62 individual trials meeting the inclusion criteria were identified. Of the 62 studies, 47 reported the effect of interventions in persons living in the community, eight were set in long-term-care institutions or in nursing homes, and four studies were in rehabilitation or geriatric assessment wards in hospitals. Three further studies included participants with specific conditions from a range of residential settings. Interventions were grouped into ten groups:
Results from this systematic review suggested interventions that were likely to be beneficial:
In contrast, many interventions were reviewed as being of unknown effectiveness:
More data are needed to confirm whether strategies apparently effective in significantly reducing the number of older people sustaining falls are also effective in reducing more serious sequelae of falls, such as fractures or other injuries.
Interventions that target risk factors for falls in three domains—medications, mobility, and medical conditions—constitute a practical approach for clinicians who are treating older patients with a high risk for or a history of falling (see Table 28.2).
Many geriatricians are beginning to consider the use of hip protectors to reduce fall-related hip fractures. Systematic reviews of the topic have concluded that there is no evidence that hip protectors are effective in reducing hip fractures in studies in which randomization was by individual patient within an institution or among patients living at home. Earlier studies that found benefits with hip protectors randomly assigned groups of patients to an intervention based on setting (such as by ward in a nursing home) and were potentially susceptible to bias on this basis; more recent studies have randomized individual patients, which is less susceptible to bias. This suggests that there may be unintended “co-interventions” that occur when whole units or facilities participate in trials and are allocated to use hip protectors, and that these co-interventions may benefit residents of these facilities. In addition, the apparent lack of efficacy of hip protectors may be due in part to low adherence rates.
At least a dozen types of hip protectors are commercially available. Many of these hip protectors have not been tested in either the laboratory or in clinical trials, yet they are being promoted by manufacturers. Despite these concerns, until the results from studies with higher rates of adherence are available, it is not unreasonable for clinicians to consider the use of hip protectors in patients at high risk of hip fractures who are willing to use them. Most of the published trials used Safe Hip.
For elderly patients who have sustained a fall, a multifactorial approach that is based on data about risk factors and a multidimensional assessment of the patient and that targets interventions on the basis of these findings is appropriate. For elderly persons who have no history of falling, it is reasonable to use traditional multidimensional geriatric assessment with targeted interventions as risk factors are identified. For a summary of the recommendations of the expert panel on falls prevention assembled by the American Geriatrics Society, the British Geriatrics Society, and the American Academy of Orthopaedic Surgeons, see the Appendix.
■ American Geriatrics Society, British Geriatrics Society, and American Academy of Orthopaedic Surgeons Panel on Falls Prevention. Guideline for the prevention of falls in older persons. J Am Geriatr Soc. 2001;49(5):664–672.
The aim of this publication is to guide clinicians in assessing fall risk and managing older patients at risk of falling or who have fallen. A review of the literature was performed to allow recommendations to be evidence based, wherever possible. Grades, based on the quality of the evidence and potential effect of the intervention, are provided for each recommendation. The recommendations cover the following areas: Approach to assessing older persons as part of routine care, not presenting with a fall; approach to assessing older persons with one or more falls, or who report recurrent falls or abnormalities of gait or balance; multifactorial interventions for the management of falls by older persons in various settings; and single interventions for managing falls by older persons. See the Appendix, for a summary of specific recommendations of the panel.
■ Bischoff-Ferrari HA, Dawson-Hughes B, Willett WC, et al. Effect of Vitamin D on falls: a meta-analysis. JAMA. 2004;291(16):1999–2006.
Vitamin D supplementation appears to reduce the risk of falls among ambulatory or institutionalized older individuals with stable health by more than 20%.
■ Fletcher PC, Hirdes JP. Risk factors for falling among community-based seniors using home care services. J Gerontol A Biol Sci Med Sci. 2002;57(8):M504–M510.
This study and the Kron study (see below) focus on the use of easy-to-administer instruments to assess fall risk, one for nursing-home residents, and one for seniors using home care. There are no surprising results here, but the sample sizes are quite large and the findings quite compatible with previously recognized fall risk factors. Application of these instruments to the appropriate populations might better operationalize the targeting of interventions to those at greatest risk of falls.
■ Gillespie L, Gillespie W, Robertson M, et al. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev. 2003;4:CD00340; 2005 update.
The Cochrane Library of systematic review is a valuable resource for evaluating clinical trials of various therapeutic interventions. Since few such reviews specifically focus on older persons, it is particularly notable that this systematic review, first published in 1995, has now been updated to 2003. This most recent update reviews the growing number of clinical trials in the area of falls prevention. The results of this systematic review demonstrated which interventions are likely to be beneficial: multidisciplinary, multifactorial, health and environmental risk-factor screening or intervention programs; Tai Chi, muscle strengthening, and balance training by trained professionals; home hazard assessment and modification that is professionally prescribed for elderly persons with a history of falling; and withdrawal of psychotropic medication.
■ Kron M, Loy S, Sturm E, et al. Risk indicators for falls in institutionalized frail elderly. Am J Epidemiol. 2003;158(7):645–653.
This is a 1-year prospective observational study of risk factors for falls in a sample of institutional frail elderly persons residing in German long-term-care facilities. The value of the study is that the authors used variables derived from the Minimum Data Set, thus permitting clinical application to nursing-home assessments commonly performed in the United States. In addition, the study identified five modifiable risks (urinary incontinence, cognitive impairment, use of restraints, depression, and transfer difficulties) that might be targeted to reduce the risk of falls in the nursing home.
■ Parker MJ, Gillespie LD, Gillespie WJ. Hip protectors for preventing hip fractures in the elderly. Cochrane Database Syst Rev. 2004;3:CD001255.
This outstanding systematic review from the Cochrane Collaboration highlights the controversy surrounding the randomized trials of hip protectors. The finding that hip protectors conferred a benefit in reducing hip fractures when considering studies that randomized facilities or units within facilities but not when considering studies that randomized individuals suggests that the hip protectors studied may not be effective. Alternatively, by being in a study, seniors wearing hip protectors may be “protected” by other safeguards used by staff within a facility housing seniors who are wearing these garments.
■ Robertson MC, Devlin N, Gardner MM, et al. Effectiveness of economic evaluation of a nurse delivered home exercise programme to prevent falls. 1: randomised controlled trial. BMJ. 2001;322(7288):697–701.
The authors used a randomized controlled trial to study the impact of a nurse-delivered home exercise program on 240 men and women aged 75 years and older. Outcome measures included number of falls, number of resultant injuries, program costs, and hospital costs associated with falls. Total falls, serious injuries, and hospitalizations due to falls were all reduced. Cost-effectiveness was demonstrated for participants 80 years and older.
Douglas P. Kiel, MD, MPH