Falls in the Nursing Home

Epidemiology

Both the incidence of falls in older adults and the severity of complications increase steadily with age and increased physical disability. Accidents are the fifth leading cause of death in older adults, and falls constitute two thirds of these accidental deaths. About three fourths of deaths caused by falls in the United States occur in the 13% of the population aged 65 years and older [1, 2]. Approximately one third of older adults living at home will fall each year, and about 5% will sustain a fracture or require hospitalization. The incidence of falls and fall-related injuries among persons living in institutions has been reported in numerous epidemiologic studies [3-18]. These data are presented in Table 1. The mean fall incidence calculated from these studies is about three times the rate for community-living elderly persons (mean, 1.5 falls/bed per year), caused both by the more frail nature of persons living in institutions and by more accurate reporting of falls in institutions.

As shown in Table 1, only about 4% of falls (range, 1% to 10%) result in fractures, whereas other serious injuries such as head trauma, soft-tissue injuries, and severe lacerations occur in about 11% of falls (range, 1% to 36%). However, once injured, an elderly person who has fallen has a much higher case fatality rate than does a younger person who has fallen [1, 2]. Each year, about 1800 fatal falls occur in nursing homes. Among persons 85 years and older, 1 of 5 fatal falls occurs in a nursing home [19]. Nursing home residents also have a disproportionately high incidence of hip fracture and have been shown to have higher mortality rates after hip fracture than community-living elderly persons [20]. Furthermore, because of the high frequency of recurrent falls in nursing homes, the likelihood of sustaining an injurious fall is substantial.

In addition to injuries, falls can have serious consequences for physical functioning and quality of life. Loss of function can result from both fracture-related disability and self-imposed functional limitations caused by fear of falling and the “postfall anxiety syndrome.” Decreased confidence in the ability to ambulate safely can lead to further functional decline, depression, feelings of helplessness, and social isolation. In addition, the use of physical or chemical restraints by institutional staff to prevent high-risk persons from falling also has negative effects on functioning.

Causes of Falls

The major reported immediate causes of falls and their relative frequencies as described in four detailed studies of nursing home populations [14, 15, 17, 21] are presented in Table 2. The Table also contains a comparison column of causes of falls among elderly persons not living in institutions as summarized from seven detailed studies [21-28]. The distribution of causes clearly differs among the populations studied. Frail, high-risk persons living in institutions tend to have a higher incidence of falls caused by gait disorders, weakness, dizziness, and confusion, whereas the falls of community-living persons are more related to their environment.

In the nursing home, weakness and gait problems were the most common causes of falls, accounting for about a quarter of reported cases. Studies have reported that the prevalence of detectable lower-extremity weakness ranges from 48% among community-living older persons [29] to 57% among residents of an intermediate-care facility [30] to more than 80% of residents of a skilled nursing facility [27]. Gait disorders affect 20% to 50% of elderly persons [31], and nearly three quarters of nursing home residents require assistance with ambulation or cannot ambulate [32]. Investigators of case–control studies in nursing homes have reported that more than two thirds of persons who have fallen have substantial gait disorders, a prevalence 2.4 to 4.8 times higher than the prevalence among persons who have not fallen [27, 30].

The cause of muscle weakness and gait problems is multifactorial. Aging introduces physical changes that affect strength and gait. On average, healthy older persons score 20% to 40% lower on strength tests than young adults [33], and, among chronically ill nursing home residents, strength is considerably less than that. Much of the weakness seen in the nursing home stems from deconditioning due to prolonged bedrest or limited physical activity and chronic debilitating medical conditions such as heart failure, stroke, or pulmonary disease. Aging is also associated with other deteriorations that impair gait, including increased postural sway; decreased gait velocity, stride length, and step height; prolonged reaction time; and decreased visual acuity and depth perception. Gait problems can also stem from dysfunction of the nervous, musculoskeletal, circulatory, or respiratory systems, as well as from simple deconditioning after a period of inactivity.

Dizziness is commonly reported by elderly persons who have fallen and was the attributed cause in 25% of reported nursing home falls. This symptom is often difficult to evaluate because “dizziness” means different things to different people and has diverse causes. True vertigo, a sensation of rotational movement, may indicate a disorder of the vestibular apparatus such as benign positional vertigo, acute labyrinthitis, or Meniere disease. Symptoms described as “imbalance on walking” often reflect a gait disorder. Many residents describe a vague light-headedness that may reflect cardiovascular problems, hyperventilation, orthostatic hypotension, drug side effect, anxiety, or depression.

Accidents, or falls stemming from environmental hazards, are a major cause of reported falls—16% of nursing home falls and 41% of community falls. However, the circumstances of accidents are difficult to verify, and many falls in this category may actually stem from interactions between environmental hazards or hazardous activities and increased individual susceptibility to hazards because of aging and disease. Among impaired residents, even normal activities of daily living might be considered hazardous if they are done without assistance or modification. Factors such as decreased lower-extremity strength, poor posture control, and decreased step height all interact to impair the ability to avoid a fall after an unexpected trip or while reaching or bending. Age-associated impairments of vision, hearing, and memory also tend to increase the number of trips. Studies have shown that most falls in nursing homes occurred during transferring from a bed, chair, or wheelchair [3, 11]. Attempting to move to or from the bathroom and nocturia (which necessitates frequent trips to the bathroom) have also been reported to be associated with falls [34, 35] and fall-related fractures [9]. Environmental hazards that frequently contribute to these falls include wet floors caused by episodes of incontinence, poor lighting, bedrails, and improper bed height. Falls have also been reported to increase when nurse staffing is low, such as during breaks and at shift changes [4, 7, 9, 13], presumably because of lack of staff supervision.

Confusion and cognitive impairment are frequently cited causes of falls and may reflect an underlying systemic or metabolic process (for example, electrolyte imbalance or fever). Dementia can increase the number of falls by impairing judgment, visual-spatial perception, and ability to orient oneself geographically. Falls also occur when residents with dementia wander, attempt to get out of wheelchairs, or climb over bed siderails.

Orthostatic (postural) hypotension, usually defined as a decrease of 20 mm or more of systolic blood pressure after standing, has a 5% to 25% prevalence among “normal” elderly persons living at home [36]. It is even more common among persons with certain predisposing risk factors, including autonomic dysfunction, hypovolemia, low cardiac output, parkinsonism, metabolic and endocrine disorders, and medications (particularly sedatives, antihypertensives, vasodilators, and antidepressants) [37]. The orthostatic drop may be more pronounced on arising in the morning because the baroreflex response is diminished after prolonged recumbency, as it is after meals and after ingestion of nitroglycerin [38, 39]. Yet, despite its high prevalence, orthostatic hypotension infrequently causes falls, particularly outside of institutions. This is perhaps because of its transient nature, which makes it difficult to detect after the fall, or because most persons with orthostatic hypotension feel light-headed and will deliberately find a seat rather than fall.

Drop attacks are defined as sudden falls without loss of consciousness and without dizziness, often precipitated by a sudden change in head position. This syndrome has been attributed to transient vertebrobasilar insufficiency, although it is probably caused by more diverse pathophysiologic mechanisms. Although early descriptions of geriatric falls identified drop attacks as a substantial cause, more recent studies have reported a smaller proportion of persons who fall because of drop attacks. This probably reflects greater precision in diagnosis and a correction of an earlier tendency to attribute falls of unclear cause to drop attacks [40]. Only one of the four nursing home studies we reviewed documented drop attack as a cause of falls.

Syncope, defined as a sudden loss of consciousness with spontaneous recovery, is a serious but relatively uncommon cause of falls that usually results from decreased cerebral blood flow or occasionally from metabolic disorders such as hypoglycemia or hypoxia. The most frequent causes in elderly persons are vasovagal reactions, cardiac arrhythmias, and orthostatic hypotension, although, in many cases, a clear cause is never determined [41]. A history of syncope may be difficult to obtain because many residents do not remember exactly what occurred during the fall and because the resident may confuse drop attacks or dizziness with syncope. Many studies of falls do not identify syncope as causing a fall but rather as a separate syndrome. This perhaps explains the extremely low proportion of falls from syncope (see Table 2).

Other important causes of falls include visual problems, arthritis, acute illnesses, disorders of the central nervous system, drug side effects, and alcohol intake. Diseases of the central nervous system (for example, cerebrovascular disease, dementia, normal pressure hydrocephalus, and parkinsonism) often result in falls by causing dizziness, orthostatic hypotension, and gait disorders. Residents with cognitive deficits often cannot recognize and avoid hazards. Drugs frequently have side effects that result in impaired thinking, stability, and gait. Especially important are agents with sedative, antidepressant, and antihypertensive effects, particularly diuretics, vasodilators, and β-blockers [42, 43]. Alcohol may be an occult cause of instability, falls, and serious injury, although it is less likely to be a problem in the nursing home than in the community. Other less common causes of falls include seizures, anemia, hypothyroidism, unstable joints, foot problems, and severe osteoporosis with spontaneous fracture.

Risk Factors for Falls

Studies have shown that among persons living in institutions, specific risk factors exist that significantly increase the likelihood of falling. Some of these risk factors are only temporary, whereas others may be related to a chronic condition. Risk factors that have been identified in case–control studies are presented in Table 3, along with an approximate mean relative risk of each factor. Of the 10 studies reviewed, 6 included physical examinations or tests for persons who have and have not fallen [21, 27, 30, 39, 44, 45], and the other 4 studies relied on data abstracted from medical chart reviews [42, 43, 46, 47].

Comparison of persons who have fallen with those who have not showed that lower-extremity weakness was a significant risk factor in all four of the studies that looked for it and increased the odds of falling about sixfold. Gait and balance impairments were found to be a significant risk factor associated with about a four- and five-fold increased risk for falling, respectively. Thus, gait and balance impairments are both the most important immediate causes and the most serious risk factors for falls in the nursing home.

Functional impairment has also been shown to be a significant risk factor for falls. Inability to do basic activities of daily living and the use of an assistive device were both associated with a threefold increased risk for falling. (Use of an assistive device is presumably a marker for impaired gait because when well-fitted, assistive devices should improve safety.) Visual impairment, postural hypotension, and cognitive impairment approximately doubled the risk for falls.

Several case–control studies have shown an association between falls and medication use, with reported odds ratios ranging from 1.0 to 5.7. In one study, residents taking four or more prescription medications had a significantly greater risk for falling [27]. Specific classes of medications found to increase the risk for falling in nursing home residents include psychotropic drugs [42, 43, 48], sedatives [17, 42, 47], cardiac drugs [42, 43, 46, 47], and nonsteroidal anti-inflammatory drugs [42, 43].

A weak relation exists between medical diagnoses and falls. In two studies, persons who have fallen had significantly more established medical diagnoses than persons who have not fallen [27, 46]. Specific medical diagnoses shown to be significant risk factors for falls include arthritis [42, 43] and depression [42].

Appreciating the interaction and probable synergism among multiple risk factors is important in making a clinical assessment. Several studies have shown that the risk for falling increases dramatically as the number of risk factors increases [27, 30, 42]. In one study, the predicted 1-year risk for falling ranged from 12% for persons with none of three risk factors (hip weakness assessed manually, unstable balance, and taking four or more prescribed medications) to 100% for persons with all three risk factors [27].

Most falls, even among frail elderly persons, do not result in serious injury. This observation has motivated several research groups to attempt to identify patients and specific risk factors associated with injurious falls. On the whole, the risk factors identified for injurious falls are the same as those for falls in general (that is, lower-extremity weakness, impaired balance, dizziness, poor vision and hearing, medications, and disorientation) [17, 18, 49, 50]. However, several additional risk factors among nursing home populations are particularly noteworthy. These include the female sex (probably largely related to osteoporosis), functional independence, the number of falls, and use of mechanical restraints. In one study [18], the risk for a fall-related injury was 10 times greater for residents who had been restrained at some time during a 1-year study period than that of those who had not been restrained. Both falls and serious injuries significantly increased as restraint use increased. This observation challenges the assumption that restraints protect patients from falls. The authors suggest that intermittent use of restraints in frail older persons may result in further deconditioning and actually increase the risk for falling.

In summary, it is possible to identify persons at substantially increased risk for sustaining a fall or fall-related injury by detecting the presence of selected risk factors.

Diagnostic Approach

The exact cause of a fall is frequently difficult to determine because most nursing home residents have more than one identifiable age-related change or medical condition that could constitute a risk factor for falling; frequently, the event is not witnessed or is poorly recalled. The diagnostic assessment of the resident who falls should focus on identifying risk factors that can be treated to reduce the likelihood of future falls. The basic components of this diagnostic approach are outlined in Table 4.

After any immediate injuries caused by the fall are treated, the diagnostic approach should begin with a well-directed history aimed at uncovering any specific cause and associated risk factors. Obtaining a full report of the circumstances and symptoms surrounding the fall is crucial; however, because the patient may poorly remember these events, reports from any witnesses are often important. Historical factors that can point to a specific cause or help to narrow the differential diagnosis include suddenly rising from a lying or sitting position (orthostatic hypotension), a trip or slip (gait, balance, or vision disturbance or an environmental hazard), antecedent cough or urination (reflex hypotension), a recent meal (postprandial hypotension), looking up or sideways (arterial or carotid sinus compression), and loss of consciousness (syncope or seizure).

Symptoms experienced near the time of falling may also indicate a potential cause—dizziness or giddiness (orthostatic hypotension, vestibular problem, hypoglycemia, arrhythmia, or a drug side effect), palpitations (arrhythmia), incontinence or tongue biting (seizure), asymmetric weakness (cerebrovascular disease), chest pain (myocardial infarction or coronary insufficiency), or loss of consciousness (any cause of syncope). Medications, especially those with hypotensive or psychoactive effects, and existence of concomitant medical problems may be important contributing factors (Table 4). Other points to elicit during the history that might be helpful in directing further work-up and care planning include the following: How long was the patient on the ground? What effect did the fall have on patient confidence, fear of further falls, and activity? Are there any effects on caregiver expectations, fears, and plans for future activities?

Pertinent items to be noted during the physical examination after the fall include orthostatic changes in pulse and blood pressure and presence of arrhythmias, carotid bruits, nystagmus, focal neurologic signs, confusion, musculoskeletal abnormalities, sensory impairments, and gait disturbances. It is often useful to attempt, under carefully monitored conditions, to reproduce the circumstances that might have precipitated the fall, such as changes in position, head turning, urination, or carotid pressure.

A careful assessment of gait and balance done as part of the assessment after the fall can provide meaningful information about impairments and functional abilities. Gait and balance should be assessed by close observation of how the patient rises from a chair, stands with eyes open and closed, walks, turns, and sits down. One should particularly note gait velocity and rhythm, stride length, symmetry, weakness, double-limb support time (the time spent with both feet on the floor), height of stepping, use of assistive devices, and degree of sway. Imbalance observed during head turning or flexion is an important finding associated with vestibular or vertebrobasilar pathologic conditions and an increased risk for falling. A useful clinical screening test for gait and balance described by Tinetti [51] provides a reproducible score helpful in detecting abnormalities, quantifying severity, localizing the type of problem, and serving as a baseline measure for assessing change over time.

The clinician should also assess the patient's mobility level and degree of independence in doing activities of daily living to realistically appraise the risk for falling. Patients with risk factors who engage in little independent physical activity may have a lower risk for falling than patients with risk factors who are ambulatory and try to do activities of daily living independently. If restraints are being used, the clinician should carefully review the rationale for their use and discuss alternative strategies with the resident and staff.

Laboratory evaluation should be considered on an individual patient basis and could include several key tests when the cause is not obvious: complete blood counts to search for anemia or infection; tests of serum sodium, potassium, glucose, and creatinine levels; electrocardiography to document arrhythmia; and thyroid function tests because occult thyroid disease may be difficult to diagnose clinically. Even then, the clinical evaluation and initial laboratory tests may not detect an intermittent problem that may have caused the fall (orthostatic changes, arrhythmias, and electrolyte disturbances, for example). More extensive testing may be indicated depending on the causes entertained after the initial history and physical examination.

An ambulatory cardiac (Holter) monitor is advisable when a transient arrhythmia is suspected by history, in cases of syncope, or when the patient with unexplained falls has a history of cardiac disease and receives cardiac medication; however, it should not be part of the routine evaluation after the fall. The likelihood of finding fall-relevant abnormalities on Holter monitoring in elderly residents, regardless of fall status, is particularly high. In one study done among nursing home residents, 82% of persons who had fallen and those who had not fallen had ventricular arrhythmias documented on Holter monitoring, and 100% had supraventricular arrhythmias [52]; none were symptomatic during the Holter monitoring. Because transient arrhythmias are so prevalent among the elderly, it is often unclear whether a monitored abnormality is related to the fall unless corresponding symptoms are noted during the monitoring process.

The value of assessing a patient after a fall has been reported by two study groups [14, 53]. In a descriptive study, an assessment done by a multidisciplinary falls consultation team in a nursing home uncovered new diagnoses and led to a reduction in the number of future falls [53]. In a randomized trial of an intervention involving assessment after a fall [14], nursing home residents who received a comprehensive assessment after they fell by a nurse practitioner experienced significant reductions in the number of hospitalizations (26%) and days spent in the hospital (52%) over 2 years compared with controls. Although patients receiving intervention were found to have lower fall and mortality rates (9% fewer falls and 17% fewer deaths), these trends did not reach statistical significance. This study strongly suggests that falls are markers of underlying disorders easily identifiable by a careful assessment after the fall, which in turn can be treated with reduction in disability.

This type of assessment to detect risk factors for falls has also been frequently recommended for inclusion in the periodic physical examination of frail nursing home residents [27, 30]. Despite the logic and appeal of such an approach, the efficacy of screening residents for these risk factors on admission or periodically as a fall prevention strategy remains to be proven.

Therapy and Prevention

Once the clinician has uncovered the possible cause or causes of the fall and additional risk factors, the most challenging aspect of the fall evaluation process begins—prescribing effective treatments and interventions to prevent future falls. Because of the multifactorial nature of falls, it should be expected that there is no standard approach to treatment and prevention. The clinician must develop an individual plan for each patient, considering intrinsic risk factors, the patient's functional level, and how treatment will affect quality of life. When a fall is caused by an obvious acute problem, treatment may be relatively simple, direct, and effective (for example, discontinuing therapy with medication that causes postural hypotension). However, most patients fall because of chronic interacting conditions, and treatment will require a combination of medical, rehabilitative, environmental, and behavioral intervention strategies (for example, treating an underlying cause of syncope, lowering bed height, and advising a patient to wear safe footwear). In other cases, the intervention required to prevent falls (for example, limiting ambulation and using restraints) may be more detrimental to the patient than a possible fall.

The types of treatment interventions for major risk factors for falls that have been proposed or studied are discussed below. Although clinical and empirical evidence suggests that risk factors for falls can be treated, few data are available to confirm that such treatment interventions actually prevent falls. The ability to establish the efficacy of fall prevention interventions is undoubtedly limited by the fact that falls have multiple confounding causes that require many different interventions. Patient compliance with interventions, especially among frail and cognitively impaired patients, is also a potential problem. In addition, because many falls are probably not preventable, studies need a large sample size to show an effect. Another consideration complicating the formulation of fall prevention strategies is the two-edged effect of physical activity on falls. Activity is and should be encouraged as a positive goal that can lead to higher function and quality of life; however, activity also facilitates the opportunity for falling and injuries. Although not well studied, active persons may have more falls overall but may also have fewer falls per unit of activity. These interactions among falls, activity levels, frailty, and injury need to be studied much more carefully. Currently, many large randomized trials are being done to study fall prevention strategies for nursing home and community-dwelling older persons [54], and these should provide some additional results that can be translated into practice guidelines. Until that time, identification of risk factors, especially those associated with injurious falls, and focused treatment interventions appear to be the most sensible approach to fall prevention.

Weakness and Impaired Function

As noted previously, muscle weakness and functional impairments are important risk factors for falls and are often interrelated. Treatment should focus on improving strength and endurance, which should improve functional ability. Unfortunately, rehabilitation efforts in the nursing home tend to focus on treating conditions after an acute event (for example, stroke and hip fracture) rather than on trying to maintain or improve strength and function. Several studies of exercise interventions to improve muscle strength have been done with mixed results in persons living in institutions. In one uncontrolled study of an 8-week exercise program among frail nursing home residents with a mean age of 90 years, remarkable increases in muscle strength, a 9% increase in the mid-thigh muscle area, and a 48% increase in mean tandem gait speed were seen [55]. In addition, two patients no longer needed an assistive device for ambulation. In a randomized trial, frail nursing home residents received 4 months of physical therapy or friendly visits. Although the rates of falls did not decrease, the experimental group did show a 15% improvement in mobility and a reduced use of assistive devices and wheelchairs [56]. Noteworthy in both studies is that mobility improved, which suggests that patients were more functionally independent.

In addition to physical therapy or exercise training programs, simple walking programs may also improve strength and function. Residents should be encouraged to be as physically active as possible, even if that only consists of walking with assistance a few minutes each day, as long as it can be done with reasonable safety.

Gait and Balance Impairments

Residents with gait and balance disturbances should be evaluated for the underlying process. This can usually be categorized after examination into problems with strength, sensation, pain, joint mobility, spasticity, or central processing or commonly a combination of these [57]. Treatment approaches should be tailored accordingly but generally involve programs of gait training, specific exercises, and prescription of assistive devices. Gait training, usually under the supervision of a physical therapist, can be particularly helpful for persons with stroke, hip fracture, arthritis, or parkinsonism. Exercise interventions for weakness have been discussed above. Assistive devices include walkers, crutches, canes, orthotics, and shoe modification. Because the assistive device must be tailored individually, it should be prescribed in consultation with a physiatrist or physical therapist. Residents should be referred to a podiatrist for foot problems such as toe deformities and calluses.

Dizziness Syndromes

As with gait and balance problems, treatment approaches to dizziness depend on the underlying cause, whether it be hypofusion (from such things as hypotension, arrhythmias, or local ischemia), vestibular problems, drug effects, or other less common disorders. A thorough evaluation to determine the cause is thus imperative [58], and useful individualized treatments can usually be devised. For example, patients with benign positional vertigo may benefit from learning desensitization exercises [59]. Antihistamines and antiemetic agents, which are often prescribed to treat vertigo, should be used with caution because of their tendency to sedate or cause confusion in older persons.

Postural Hypotension

As noted previously, postural hypotension can also stem from many causes. Hypotension caused by hypovolemia or drugs has obvious therapeutic approaches. More difficult is treating postural hypotension that is caused by autonomic dysfunction, which is particularly common among patients with diabetes and parkinsonism. Several techniques may help residents with persistent orthostatic hypotension caused by autonomic dysfunction: sleeping in a bed with the head raised to minimize a sudden decrease in blood pressure on rising; wearing elastic stockings to minimize venous pooling in the legs; rising slowly or sitting on the side of the bed for several minutes before standing up; and avoiding heavy meals and activity in hot weather. If such conservative measures are ineffective, the volume of circulating blood can be increased by liberalizing dietary salt, provided that associated medical conditions are not a contraindication. Rarely, if disabling postural hypotension persists, mineralocorticoid therapy can be initiated with low doses of fludrocortisone acetate (Florinef; Apothecon, Princeton, New Jersey); a dose of 0.1 to 0.2 mg/d will usually suffice in elderly patients. Extreme caution must be used to avoid precipitating congestive heart failure, fluid overload, hypokalemia, and hypertension.

Polypharmacy and Inappropriate Medications

The goal of the clinician should be to reduce the number of medications taken by the resident. When choosing medications, the clinician should consider drugs that are least associated with postural hypotension, are short-acting, and are less likely to have sedative effects. Numerous interventions to improve prescribing practices and reduce the incidence of polypharmacy in nursing homes have been studied [60]. In most of these studies, the intervention has consisted of a consultant pharmacist reviewing medication orders and recommending treatment to physicians and nurses. Positive outcomes shown by these interventions include reductions in the number of drugs prescribed per patient and the number of doses taken per patient; however, a corresponding reduction in the number of falls has not been documented specifically.

Although we have focused on specific interventions for individual residents, fall prevention strategies have also been recommended and tested at the institutional level. These include environmental assessment, nursing interventions, and technological devices.

Environmental Assessment

The environmental assessment identifies and removes potential hazards (for example, clutter, poor lighting, and uneven floor surfaces) and modifies the environment to improve mobility and safety (for example, installation of grab bars and raised toilet seats and the lowering of the bed height). Specific environmental interventions should include the following: adequate lighting in all hallways and stairwells, bathroom grab bars next to the toilet and in the tubs or shower, nonskid mats in tubs or shower, raised toilet seats, handrails in the hallways, secure stairway banisters, and furniture that is easy to rise from. Of special importance is bed height. Most beds are adjustable and are often in an inappropriately raised position for the convenience of the staff. Proper bed height is such that when the patient sits on the side of the bed with feet touching the floor, the knees are bent at a 90-degree angle. Furniture can also be rearranged to support an unstable patient for ambulation to the bathroom. Table 5 lists the most important items to include in an environmental assessment, which are based on published safety checklists and guidelines [61]. Although not usually considered a part of the environment, proper footwear is important for safety. Ill-fitting shoes, shoes with worn soles and heels, heels that are too high or too narrow, or shoes left untied or unbuckled are unsafe. The habit of many nursing home residents of wearing slippers without soles or backs or wearing stockings without shoes is particularly hazardous. A study that examined the stability of older men when barefoot and when wearing different types of shoes showed that shoes with thin, hard soles provide the most stability [62].

Nursing Interventions

Nursing practice interventions are probably the most widely used fall prevention strategies in institutional settings. The goal of most current nursing interventions is to identify at admission residents who are at high risk for falling and to institute appropriate precautions. To assist in the identification of such high-risk residents, fall assessment tools have been developed and described in the literature [12, 63]. Using these tools, the nursing staff assesses such factors as mental status, history of falls, ambulation status, medications, physical status, continence, and sensory deficits. A resident's status for risk for falling is determined by either the number of risk factors present or by a summary score [11, 62]. Once a resident has been identified as being at high risk for falling, a nursing care plan is usually developed that includes interventions aimed at injury prevention. Such interventions include indicating on the medical chart and the resident's door that the resident is at high risk for falls, moving high-risk residents to rooms close to the nursing station to increase observation, periodically reassessing residents after new episodes of illness or change in medication, lowering side-rails and bed heights for residents who climb out of bed, increasing the nurse-to-resident ratio, and instituting fall prevention education for residents and staff.

Few published reports have supported the validity of assessment tools or the effectiveness of these types of prevention programs. However, one recent study [63] reported high reliability and validity of a scored fall assessment tool developed to identify hospitalized patients at high risk for falling. Implementing this tool, using standardized nursing care plans for high-risk patients, and installing new safety equipment (such as safety vests and bed alarm systems) resulted in an average 20% decrease in the number of falls during the first year of the program.

Fall Prevention Devices

Until recently, the most common “devices” that were used in the nursing home to prevent falls were physical restraints such as poseys and bedrails. In U.S. nursing homes in the 1980s, the reported prevalence of restraint use ranged from 25% to 85% of residents [64]. This is in marked contrast to rates in many European countries and to current recommendations that restraints be used only in relatively rare and well-documented situations in which less restrictive measures are clearly unfeasible [65]. Since the new federal regulations went into effect in 1990 [65], there has been a major move away from the use of physical restraints, and research has shown that the adverse effects of physical restraints on functional status and quality of life are worse than any potential benefit in preventing falls [66]. Specifically, evidence suggests that physical restraints actually contribute to falls [7], injuries [18], and death [67]. Nursing homes are now beginning to look at reducing the use of physical restraints by implementing individualized care alternatives such as alterations in nursing care (as described previously) and environmental changes. In one nursing home where care alternatives were implemented, physical restraints were successfully removed in 92% of previously restrained residents [68]. One of the reasons that many nursing homes have been reluctant to discontinue the use of restraints has been the fear of potential legal liability for injuries sustained by an unrestrained patient [65]. However, no nursing home has ever been held legally liable for injuries to a resident solely on the basis of failing to apply restraints. Rather, lawsuits against nursing homes have been successful on the basis of injuries associated with the use of restraints [65]. Given these legal realities, clinicians should be more aggressive in finding alternatives to restraints and should help residents and their families understand that all falls may not be preventable and that the risk for a fall is often preferable to restraints.

Several devices that alert caregivers to patient movement or that protect patients from injuries from falls are currently being developed and marketed as possible alternatives to restraints for many high-risk patients. The most widely available devices are various alarm systems that are activated when patients try to get out of bed or move unassisted. One such alarm system was pilot-tested on an orthopedic and a general medicine hospital ward. Preliminary 5-month data indicated that the number of patient falls was reduced 33% and 45% on each ward, respectively [69]. An infrared scanning system that activates an alarm in the nursing station when a patient sits up or gets out of bed was found to reduce the incidence of nighttime falls from 2.8 to 1.0 falls per month when installed on a psychogeriatric unit [70]. Video recording systems are also being used as a means of providing closer monitoring of patient activities [71].

Injury-prevention alternatives such as hip protective pads are also being tested but are not generally available. A recent Danish study [72] tested hip protective pads in a nursing home where hip fractures had been extremely common. In wards randomly assigned to use the pads, hip fracture rates were markedly lower than those in the comparison wards, with a risk reduction of almost 60%. Even more striking was that no hip fractures occurred among patients actually wearing external hip pads.

Conclusion

Research in the nursing home has verified the complex multifactorial causes of falls. Many specific identifiable risk factors substantially increase the risk for falls and fall-related injuries. The clinician should focus on identifying these risk factors and intervening actively to prevent future falls. Although further research is needed, evidence suggests that a vigorous diagnostic, therapeutic, and preventive approach should be used for all nursing home residents who fall, as well as for those identified as being at high risk for falling.