Clinical manifestations

Presentation and course

Falls are the leading cause of both nonfatal injuries and unintentional injury deaths among older people in the United States, and dizziness is a major contributor to such falls. More than a third of community-dwelling older people (ie, 65 years or older) fall each year, and 10% of these falls result in serious injury (193). Injuries from falls include hip fracture, other fractures, subdural hematoma, and other head injury. Many elderly people have multiple medical problems for which they are receiving medications, which, in turn, can often increase the risk of dizziness and falls. The risk of falling and confusion increases with increasing numbers of medications, independent of the types of medications.

Community physicians, including neurologists, frequently fail to detect falls and gait disorders, and detected falls often receive inadequate evaluation, leading to a paucity of recommendations and treatments (164). Adhering to guidelines may improve outcomes in community-dwelling older adults (164).

A focused history and physical examination after a fall can usually determine both the immediate underlying causes of the fall and contributing risk factors. In addition, regular evaluations can help identify patients at high risk who can then be targeted for specific treatment and prevention strategies.

Assessment should include evaluating the circumstances of the fall and a complete history and physical examination, looking for potential risk factors. Patients who have fallen or who have a gait or balance problem are at higher risk of future falls. Therefore, screening for the risk of falling during the clinical examination should begin with determining if the patient has fallen in the past year and should include an assessment of gait and balance (154; 65; 190; 193). A history of falling in the past year strongly predicts the likelihood of future falls (190). Patients with neurologic or other medical conditions associated with an increased risk of falling should be asked about recent falls and should be examined for the presence of specific neurologic deficits that predict falls, including cognitive impairments, gait and balance disorders, and deficits of lower extremity strength, sensation, and coordination (190). It is also essential to review all medications (including over-the-counter medications) for potential contributors to falls, especially in the elderly.

Examination should include (1) orthostatic vital signs to assess for orthostatic hypotension (ie, pulse and blood pressure measured supine, standing, and standing after 3 minutes); (2) cardiac rhythm and rate; (3) examination of cognition, vision, hearing, heart, lungs, joints (mechanics and range of motion, especially for the spine, hips, knees, and ankles), muscle strength, and sensory function (especially tactile and proprioceptive function in the feet and Romberg sign); and (4) observation of the patient rising from a chair, standing, walking, and turning (either as part of the standard neurologic examination or in conjunction with the Timed Up-and-Go [TUG] test) (124; 110).

Although orthostatic hypotension is most prevalent and symptomatic within 1 to 2 minutes after standing, delayed or persistent orthostatic hypotension (after 4 to 5 minutes) is more informative for fall risk (147).

Some of the abnormalities of standing and gait that may cause or contribute to falls include (1) difficulty standing from a seated position without use of arms, proximal greater than distal weakness, and Trendelenburg sign(s) (ie, when the leg supports the weight of the body on the lesioned side, the pelvis dips on the contralateral side), which suggest a myopathic process; (2) difficulty with gait initiation, which suggests a fronto-subcortical disorder (eg, Parkinson disease or normal pressure hydrocephalus); (3) worse performance (stance or gait) with eyes closed than open (ie, Romberg sign), which suggests proprioceptive, vestibular, or possibly cerebellar impairment (110); (4) asymmetric step length, which may indicate a muscle, joint, or nervous system problem; (5) various movement disorders (eg, parkinsonism, ataxia, chorea); (6) spastic weakness (eg, hemiparesis, paraparesis); and (7) a steppage or semi-steppage gait.

Screening measures and functional assessments have been developed to assess risk of falls, which typically assess overlapping neurologic functions (eg, gait, mobility, and balance) (124; 192; 196; 202). Although their marginal value beyond that provided by a standard neurologic examination has been questioned (190), some do provide useful qualitative and quantitative information that is useful for assessment and monitoring.

No single factor can be relied on as a sole predictor of fall risk or fall injury risk because so many diverse factors are involved in falling; however, one-leg balance is a useful predictor of injurious falls and is easy to administer (202).

A very useful algorithm for falls risk assessment and intervention has been developed by the Centers for Disease Control and Prevention/STEADI (Stopping Elderly Accidents, Deaths, and Injuries) Program (182; 97; 116). This algorithm categorizes patients into three risk categories (Low, Medium, and High) on the basis of a history of falls in the past year, fall-related injuries, feelings of unsteadiness while standing or walking, worries about falling, and performance on three specific, easily administered, functional tests of balance, coordination, and locomotion: the 4-Stage Balance Test, the 30-Second Chair Stand Test, and the Timed Up-and-Go (TUG) Test.

These have been commonly used in studies of falls and fall-related injuries and are frequently employed in geriatrics clinics. The risk category then drives further assessment and management in a tiered fashion. The “Quick STEADI,” a simplified two-level fall risk screening algorithm for predicting falls among older adults, has also been proposed (127).

Medication review

Many elderly people have multiple medical problems for which they are receiving medications, which, in turn, can often increase the risk of dizziness and falls. The risk of falling and confusion increases with increasing numbers of medications, independent of the types of medications.

Certain medications are frequently associated with an increased risk of falls or falls with serious injury, including sedatives, hypnotics, anxiolytics, antidepressants (particularly tricyclics and selective serotonin reuptake inhibitors [SSRIs], but not serotonin and noradrenalin reuptake inhibitors [SNRIs]), neuroleptics, antihypertensives, anticonvulsants, muscle relaxants, anticoagulants, diuretics, sodium-glucose cotransporter-2 (SGLT2) inhibitors, urinary antimuscarinics (for overactive bladder), or alpha-blockers (for benign prostatic hypertrophy) (135; 155; 96; 95; 93; 191; 62; 203; 193; 16; 89; 198). The risk of falling and confusion increases with increasing numbers of medications, independent of the types of medications (191; 193).

Centrally acting antihypertensives and alpha-blockers significantly increase the risk of falls through a variety of mechanisms (ie, hypotension, orthostatic hypotension, arrhythmias, and sedation). These agents are no longer recommended for the treatment of hypertension unless all other agents are contraindicated or not tolerated and should be prioritized for deprescribing in older frailer individuals (209).

Table 1. Medications Particularly Likely to Increase the Risk of Falling or Falling with Serious Injury

Medication management often involves consideration of complex trade-offs between the benefits and risks of medications, particularly in the common geriatric clinical situation of multiple medical problems: treatment of one condition may worsen another (193). This is augmented by fragmentation of care across multiple healthcare providers, poor provider communication, and "push-button" prescribing, sometimes based on narrow interpretations of clinical guidelines. Antihypertensives, anticoagulants, and antidepressant medications commonly pose such trade-offs for patients at increased risk for falling. Ideally, medication optimization will consider the patient's medical problems in a holistic rather than specialty-specific manner and will also prioritize management of factors most likely to affect the quality of life of the patient. So, for example, it may be necessary to relax blood pressure treatment goals in someone with supine hypertension but with severe orthostatic hypotension that is causing syncope and falls. By eliminating unnecessary medications and reducing the dose of necessary medications, it is often possible to satisfactorily manage coexisting conditions while minimizing the risk of medication-related fall or injury.

Selected fall-related information for specific neurologic diseases

Multiple sclerosis. Individuals with multiple sclerosis have an increased risk of falls and falls with injury, with the latter being exacerbated by a higher prevalence of osteoporosis at an earlier age (due to mobility issues, low vitamin D levels, and acute glucocorticoid use for multiple sclerosis exacerbations) (69).

Parkinson disease. Reported difficulty turning is a sensitive indicator of freezing or falling in Parkinson disease, both indicators for physiotherapy (177). Patients with severe Parkinson disease who appear unstable when reaching are likely to be repeat fallers and are at risk of further falls during more demanding activities (176). Standard in-office turning tests for Parkinson disease poorly reflect step counts during real-life turning, whereas unobtrusive structured observation reveals the turning strategies patients habitually use (177).

A systematic review of prognostic factors associated with falls in Parkinson disease across 34 studies involving a total of 3454 individuals found that a previous history of falls, gait disorders, and poor balance are robust prognostic markers for falls in Parkinson disease (134).

A meta-analysis found that exercise is effective in reducing falls for frail aging people with Parkinson disease (RR: 0.49, 95% CI: 0.39–0.69) (59).

Dementia, Alzheimer disease, and cognitive frailty. In a systematic review and meta-analysis of randomized controlled trials of cholinesterase inhibitors (donepezil, galantamine, rivastigmine) in patients with cognitive impairment, 53 studies (30 donepezil, 14 galantamine, 9 rivastigmine) were included, providing data on 25,399 patients (03). In patients with neurocognitive disorders, cholinesterase inhibitors were associated with a decreased risk of falls and an increased risk of syncope but were not associated with accidental trauma or fractures.

A meta-analysis found that exercise is effective in reducing falls for frail aging people with dementia (RR: 0.64, 95% CI: 0.51–0.82) (59).

Antipsychotic use in Alzheimer disease is associated with adverse events and increased mortality. In an analysis of longitudinal data from an 18-month trial, long-term antipsychotic use in individuals with mild-to-moderate Alzheimer disease was associated with (1) sit-to-stand orthostatic hypotension and (2) a greater risk of incident falls and syncope over 18 months (55).

"Cognitive frailty" is the combination of cognitive decline, physical frailty, and a reduced ability to perform daily activities in older adults (137). In a systematic review and meta-analysis, cognitive frailty was significantly associated with falls (206).

Stroke and complications of stroke prevention. Stroke is associated with an increased risk of falls and falls with injury (24). A systematic review and meta-analysis found that a small, but significant, association exists between falls and the fear of falling in both acute and chronic stroke patients (149).

Anticoagulants are often underused in older people with atrial fibrillation due to concerns about the risk of falls and intracranial hemorrhage (130). For most individuals with atrial fibrillation, the advantages of stroke prevention with oral anticoagulation outweigh the risks of fall-related bleeding, especially with the use of non-vitamin K oral anticoagulants, which are associated with fewer intracranial hemorrhages and thromboembolic complications than vitamin K anticoagulants (111; 130). Direct-acting oral anticoagulants (DOACs, such as the direct thrombin inhibitor dabigatran and various factor Xa inhibitors, including rivaroxaban, apixaban, and edoxaban) are now recommended as first-line anticoagulant therapy for most patients with atrial fibrillation due to their favorable safety profile (130). Off-label dose reduction of DOACs is not recommended because efficacy is reduced with limited reduction in bleeding risk. Medication review and fall prevention strategies should be implemented before prescribing anticoagulation. Anticoagulant deprescribing should be considered in individuals with severe frailty, limited life expectancy, or increased bleeding risk (eg, cerebral microbleeds).

A post hoc retrospective analysis of intracranial hemorrhage and major bleeding outcomes among 18,113 individuals with atrial fibrillation found that falls were frequent, occurring in 4% of study subjects during the RE-LY Trial (30). Subjects who fell were older, had more frequent comorbidities (eg, diabetes, previous stroke, or coronary artery disease), and had significantly higher risks of major bleeding, intracranial hemorrhage, and mortality compared to those who did not have reported falls or head injury. Among patients who had falls, those allocated to dabigatran showed a significantly lower intracranial hemorrhage risk compared with warfarin.

The rate of delayed intracranial hemorrhage after a ground-level fall in patients taking anticoagulants or antiplatelet agents ranges from 0.6% to 6% (27). Patients on warfarin have a recognized increased risk of delayed intracranial hemorrhage, and there is no clear indication for follow-up imaging in patients taking DOACs who fall. In one study, no delayed hemorrhages were detected among 159 neurologically intact patients who were evaluated for falls after a fall-associated head strike, who were on DOACs, and who had a repeat head CT done (27).

Benign paroxysmal positioning vertigo (BPPV). BPPV increases the risk of falls and negatively impacts spatiotemporal parameters of gait. Individuals with BPPV walk slower and demonstrate more sway during tandem walking compared with controls and also walk slower during head rotations (142). Particle repositioning maneuvers improve the frequency of falls, the fear of falling, and gait during level walking (142).

Prognosis and complications

Falls are the leading cause of both nonfatal injuries and unintentional injury deaths among older people in the United States, and dizziness is a major contributor to such falls. Injuries from falls include hip fracture, other fractures (eg, wrist, skull), subdural hematoma, and other head injury.

Inability to rise without help, which is experienced by half of older persons after at least one fall, may result in dehydration, pressure ulcers, and rhabdomyolysis (193).

Persons at increased risk for falling face trade-offs between safety and functional independence: to reduce their risk of falling, they may have to avoid desired activities or rely on help. Conversely, they may have to accept a high risk of serious injury if they wish to continue performing activities beyond their current capabilities (193).

Biological basis

Etiology and pathogenesis

Classification of falls is helpful in elucidating the precipitating causes and in predicting future falls and fall-related injuries (109). A useful scheme for classifying falls is the St. Louis Oasis Study Fall Classification, which utilizes a 4-part categorization: extrinsic, intrinsic, non-bipedal, and unclassifiable (Table 2).

Table 2. St. Louis Oasis Study Fall Classification

Table 3. Patient-Related Factors Associated with An Increased Risk of Falling*

Parkinson disease. Gait hypokinesia in Parkinson disease results from difficulty in activating the motor control system causing an inability to generate appropriate stride length (ie, sufficiently large steps), whereas cadence control (ie, regulation of the timing of consecutive steps) is intact and is used as a compensatory mechanism (132; 133). Normal stride length can be elicited in Parkinson disease using attentional strategies and visual cues, but the effect of attention requires constant vigilance to prevent reverting to more automatic but suboptimal control mechanisms (133).

Epidemiology

Falls are common among older persons and frequently threaten their independence. Between 30% and 40% of community-dwelling older adults (aged 65 years and older) fall each year, and the rates are higher for nursing home residents (196; 140; 84; 80; 154; 193). In 2014, 29% of older adults reported falling at least once in the preceding 12 months, resulting in an estimated 29.0 million falls (20).

According to data from the National Center for Health Statistics, falls in the elderly are more common among women, increase with age, and vary by race and ethnicity. Falls also vary considerably by state, but without a clear pattern or clustering.

Some of the age-associated increase in falls is due to multifactorial changes in sensory and motor function in the elderly (51; 120). The elderly have high rates of visual, auditory, and tactile impairment, and some three quarters of those over 70 years of age have balance impairment (51). Sensory impairments differ between men and women. All forms of sensory impairment and balance impairment are significantly worse among individuals aged 80 years and older compared to individuals aged 70 to 79. Balance impairment is significantly worse among elderly Mexican Americans than among non-Hispanic whites and blacks. Balance impairment is significantly worse among elderly people living below the poverty level than among elderly people living at or above the poverty level.

Muscle strength also markedly declines with age, particularly among individuals aged 80 and over (120). The age-associated decline in muscle strength applies to both sexes but is worse among women aged 80 and older. Reduced muscle strength in adults aged 60 and over varies by race and Hispanic origin and is worse among non-Hispanic Asians and Hispanics than among whites or blacks. In addition to contributing to the increased risk of falls among the elderly, these changes in muscle strength negatively impact functional abilities and quality of life. Even simple tasks, such as rising from a chair, become progressively more difficult with worsening degrees of weakness.

Noninstitutionalized adults. To obtain national estimates of falls with injury for noninstitutionalized adults who are at least 65 years of age, epidemiologists from the Centers for Disease Control and Prevention combined data from the 2001–2003 National Health Interview Surveys (NHIS), conducted by the National Center for Health Statistics (166). NHIS is an ongoing continuous survey of a nationally representative sample of the civilian noninstitutionalized U.S. population. Three years of data were combined to provide enough cases for subgroup analyses. Data on nonfatal medically attended injuries in the 3-month period prior to the survey were obtained from an adult family member. Analysis was restricted to adults aged 65 years of age or older. Annually, 51 falls with injury per 1000 older adults required medical evaluation. The rates of falls with injury were similar in men and women but increased with increasing age and were higher in those with fair or poor health status, in those with various comorbidities (eg, stroke, diabetes, visual impairment, heart disease) or functional impairments (eg, marked difficulty or inability to walk a quarter mile; stand for 2 hours; stoop, bend, or kneel; or climb 10 steps without resting), and among those whose health status declined in the previous year. Subgroups with most (57%) of the falls with injury occurred due to slipping, tripping, or stumbling, but a significant minority (27%) occurred due to more specific neurologic issues (eg, loss of balance, dizziness, syncope, or seizures). Approximately three quarters (74%) of fall injuries among older adults occurred at home, either inside (50%) or outside (24%) of the house. Approximately one third (32%) of falls with injury resulted in a need for assistance with activities of daily living, and for most of these, the help was expected to be necessary for at least 6 months.

Institutionalized older adults. Falls in skilled nursing facilities are common and cause significant morbidity, mortality, and health system use (105). Institutionalized older adults are, as a group, much more likely to fall because this group suffers greater morbidities. They tend to be on more medications that increase the risk of falls, and this population is relatively older (ie, there is a higher proportion of adults aged 85 years and older among institutionalized versus noninstitutionalized populations of older adults). The mean incidence of falls in nursing homes is 1.5 falls per bed per year (162; 161).

In a prospective year-long cohort study of 140 women and 59 men (mean age 82 years; range 65–97) in five residential-care facilities, more than half (57%) of the residents had at least one fall during the 12 months of the study (95); overall, there were 482 registered falls, representing an average of 2.4 falls per resident per year.

Of 143 patients studied in a rehabilitation setting, 32% fell at least once (128). Of those who fell, only a quarter (26%) called for assistance prior to the fall; 68% of the falls were from wheelchairs. None sustained a serious injury or morbidity from the falls.

In a retrospective study on a 30-bed neurorehabilitation unit over a 4-year period, 117 falls occurred in 82 patients, with a reported rate of 1439 falls per 1000 patient years (04). Most falls were associated with no injury (n = 96) or minor injury (n = 18), whereas three cases sustained lacerations requiring sutures. Falls occurred with greatest frequency during the evening and while bed transfers were being performed. An increased risk of falling was associated with physician orders for Posey restraints.

In a systematic review and meta-analysis of eight prospective studies on the incidence and risk factors for falls among older people in nursing homes, the pooled incidence of falls was 43% (170). Risk factors with a strong association with falls included fall history, impaired performance of activities of daily living, insomnia, and depression. Risk factors with low to moderate associations included vertigo, walking aids, poor balance, use of antidepressants, use of benzodiazepines, use of antipsychotics, use of anxiolytics, polypharmacy, dementia, unsteady gait, hearing problems, and gender (being male). Having bed rails was identified as a protective environmental factor.

Conclusions and implications. The results from our meta-analysis suggest that the incidence of falls of older adults living in nursing homes is high, and the risk factors for falls are various. Assessments of balance and mobility, medical condition, and use of medications should be included as key elements in the fall risk assessments of older people in nursing homes. Environmental risk factors still need to be explored in future studies. Tailored fall prevention strategies should be implemented by addressing the modifiable risk factors.

Consequences of falls. Falls are associated with increased morbidity (eg, numerous physical injuries, disability), lower functional independence (eg, decreased ability to perform activities of daily living, nursing home placement), decreased quality of life, high healthcare utilization and costs, and increased mortality.

Annual medical costs in the United States for nonfatal fall-related injuries in older adults are estimated to be around $50 billion, whereas costs for fatal falls are estimated at $754 million (60). According to the Centers for Disease Control and Prevention, of the annual cost of $50 billion for nonfatal falls, $29 billion is paid by Medicare, $12 billion is paid by private or out-of-pocket payers, and $9 billion is paid by Medicaid. State costs vary considerably, without a clear pattern or evident clustering. These direct costs do not account for long-term consequences of these injuries (eg, lost time from work, disability, functional dependence, or reduced quality of life) (73).

Occupational fall-related injuries contribute to significant lost workdays (37). Among workers, approximately 20% of fall injuries involve ladders, and among construction workers, about 80% of fall injuries treated in U.S. emergency departments involve a ladder (174). The most common fall height from a ladder to result in injuries or death is just 6 to 10 feet.

A previous self-reported fall confers an increased risk of fracture that is largely independent of bone mineral density (200).

Fall-related injuries. Different survey approaches yield somewhat different estimates of rates of medically attended falls with injury among older adults. Those that utilize emergency room records, such as the National Hospital Ambulatory Medical Care Survey (NHAMCS), give somewhat higher estimates because they include both noninstitutionalized and institutionalized individuals and, to a degree, because they avoid some reporting and recall bias. Several studies have noted that older adults may underreport medically attended falls with injury, particularly those that are relatively minor (46; 207).

By some estimates, at least 10% of falls among older adults (aged 65 and older) result in serious injury (191), but rates are higher among institutionalized older adults.

Falls are the leading cause of nonfatal injuries and hospital admissions for trauma among older adults (05; 02), accounting for 1.7 to 1.8 million emergency room visits and more than 420,000 hospitalizations in the United States annually (36; 166). The risk of being seriously injured by a fall increase with age (183), and nonfatal falls with injury as well as nonfatal falls with injury requiring hospitalization are significantly more frequent in women than in men (36).

Annually, over 1.7 to 1.8 million older adults in the United States are treated in emergency rooms, and more than 420,000 are hospitalized for fall-related injuries (166). In the United States in 2014, of the elderly who fell, 38% reported at least one fall that required medical treatment or restricted their activity for at least 1 day, resulting in an estimated 7.0 million fall injuries (20).

Among states in which falls and fall injuries were consistently reported over the years (excluding Michigan, Oregon, and Wisconsin where data in 2012 were reported differently than in other years), the percentage of those older adults reporting at least one fall increased from 27.9% in 2012 to 29.6% in 2016 (p < 0.001) and decreased to 27.4% in 2018 (p < 0.001) (131). Age-specific percentages all showed similar peaks in 2016. The rates of falls and fall-related injuries and the percentages of older adults reporting a fall-related injury did not significantly change from 2012 to 2018.

In a prospective year-long cohort study involving 140 women and 59 men (mean age 82 years; range 65–97) in five residential care facilities, approximately one third of these falls resulted in injury (32%), and two thirds of the fallers (65%) sustained at least one injury.

Falls are the leading cause of traumatic brain injuries in older adults (90). Traumatic brain injuries are responsible for approximately half of all fatal falls among older adults (178).

Most fractures among older adults are caused by falls (19). Fall-related fractures among older adults are more than two times higher in women than in men (36). Fall-related fractures can lead to reduced quality of life and premature death (212; 76). The most common fracture sites are the spine, hip, forearm, leg, ankle, pelvis, upper arm, and hand (169).

Falls cause most hip fractures, which often result in long-term functional impairment and prolonged rehabilitation and nursing home stays, or even permanent nursing home placement. More than 95% of hip fractures among older adults are caused by falls (71), and 80% of these occur among women (181). Among both sexes, hip fracture rates increase markedly with age, so people aged 85 years and older are 10 to 15 times more likely to fracture a hip than are people aged 60 to 65 years, a gender difference that in part reflects a much higher rate of osteoporosis among women (70). Men also fracture their hips, but most commonly after the age of 80 years (36). A quarter of older adults who lived independently before fracturing a hip spend at least a year in a nursing home, and more than 20% of older adults who fracture a hip die within a year of their injury (117).

Many older people who fall develop a fear of falling, even if uninjured, which causes them to limit their activities; this results in reduced mobility and physical fitness and an increased risk of further falls (201).

Fall-related changes in independence and quality of life. Falls and fall-related injuries are among the most common causes of decline in the ability to care for oneself and to participate in social and physical activities (193); indeed, falling without a serious injury increases the risk of nursing home placement 3-fold (after accounting for cognitive, psychological, social, functional, and medical factors), and a serious fall-related injury increases the risk 10-fold.

Among 1815 community-dwelling subjects over the age of 75 who had a standardized and validated primary care health check in the United Kingdom, risk of admission to long-term care over 1 year was markedly increased, both for single fallers (odds ratio 3.8, 95% CI: 1.8–8.3) and recurrent fallers (odds ratio 4.5, 95% CI: 1.7–12) (52).

Fall-related deaths. The Web-based Injury Statistics Query and Reporting System (WISQARS™) of the Centers for Disease Control and Prevention (CDC) is an interactive, online database that provides fatal and nonfatal injury, violent death, and cost of injury data (17; 60; 114; 119; 144; 145; 146; 44).

Analysis of WISQARS data on falls demonstrates that fall-related deaths are predominantly unintentional rather than suicidal. Indeed, a histogram of total, suicidal, and unintentional falls by age shows a marked age-associated increase in total and unintentional fall-related deaths for age groups over 60 years, but suicidal deaths are so infrequent as to be hardly noticeable. The age-specific rates of fall-related deaths follow essentially the same pattern as total fall-related deaths. Fall-related deaths vary by intent across different races, but suicide makes up a small minority of fall-related deaths in every race group. In the interval from 2001 to 2020, males had higher age-adjusted rates of fall-related deaths overall for unintentional falls and for suicidal falls; the most recent available data (ie, for 2020) show that men had an age-adjusted unintentional fall-related death rate of 12.57 per 100,000, compared to 8.44 per 100,000 for women. Age-adjusted fall-related death rates have steadily increased overall and for unintentional falls from 2001 to 2020, whereas suicide-related falls have changed very little in that interval. The increases in age-specific deaths from unintentional falls in the interval from 2001 to 2020 are most prominent for individuals aged 70 and over. In the interval from 2001 to 2020, age-adjusted unintentional fall-related death rates varied by race and were more than two times higher in whites than in blacks, with intermediate values for American Indian/Alaska natives and Asian/Pacific Islanders. In this interval, age-adjusted death rates for unintentional falls increased markedly for non-Hispanic whites and progressively less for Hispanic whites and Hispanic and non-Hispanic blacks. The spatial pattern of age-adjusted fall-related mortality rates by state shows considerable variation but no clear pattern or clustering.

Both the number of fall-related deaths and age-adjusted rates of fall-related deaths have been rising in the United States for persons aged 65 and older, and this applies at a national level and for nearly every state (except Hawaii) (29). A major component of fall-related deaths is traumatic brain injury, and it is no surprise that age-adjusted rates of traumatic brain injuries due to unintentional falls have been increasing for most states (143).

Falls are the leading cause of injury-related deaths, accounting for approximately 14,000 deaths among older adults in the United States each year (36). The risk of fatal falls increases with age; indeed, approximately six of seven deaths from falls in the United States were among those aged 75 years and older (36). Fatal falls are more common among older men than older women after adjusting for age differences (183; 36). Rates of fall-related deaths have increased significantly over the past decade (180; 178).

Among 1815 community-dwelling subjects over 75 years of age who had a standardized and validated primary care health check in the United Kingdom, the risk of death was increased at 1 year (odds ratio 2.6, 95% CI 1.4–4.7) and 3 years (odds ratio 1.9, 95% CI 1.2–3.0) for recurrent fallers but not for single fallers (52).

In a review of U.S. mortality data from 1979 to 1986, falls accounted for 12% of the deaths associated with head injury (175).

Risk factors for falls. Risk factors for falls include a history of falls; muscle weakness or rigidity; use of four or more prescription medications; use of specific medications; use of an assistive device; symptomatic knee and hip osteoarthritis; depression; various neurologic disorders; prior hip fracture; age older than 80 years; impairments in gait, balance, cognition, vision, and activities of daily living; and multiple comorbidities ("multimorbidity") (154; 190; 74; 193; 86; 216). An increased risk of falls is established among persons with diagnoses of stroke, dementia, and disorders of gait and balance and is probable among patients with Parkinson disease, peripheral neuropathy, lower extremity weakness or sensory loss, and substantial vision loss (190; 24). Older adults have a 5- to 8-fold increased risk of all-cause mortality during the first 3 months after a hip fracture and a persistent excess mortality of 10% to 20% in the first 1 to 10 years after hip fracture (74).

Mechanical factors (eg, arthritis, amputation) can also interfere with ambulation and may greatly increase the risk of falls (160; 18; 39). A self-reported injurious fall within the preceding year was associated with difficulty climbing 10 steps without special equipment, loss of dependence for an activity of daily living, pain in the legs and below the knees, and moderate visual impairment (63).

Ganz and colleagues performed a meta-analysis of studies identifying the prognostic value of risk factors for future falls among community-dwelling persons aged 65 years and older (65). The estimated pretest probability of falling at least once in any given year for individuals 65 years and older was 27%. Patients who had fallen in the past year were two to three times more likely to fall again. In this study, the most consistent predictors of future falls were clinically detected abnormalities of gait or balance, whereas visual impairment, medications, functional limitations (ie, decreased activities of daily living), and impaired cognition did not consistently predict falls across studies, and orthostatic hypotension did not predict falls after controlling for other factors.

Inpatient rehabilitation setting. In a rehabilitation setting, impaired ability to follow directions, impaired judgment, impaired proprioception, presence of physical restraints, use of major tranquilizers, use of sedatives, and presence of psychiatric diagnosis were all individually associated with patients who fell (128). Men fell more than women. In a multivariate analysis, altered proprioception was the only major predictor of falling.

The most common precipitating causes of falls in nursing homes include gait and balance disorders, weakness, dizziness, environmental hazards, confusion, visual impairment, and postural hypotension (162; 161). The most important underlying risk factors for falls and injuries include some of these same items and others, such as lower extremity weakness, gait and balance instability, poor vision, cognitive and functional impairment, and sedating and psychoactive medications (162; 161).

Nursing home setting. A prospective cohort study with 1-year follow-up conducted in 12 community nursing homes in Tennessee included 1228 residents, 65 years of age or older; 725 (59%) were nonambulatory and 503 (41%) were ambulatory (189). As ascertained from facility incident reports and nursing home charts, there were 111 first falls resulting in serious injury (eg, head injury with altered consciousness, fracture, joint dislocation or sprain, or sutured laceration) that required medical treatment (hospitalization, emergency room visit, physician visit, or onsite radiological examination). The incidence of injurious falls in nonambulatory residents (6.7 per 100 person-years) was less than half that in ambulatory residents (17.0 per 100 person-years). The circumstances of injurious falls differed significantly for the two groups, with injurious falls in nonambulatory residents more likely to involve equipment (87% vs. 45%), to occur while seated or during transferring (82% vs 21%), and to occur from a chair/bed level (54% vs. 6%). After controlling for other factors, the nonambulatory residents at highest risk were those who were not bed-bound and with capacity for independent transfer; the ambulatory residents at highest risk were those receiving psychotropic drugs.

In a cohort study of older people (58 women and 25 men) in residential care followed for a period of 1 year after baseline assessments, the incidence rate of falls was 2.29 falls/person-year (96). Antidepressants (selective serotonin reuptake inhibitors, SSRIs), impaired vision, and being unable to use stairs without assistance were independently associated with being a “faller.” Fifty-four percent of fallers suffered injuries as a result of their falls, including 21 fractures; 27% of the falls were judged to be precipitated by an acute illness or disease and 9% by a side effect of a drug. Acute symptoms of diseases or drug side effects were associated with 58% of the falls that resulted in fractures.

In a prospective observational study of institutionalized, frail elderly in southern Germany with a 1-year follow-up, the study population included 472 long-term-care residents whose mean age was 84 years (108). Risk indicators for accidental falls were analyzed using logistic regression. The incidence of falls was 2.6 per resident-year. Multivariate analysis revealed short-term memory loss, needing transfer assistance, urinary incontinence, history of a prior fall, and use of trunk restraints as predictors of falls. Depressive symptoms, transfer assistance, urinary incontinence, and positive fall history were associated with frequent falls.

In a prospective year-long cohort study of 140 women and 59 men, (mean age 82 years; range 65–97) in five residential care facilities, previous falls and treatment with antidepressants were the most important predisposing factors for falls (95). Probable precipitating factors could be determined in 69% of the 482 registered falls. When precipitating factors were judged to be involved, typically only one precipitating factor was identified (90%). Factors associated with falls included previous falls, disability, impaired cognition, depression or treatment with antidepressants, delirium, and use of laxatives. Independent predictors of falls in multivariate statistical models were falls within the last 6 months and treatment with antidepressants. Acute disease or symptoms of disease were judged to be precipitating, alone or in combination, in 39% of falls; delirium was a factor in 10%, and infection (most often urinary tract infection) was a factor in 8%. Of the falls precipitated by drugs (8%), most involved benzodiazepines or neuroleptics. External factors, such as material defects and obstacles, precipitated 8% of the falls. Other conditions both related to the individual and the environment, such as misinterpretation (eg, overestimation of capacity or forgetfulness), misuse of a roller walker, or mistakes made by the staff, were precipitating factors in 17% of the falls.

Table 4. Precipitating Factors for Falls

In a population-based cross-sectional study of all geriatric care settings in the county of Västerbotten, Sweden, (residential care facilities, nursing homes, group dwellings for people with dementia, rehabilitation/short-stay units, geriatric, and psychogeriatric clinics) collectively involving 3604 residents with a mean age of 83 years (range 65–103), 8.4% sustained a fall at least once during the preceding week (93). Multivariate analyses showed that a history of falls, the ability to get up from a chair, the need for a helper when walking, pain, cognitive impairment, and use of neuroleptics or antidepressants were all associated with falls. Among the antidepressants, selective serotonin reuptake inhibitors (SSRIs), but not serotonin and noradrenalin reuptake inhibitors (SNRIs), were associated with falls. In a subsequent study of the same population focusing on those with cognitive impairment, 9.4% of 2008 cognitively impaired residents had fallen at least once during the preceding week (94). Being able to get up from a chair, previous falls, needing a helper when walking, and hyperactive symptoms were the factors most strongly associated with falls.

In a population-based cohort study of 253 people aged 85 and older in Umeå, Sweden, 87% were followed up for falls for 6 months; 109 lived in ordinary housing and 111 lived in institutional housing. Forty percent of the participants fell. The rate of falls was 2.17 per person-year, with 0.83 fall-related injuries per person-year, including 0.14 fractures per person-year. In a multivariate analysis, independent explanatory risk factors for time to first fall were dependency in activities of daily living, thyroid disorders, treatment with selective serotonin reuptake inhibitors (SSRIs), and occurrence of falls in the preceding year.

Osteoporosis. Epidemiologic studies indicate that at least half the population burden of osteoporosis-related fractures affects persons with osteopenia (low bone density), who comprise a larger segment of the population than those with osteoporosis (173).

Parkinson disease. Parkinson disease is associated with an increased incidence of osteoporosis, vitamin D deficiency, falls, and fall-related fractures.

In a population-based retrospective cohort study, 138 residents of Olmsted County, Minnesota, who were first diagnosed with Parkinson disease during 1967 to 1979 were matched by age and sex to an equal number of control subjects from the community (92). Patients with Parkinson disease experienced significantly more new fractures than controls (33% vs. 20%) during 696 person-years of follow-up. The greatest risk was seen for proximal femur fractures: by 10 years after diagnosis, an estimated 27% of the Parkinson disease cohort experienced a new hip fracture.

In a case-control study of 82 patients with established Parkinson disease aged 65 years or older and 68 age-, sex-, and body mass index (BMI)-matched healthy control subjects, bone mineral density was significantly lower in patients with Parkinson disease than in controls (01). Compared to healthy controls, patients with Parkinson disease had significantly decreased vitamin D levels, significantly increased levels of bone alkaline phosphatase and N-terminal telopeptide of type I collagen, reduced physical and mental performance, and more falls or fractures.

In 186 patients with early-stage Parkinson disease, (Hoehn and Yahr stage 1–2.5), osteoporosis (12%) and osteopenia (41%) were common (197). Female gender, low weight, and low 25-OH-vitamin D level were significantly correlated with bone mineral density of the hip and lumbar spine. More than half of the patients with early-stage Parkinson disease had an abnormal bone mineral density. Female gender, low weight, and low vitamin D concentration were associated with bone loss.

In a consecutive series of 182 patients with Parkinson disease, patients with lower vitamin D levels had a significantly higher frequency of falls (215).

Epidemiology of falls with selected neurologic diseases

Stroke. Risk factors for falls among stroke patients include instability in hospital, near falls, and poor upper limb function (impaired ability to self-correct or prevent a fall) (13). Among stroke patients, repeat fallers have greater mobility deficits, significantly reduced arm function, and greater disability (impairments in activities of daily living) compared to those who have not fallen or experienced near falls (88). Loss of balance, misjudgment, and foot dragging during walking, turning, and standing contribute to falls in these patients. Attention deficits in stroke patients may contribute to accident-prone behavior and falling because deficits in sustained and divided attention correlate with functional impairments and falls (87).

Bradykinetic rigid syndromes. In a study of 782 cases with pathological diagnosis of various bradykinetic rigid syndromes (474 Parkinson disease; 127 progressive supranuclear palsy; 91 multiple system atrophy; 46 dementia with Lewy bodies; 27 vascular parkinsonism; nine Alzheimer disease; eight corticobasal degeneration), falls were recorded in 78% and fractures in 17% (211). In Parkinson disease, female gender, symmetrical onset, postural instability, and autonomic instability all independently predicted time to first fall. In Parkinson disease, progressive supranuclear palsy, and multiple system atrophy, latency to first fall was shortest in those with older age of onset. Median latency from disease onset to first fall was shortest in progressive supranuclear palsy (12 months), multiple system atrophy (42 months), and progressive supranuclear palsy–parkinsonism (47 months) and longest in Parkinson disease (108 months). Across the group of all cases, fractures of the hip were more than twice as common as wrist and forearm fractures. Fractures of the skull, ribs, and vertebrae occurred more frequently in progressive supranuclear palsy than in other diseases.

Parkinson disease. About three fourths of patients with Parkinson disease fall during their illness, and most patients fall within about 9 years of disease onset. Falls tend to occur later in patients with Parkinson disease than in patients with other bradykinetic rigid syndromes. Falls in patients with Parkinson disease are associated with female gender, postural instability, no tremor, and late clinical features of cognitive dysfunction, speech disturbance, dysphagia, autonomic dysfunction, and hallucinations. Independent predictors of earlier falls include (in decreasing order of the strength of the association) autonomic dysfunction, symmetric onset, female gender, and older age of onset. Clinical features associated with fractures include falls, late postural instability, and female gender. About half of fractures are of the hip, and a third are of the upper limb. Hip fractures are more frequent in patients with Parkinson disease than in patients with progressive supranuclear palsy or multiple system atrophy.

Women with Parkinson disease are more prone to fall, fall earlier in the disease course, and are more likely to sustain a fracture. The higher risk of fractures in female patients with Parkinson disease likely results from a combination of a greater propensity to fall and a lower bone mineral density (including greater frequency of osteoporosis), making hip fractures particularly more likely when falls occur (Aita 1982; 71; 40; 92; 99). Several studies have documented low bone mineral density in Parkinson disease patients, with a greater prevalence of severe osteoporosis at higher Hoehn and Yahr stages (82; 99).

In patients with Parkinson disease, the use of multiple medications, greater physical disability, greater disease severity, postural instability (eg, shorter functional reach and greater postural sway), dyskinesias, and on–off phenomena are associated with falls (14).

Patients with Parkinson disease experience a high frequency of falling, with almost half of patients falling during a 3-month period (148) and about two thirds falling during a 1-year period (15). The strongest predictor of falling is a history of prior falls (two or more in the preceding year), but even subjects without any prior falls have a considerable risk of sustaining future falls (approximately 20% in 3 months) (15; 148). Disease severity alone is not a good predictor of falls due, in part, to the complex U-shaped relation with falls [ie, falls increase with disease severity through a Unified Parkinson's Disease Rating Scale (UPDRS) score of 25 to 35 and then decrease with more severe disability as patients become less ambulatory] (148).

In a study of 87 patients with Parkinson disease, fallers had longer disease duration, higher levodopa-equivalent doses, greater “on” time with dyskinesia, and higher axial scores on the UPDRS (168). Severity of psychosis, executive cognitive impairment, autonomic (particularly cardiovascular) dysfunction, and sleep disturbances (particularly REM sleep behavioral disorder) were significantly associated with falls. Fallers more frequently reported use of antidepressants (both tricyclics and SSRIs) and neuroleptics but not hypnotics. In logistic regression analysis, cardiovascular dysfunction, antidepressant use, and REM sleep behavior disorder were significantly associated with falls.

Progressive supranuclear palsy. Almost all patients with progressive supranuclear palsy fall at some point in their illness, and most fall within 2 years of onset. Falls occur earlier in patients with progressive supranuclear palsy than in patients with any other bradykinetic rigid syndrome, and falls in patients with progressive supranuclear palsy are more likely to result in a fracture than falls in patients with other bradykinetic rigid syndromes. Early falls (in the first 2 years after disease onset) were associated with cognitive dysfunction, absence of tremor, axial rigidity, symmetric onset, postural instability, eye movement abnormalities (vertical gaze paresis or abnormal saccades), and speech disturbance. Progressive supranuclear palsy patients with falls in the first 2 years after disease onset had an average life expectancy of 3 years less than other progressive supranuclear palsy patients. Nearly a third of patients with progressive supranuclear palsy fractured at least one bone, and the frequency of fractures was greater in progressive supranuclear palsy than in any other bradykinetic rigid syndrome. Interestingly, early bradykinesia, early limb rigidity, and early pyramidal signs in patients with progressive supranuclear palsy were associated with a lower frequency of fractures—presumably because these patients were less likely to ambulate independently. About a third of fractures in patients with progressive supranuclear palsy were of the hip, about a third were of the upper limb, and the remainder were at other locations.

Multiple system atrophy. More than three quarters of patients with multiple system atrophy fall during the course of their illness, and most fall within the first 4 years of disease onset. Falls in patients with multiple system atrophy are associated with late clinical features of limb rigidity, speech disturbance, dysphagia, and pyramidal tract signs. Independent predictors of earlier falls include early postural instability and older age of onset. Clinical features associated with fractures include female gender.

Prevention

Current guidelines recommend that individuals who report a fall in the prior year, have concerns about falling, or have a gait speed less than 0.8 to 1 m/s should receive fall prevention interventions (43).

Interventions to prevent falls in older adults are effective in reducing both the risk of falling and the monthly rate of falling (42; 48; 38; 75; 32; 65). Although effective multifactorial interventions reduce the frequent falling rate of older patients by 30% to 40%, these interventions are often reserved for high-risk patients (38; 65). Furthermore, the results across studies show considerable heterogeneity, and the specific components of multifactorial interventions that are helpful are generally unknown.

Effective fall prevention strategies for community-dwelling adults include multifactorial interventions targeting identified risk factors, exercises for muscle strengthening combined with balance training, and withdrawal of psychotropic medication (154; 193). Exercise programs and physical therapy can also be effective in reducing the risk of falling (38; 193; 43). Among older people living in their own homes, group-based exercise was the most potent single intervention tested, and the associated reduction in falls was associated with improved balance (48). A meta-analysis of 59 randomized clinical trials in average-risk to high-risk populations found that exercise interventions to reduce falls were associated with significantly fewer falls (655 falls per 1000 patient-years in intervention groups vs. 850 falls per 1000 patient-years in nonexercise control groups), with most trials assessing balance and functional exercises (43).

The most effective intervention may be a multifactorial falls risk assessment and management program (38; 160; 193), although implementation hurdles can limit effectiveness (06). A meta-analysis of 43 randomized clinical trials of interventions that systematically assessed and addressed multiple risk factors among individuals at high risk found that multifactorial interventions lowered the number of falls (multifactorial interventions were associated with 1784 falls per 1000 patient-years in intervention groups vs. 2317 falls per 1000 patient-years in control groups), without a significant difference in the number of individuals who fell (43).

In addition to multifactorial assessment and intervention and exercise programs, other fall prevention approaches may also be useful, including (1) evaluation and management of comorbid conditions predisposing to falls (eg, orthostatic hypotension, parkinsonism, delirium, leg weakness, arthritis, and cataracts); (2) medication management (eg, limiting or curtailing the use of certain medications, particularly those with sedative effects or those contributing to orthostatic hypotension or motor disequilibrium or ataxia); (3) provision of appropriate footwear; (4) proper use of appropriate assistive devices (eg, cane, walker); and (5) elimination of home hazards, etc. Some approaches can decrease the risk of injury from falls without necessarily impacting the risk or frequency of falls: evaluation and management of osteoporosis, and provision of protective orthotics (eg, hip protectors). These prevention methods are largely already accepted in clinical practice and have been incorporated into "usual care," so a direct measure of efficacy or utility is not possible.

Fall risk may be reduced by interventions to address hazards in the home and hazards related to reduced vision. A structured interdisciplinary medical and occupational therapy assessment of elderly people who have fallen with referral to relevant services, if indicated, can significantly decrease the risk of further falls and limit functional impairment (42). A home safety program may reduce falls and may be more cost-effective than an exercise program in elderly people with poor vision (32). A systematic review and meta-analysis of home hazard modification programs for reducing falls in older adults found that falls can be significantly reduced with the use of home modification interventions that are thorough, well-focused, have an environmental-fit perspective, and have adequate follow-up (118).

At least on a yearly basis, medical care providers should ask elderly patients about any falls or difficulty with balance and gait. They should critically review all of their medications (both prescription and over-the-counter medications) and substance use. They should also observe their patients while rising from a chair, while standing, and while walking. Any elderly patients who have observed difficulty with getting around should be considered for professionally supervised balance, gait, and muscle-strengthening programs. Such programs, coordinated by physical, occupational, or kinesio-therapists, can decrease the risk of falls by 10%.

Table 5. Recommendations For Community-Dwelling Older Adults to Decrease the Risk of Falls and Fall-Related Injuries*

Multifactorial assessment and intervention. Strategies utilizing multifactorial assessment and intervention can significantly reduce the rate of falling and the risk of fall-associated injury, including hip fractures, especially in those who are dizzy and vertiginous, elderly, frail, or infirm. A meta-analysis found that there was only a 4% decrease in the overall rate of falls for individuals receiving various fall prevention interventions (81). However, results of multifactorial fall prevention programs depend on the components of the programs, the specific patient populations, and the settings (66).

Many strategies for the prevention of falls have been tried, with mixed success. The most successful consider the multifactorial causes of falls and include interventions to improve strength and functional status, reduce environmental hazards, and allow staff to identify and monitor high-risk residents (162).

Although there is no consensus on which components are necessary in such multifactorial intervention programs, components can include educating and guiding staff, modifying the environment, implementing exercise programs, supplying and repairing aids, reviewing and modifying drug regimens, providing hip protectors, and post-fall problem-solving conferences (112; 58; 188). The choice of components should be appropriately targeted to specific patient issues (eg, type of dizziness). Passive wearables (eg, hip protectors), active wearables (eg, instrumented belts with deployable airbags), and compliant flooring have all shown promise in reducing fall-related hip injuries in older adults, but each of these products is accompanied by limited real-world data and significant challenges that must be overcome to maximize effectiveness (188).

Community settings. A community-based multifaceted intervention (incorporating group exercise three times a week, 6 hours of fall prevention education, comprehensive falls risk assessment results sent to primary health care provider) was effective in improving balance, mobility, and leg strength—all known fall risk factors (172).

A low-cost, home-based fall risk reduction program incorporating fall risk education, home-based exercise programming, nutrition counseling, and environmental hazards education for community-dwelling older adults was effective in reducing some of the studied fall-related risk factors over a 10-week period (214).

A modest, one-time prevention program appeared to confer short-term health benefits on ambulatory HMO enrollees, although benefits diminished by the second year of follow-up (205).

Subacute hospital settings. A randomized controlled trial compared a falls prevention program plus usual care with usual care alone in a subacute geriatric/rehabilitation hospital setting (75). Six hundred twenty-six adult patients were recruited from consecutive admissions to three subacute hospital wards in a metropolitan hospital that specialized in geriatric care and rehabilitation; 310 patients were allocated to the intervention group, and 316 patients were allocated to the usual care (control) group. Baseline characteristics, admission diagnoses, and participant days of observation were similar in the two treatment groups. The age of the participants ranged from 38 to 99 years old, with a mean age of 80 years. Due to the study design, complete blinding of hospital staff and participants was not considered feasible. Usual care included weekly medical assessments, daily hour-long physical and occupational therapy sessions, nursing assistance, and other allied health services as required. Participants in the intervention group received a targeted falls prevention program in addition to usual care. The falls prevention program included a falls risk alert symbol placed above the patient's hospital bed, an informational brochure targeted to family members and caregivers, 45-minute exercise sessions three times a week, individualized educational sessions twice a week, and hip protectors. Participants in the intervention group had 30% fewer falls, 22% fewer fallers, and 28% fewer falls resulting in injury than participants in the usual care group. Differences in the number of falls between the two study groups were most apparent after approximately 6 to 7 weeks.

Acute hospital settings. A targeted multifactorial falls prevention program was not effective among older people in hospital wards with relatively short lengths of stay (45). Hospital-based multifactorial falls prevention programs may have initial success followed by a waning of benefit due to suboptimal compliance by staff and patients resulting from staff turnover, high patient volume and turnover, high patient-to-nurse ratios, competing demands on nursing staff, and lack of buy-in from providers (107). Further, scheduled toileting is difficult to implement because patients often choose not to use facilities at their scheduled times. A targeted multifactorial falls prevention program, including an exercise program, education program, and hip protectors, can reduce the incidence of falls and fall-related injuries in the subacute hospital setting (75)

Residential care settings. In a study of residential care facilities in Sweden, an intervention program resulted in a significant reduction in the number of falls, fallers, and hip fractures (91). The intervention program included educating and guiding staff, modifying the environment, implementing exercise programs, supplying and repairing aids, reviewing and modifying drug regimens, providing hip protectors, and post-fall problem-solving conferences.

Meta-analysis of the effectiveness of interventions to prevent falls in older adults. In a meta-analysis of the effectiveness of interventions to prevent falls in older adults, interventions considered included multifactorial falls risk assessment and management programs, exercise, environmental modifications, and education (38). Environmental modifications included a home visit by a professional who recommended modifications to address various hazards, such as poor lighting or loose rugs or carpets. Educational interventions considered were diverse and included general educational efforts directed at high-risk groups in the community (such as pamphlets and posters at nursing homes) as well as studies of targeted one-on-one counseling sessions. Forty relevant trials were identified comparing various interventions with usual care. None of the studies directly assessed the relative effectiveness of different fall prevention approaches compared with usual care.

A multifactorial falls risk assessment and management program was the most effective management approach to decrease the risk of falling and the monthly rate of falling. Among patients who fell at least once, such programs decreased the risk of falling by 18%. The monthly rate of falling was decreased by 37%, with 12 fewer falls overall per 100 patients per month.

Exercise programs were also beneficial and decreased the risk of falling by 14% among patients who fell at least once. The monthly rate of falling was decreased by 14%, with three fewer falls overall per 100 patients per month.

Environmental modification and education were the primary interventions in only a few trials, and the pooled effect estimates for these interventions were not statistically significant.

Restraint use as a strategy for prevention of falls. The U.S. Centers for Medicare & Medicaid Services (CMS) defines physical restraints as "any manual method or physical or mechanical device, material, or equipment attached to or adjacent to the resident's body that the individual cannot remove easily which restricts freedom of movement or normal access to [one's] body." CMS defines chemical restraints as "any drug used for discipline or convenience and not required to treat medical symptoms."

Typically, restraints have been used as a strategy to prevent falls or wandering, to control agitation, and to protect medical devices (121; 28; 106; 102; 101; 186; 77), even if the use of physical restraints is poorly connected with the estimated fall risk (106; 102; 101). Restraint use has often been portrayed as a means of limiting injury risk, especially in confused or agitated patients (129; 64), but the act of restraining may itself contribute to manifestations of agitation (210). The "archetypical restrained patient was older, new on the unit, had altered thought processes and a high potential for injury, and required extensive nursing care" (122). Often an unstated rationale is limiting perceived liability risk for staff and institutions responsible for the care of impaired and frail elderly people, with seldom a consideration of how this adversely impacts their independence, autonomy, or rights (129). Nurse and physician communication regarding restraints is often poor, with poor agreement between the nurse and physician as to the reason for restraint in an individual patient (121).

A systematic review identified nine studies addressing the effectiveness of physical restraints in reducing falls among adults in acute care hospitals and nursing homes: most of the studies found that physical restraints do not reduce falls, and decreased restraint use does not contribute to more falls (185). Some studies have found that removal of restraints increases "nonserious falls" but not serious falls (57; 139), but even if minor injuries and falls may increase in some environments, restraint-free care is safe when a comprehensive assessment is done and restraint alternatives are used (139). Nevertheless, physical restraints are, in general, not effective in reducing falls or injuries among adults in acute care hospitals and nursing homes, and "restraint use imposes more risk of falls and other undesirable outcomes than it prevents" (151; 41; 67; 33; 34; 35; 61; 184; 72; 54; 185). Observational studies relating use of restraints to the risk of falls are subject to potential biases and limitations because orders for restraint use may not reflect actual restraint use at the time of a fall (171). In any case, no definitive evidence that restraints protect hospitalized patients from falling has been forthcoming (171), and restraint use may be associated with increased severity of injury among hospitalized patients who fall (194; 187).

In inpatient rehabilitation, male sex, decreased mental status, low admission functional independence measure score, stroke, or traumatic brain injury were closely associated with restraint use (167). Falls occurred in 25% of restrained and 10% of unrestrained patients. Although physician orders were required to apply restraints, nursing staff initiated, monitored, and discontinued restraint use independently.

In a cross-sectional survey of 1710 nursing home residents in Hong Kong examining factors associated with falls and use of restraints, men, older aged individuals, and those with impaired vision, dementia, or taking psychotropic drugs were more likely to fall (115). Restraints are very commonly (68%) used in Hong Kong nursing homes, and the use of any type of restraint in this population was only very marginally associated with fewer falls.

Furthermore, restraints may produce serious adverse outcomes, including functional loss, immobility, delirium, pressure sores, falls or injuries, other physical or psychological harm, loss of dignity, and violation of a patient's rights (07; 10; 09; 08; 11; 165). A public awareness campaign in Colorado used the slogan, "Restraints Have Risks!" (141).

In a case-control study of fall-related risk factors for elderly hospitalized patients, 252 elderly patients (aged 60 to 85 years) who fell during the period between March and December 1993 in a large metropolitan hospital were matched with 250 randomly selected elderly patients by length of stay to the day of the fall (11). Individuals who had been placed in a mechanical restraint during their hospital stay prior to the fall (for the cases) or the matched day of hospitalization (for the controls) had approximately twice the risk of falling as did patients who had not been placed in restraints. Cases and controls did not differ in participation in occupational therapy, physical therapy, or cardiac rehabilitation, and staffing adequacy was similar for both groups. The risk of falling was highest soon after a patient was placed in a mechanical restraint.

Despite the lack of clear benefits to restraints, the many negative potential effects of restraints, and the alternatives available, physical restraints were, and still are, commonly used in hospitals and nursing homes (195; 33; 34; 35; 100; 101; 156; 157; 156; 157; 11; 77; 26; 78; 123; 115; 185).

The Centers for Medicare & Medicaid Services (CMS) implemented the Hospital-Acquired Conditions (HACs) initiative in October 2008, and the CMS no longer reimburses hospitals for fall injury. In a 9-year retrospective cohort study (July 2006–December 2015) involving 2862 adult medical, medical-surgical, and surgical nursing units from 734 hospitals, physical restraint use prevalence decreased from 1.6% to 0.6% (171). Changes in the rates of falls, injurious falls, and restraint use varied according to hospital bed size and teaching status. However, since the HACs initiative, there has been a modest decline in the rates of falls and injurious falls, primarily in larger teaching hospitals. Nevertheless, falls remain a difficult patient safety problem for hospitals without cost-effective, generalizable strategies for their prevention.

Clinicians need to be aware of appropriate, albeit quite limited, indications for the use of physical restraints, risks and perceived benefits of physical restraints, and the effectiveness of alternative measures. The use of physical restraints should be assessed by trained staff based on individualized need and consideration of patient autonomy and integrity. Health care providers and institutions need to employ preventive strategies and innovative alternatives as well as process improvements to avoid restraints and thereby improve patients' overall well-being, health, and safety (28; 25; 139; 165).

One reasonable alternative to restraints is a nonintrusive monitor to reduce falls (103).

Differential diagnosis

Confusing conditions

Falls are generally fairly obvious, with the clinical unknowns most often involving the circumstances or causes of a fall rather than whether or not there was a fall.

Diagnostic workup

Table 6. Key Diagnostic Tests

A meta-analysis of the effects of active vitamin D analogues on muscle strength and falls in elderly people found that the use of active vitamin D analogues does not improve global muscle, hand grip, or back extensor strength but improves quadriceps strength and reduces the risk of falls (213). However, intermittent or single high-dose vitamin D supplementation has no preventive effect on the risk of falls and fractures and might even increase the risk of falls (136).

Osteoporosis

After menopause, both the quantity and quality of bone decline rapidly, resulting in a dramatic increase in the risk of fracture in postmenopausal women. Unfortunately, osteoporosis is frequently undetected until a fracture occurs. Identification of postmenopausal women at high risk of fracture is especially important for early postmenopausal women who can benefit from early intervention to maintain or to increase bone mass and, thus, reduce the risk of fracture.

Most authorities recommend osteoporosis risk-factor assessment for all postmenopausal women and bone mineral density measurements for women at highest risk (ie, women aged 65 years or older, postmenopausal women younger than 65 years old with one or more additional risk factors for osteoporosis, and postmenopausal women with fragility fractures) (49).

Risk stratification for medically appropriate and cost-effective treatment is facilitated by the World Health Organization (WHO) FRAX algorithm, which uses clinical risk factors, bone mineral density, and country-specific fracture and mortality data to quantify a patient's 10-year probability of a hip or major osteoporotic fracture (173). Included risk factors comprise femoral neck bone mineral density, prior fractures, parental hip fracture history, age, gender, body mass index, ethnicity, smoking, alcohol use, glucocorticoid use, rheumatoid arthritis, and secondary osteoporosis. FRAX was developed by the WHO to be applicable to both postmenopausal women and men aged 40 to 90 years; the National Osteoporosis Foundation Clinician's Guide focuses on its utility in postmenopausal women and men over 50 years of age. It is validated to be used in untreated patients only. The current National Osteoporosis Foundation Guide recommends treating patients with FRAX 10-year risk scores of 3% or more for hip fracture or 20% or more for major osteoporotic fracture to reduce their fracture risk. Additional risk factors not represented in FRAX, such as frequent falls, warrant individual clinical judgment.

Bone densitometry

Bone densitometry is the integration of bone densitometry (DXA studies) and clinical risk factors (WHO FRAX) in clinical decision-making for low bone density and osteoporosis. For postmenopausal women and men aged 50 years and older, T-score criteria from bone densitometry are used to categorize patients into three groups:

1. Normal: T-score > -1.0 at all sites
2. Low bone density (osteopenia): T-score -1.0 to > -2.5
3. Osteoporosis: T-score </= -2.5

Table sites utilized for bone mineral density measurement include the following:

Premenopausal women and men under 50 years of age. The World Health Organization diagnostic classification for bone mineral density should not be applied to premenopausal women and men under 50 years of age. For these groups, the diagnosis of osteoporosis should not be made based on densitometric criteria alone. In such individuals, ethnicity-, race-, and age-adjusted Z scores should be used. Z-scores of -2.0 or lower indicate "low bone mineral density for age." Z-scores greater than -2 are "within the expected range for age."

Trends. The following trends have been identified:

1. The purpose of monitoring medical management of low bone mineral density is to "ensure reduction of future fracture risk, stabilize or increase bone mass, and preserve or improve bone quality and strength."

2. Central DXA assessment of the hip or spine is the gold standard for serial assessment of bone mineral density.

3. Changes of 3% to 6% at the hip and 2% to 4% at the spine may be due to imprecision of the test.

4. Precision varies with the confidence intervals used, DXA scanner, measurement site, patient population, and technologist skill with patient positioning.

5. Serial DXA studies should be performed according to medical necessity, but in general, usually after intervals of at least 1 to 2 years. Serial studies for patients with osteoporosis can be considered at 1- to 2-year intervals, whereas serial studies for those with severe osteopenia, especially with risk factors for osteoporosis, can be considered at 2- to 3-year intervals, and serial studies for other at-risk individuals can be considered at 3- to 5-year intervals.

Treatment thresholds (138). Recommendations for pharmacological treatment of low bone mineral density are based on the U.S. adaptation of the WHO 10-year fracture probability model that includes U.S. hip fracture incidence and mortality rates and algorithms that incorporate the costs and health consequences of clinical osteoporotic fractures (208).

For postmenopausal women and men 50 years of age and older, treatment is supported by ANY of the following:

Notes. The following notes should be considered:

1. Although the guidelines support treating patients with low bone mass with associated risk factors, there is no evidence for reduction in fracture risk in treated patients with T-scores greater than -1.5, and there is no strong evidence for effectiveness at reducing fracture risk in patients with T-scores better than -2.0 without a previous fracture.

2. Sometimes recommendations are inconsistent between T-score-based recommendations and those based on the WHO FRAX Model. In such situations, it is reasonable to use the more liberal model (ie, consider treatment if either criterion is met).

3. Other factors should be considered when making treatment decisions for individual patients, including risk of falling, severity of concurrent medical conditions, duration and dosage of adjunctive medical therapy for treatment of various diseases, toxicity of agents considered (especially in light of individual comorbidities), etc.

4. Secondary etiologies for loss of bone mineral density are not included in the FRAX calculation of risk when bone mineral density is inserted in the model. The assumption is that the associated risk is mediated through changes in bone mineral density. Thus, other potential contributing secondary etiologies are not always included as a clinical risk factor in calculating fracture risk.

Management

Neurologists are frequently asked to see elderly people who have fallen to identify neurologic contributing factors for their falls and to evaluate neurologic complications of the falls. It is helpful to have a systematic approach to this problem and to be familiar with the potential benefits of a multidisciplinary assessment and intervention.

An extensive resource from the U.S. Centers for Disease Control and Prevention lists effective fall interventions for community-dwelling older adults (179).

Table 7. Management Considerations for Patients Who Fall

Calcium and vitamin D supplementation. Vitamin D3 and calcium supplements are helpful in the prevention of hip fractures in elderly women living in nursing homes, with a reduction of about 25% in the number of hip fractures noted in a 3-year controlled study (126).

Patients with a history of falling and vitamin D insufficiency living in sunny climates benefit from ergocalciferol supplementation in addition to calcium, which is associated with a 19% reduction in the relative risk of falling, mostly in winter (152).

An oral vitamin D dose of 400 IU/day is not sufficient for fracture prevention (23), and doses of supplemental vitamin D of less than 700 IU/day or serum 25-hydroxyvitamin D concentrations of less than 60 nmol/l may not reduce the risk of falling among older individuals (23; 21).

Oral vitamin D supplementation between 700 to 800 IU/d appears to reduce the risk of hip and any nonvertebral fractures in ambulatory or institutionalized elderly persons (23). Supplemental vitamin D in a dose of 700 to 1000 IU/day reduced the risk of falling among older individuals by about 20% and to a similar degree as active forms of vitamin D [eg, 1alpha-hydroxy vitamin D3 (1alpha-hydroxy calciferol) or 1,25-dihydroxy vitamin D3 (1,25-dihydroxy cholecalciferol)] (21a; 21b). Vitamin D2 at a dose of at least 800 IU is the only medication with compelling evidence of effectiveness for fracture prevention in older men (21b; 193).

There is not uniform agreement on the utility of calcium and vitamin D supplementation in elderly individuals, with or without a history of fracture. Some studies did not find sufficient evidence to support routine oral supplementation with calcium and vitamin D3, either alone or in combination, for the prevention of further fractures in previously mobile elderly people (68).

In addition, a randomized controlled clinical trial found no evidence that calcium and vitamin D supplementation (at the doses given) reduces the risk of clinical fractures in women with one or more risk factors for hip fracture (150). The trial included 3314 women aged 70 and over with one or more risk factors for hip fracture [ie, any previous fracture, low body weight (< 58 kg), smoker, family history of hip fracture, or fair or poor self-reported health]. One group received daily oral supplementation using 1000 mg calcium with 800 IU cholecalciferol and an information leaflet on dietary calcium intake and prevention of falls; the control group received only the leaflet. After a median follow-up of 25 months, clinical fracture rates were lower than expected in both groups but did not significantly differ for all clinical fractures or for hip fractures.

A randomized controlled trial with a cluster design found no evidence that vitamin D prevents fractures or falls in elderly people in an elder care home setting (113). The study included 3717 residents from 118 elder care homes throughout Britain (average age 85 years). The units provided mainly or entirely residential care (35%), nursing care (42%), or care for elderly mentally infirm residents (23%). Participants were randomly allocated by residential unit to a treated group that was offered ergocalciferol 2.5 mg every 3 months (equivalent to a daily dose of 1100 IU) or to a control group. After a median follow-up of 10 months, 3.6% of 1762 vitamin D–treated residents and 2.6% of 1955 controls had one or more nonvertebral fractures, and 1.3% and 1.0%, respectively, had a hip fracture. The proportion reporting at least one fall was 44% in vitamin D-treated residents and 43% in control residents. The differences between the vitamin D and control groups were not statistically significant. The pretreatment serum 25-hydroxy vitamin D concentration was high (median 47 nmol/l, measured in a 1% sample).

Osteoporosis. The aim of treatment of postmenopausal osteoporosis is to reduce the frequency of vertebral and nonvertebral fractures (especially at the hip), which are responsible for morbidity associated with the disease (50).

Nonpharmacologic interventions can reduce the risk of fracture, including a balanced diet with adequate intake of calcium and vitamin D, regular exercise, measures to prevent falls or to minimize their impact, smoking cessation, and moderation of alcohol intake (49).

It is cost-effective to treat patients with a fragility fracture and those with osteoporosis, as well as older individuals at average risk and osteopenic patients with additional risk factors, using WHO criteria (47).

Several pharmacologic agents, including the bisphosphonates (eg, alendronate, risedronate, and ibandronate) and the selective estrogen receptor modulator, raloxifene, increase bone mass and reduce fracture risk and have acceptable side-effect profiles (49). Accelerated bone loss and increased fracture risk occur after withdrawal of estrogen treatment.

Results of large placebo-controlled trials have shown that alendronate, raloxifene, risedronate, the 1-34 fragment of parathyroid hormone, and nasal calcitonin greatly reduce the risk of vertebral fractures (50). A large reduction of nonvertebral fractures has been shown for alendronate, risedronate, and the 1-34 fragment of parathyroid hormone. Calcium and vitamin D supplementation is not sufficient to treat individuals with osteoporosis but is useful, especially in elderly women in nursing homes. Hormone replacement therapy remains a valuable option for the prevention of osteoporosis in early postmenopausal women. The choice of treatment depends on age; the presence or absence of prevalent fractures, especially at the spine; and the degree of bone mineral density measured at the spine and hip. Nonpharmacological interventions include adequate calcium intake and diet, selected exercise programs, reduction of other risk factors for osteoporotic fractures, and reduction of the risk of falls in elderly individuals.

Exercise program. Multiple studies have found that exercise programs can decrease falls and falls with injury among most fall-prone elderly individuals across a variety of settings, with a noted exception being those with significant cognitive impairment (153; 163; 91; 79; 104; 56). However, exercise needs to be maintained to be effective; for example, although exercise is recommended as a fall prevention strategy for older people living in aged care who are willing and able to participate (moderate certainty evidence), exercise has little or no lasting effect on falls after the end of a program (high certainty evidence) (56).

A randomized controlled trial studied the effects of a low- to moderate-intensity group exercise program on strength, endurance, mobility, and fall rates in 59 fall-prone elderly community-living men (mean age = 74 years) with chronic impairments and specific fall risk factors (ie, leg weakness, impaired gait or balance, previous falls) (163). Subjects were randomly assigned to a control group (n = 28) or to a 12-week group exercise program (n = 31). Exercise sessions (90 minutes, three times per week) focused on increasing strength and endurance and improving mobility and balance. Increased physical activity was associated with reduced fall rates when adjusted for level of activity (6 falls/1000 hours of activity vs. 16.2 falls/1000 hours, p < .05).

Compared with a general wellness program, an 18-month exercise program with an emphasis on exercise intensity significantly improved bone mineral density and fall risk in elderly women without increasing direct costs (104).

A 6-week, supervised, low-to-moderate intensity program of stretching, postural control, endurance walking, and coordination exercises designed to improve balance and mobility did not significantly improve rates of falls or injuries relative to baseline levels (125).

Regular, moderate physical activity in people with osteoporosis can reduce the risk of falls and fractures, decrease pain, and improve fitness and overall quality of life. It may also stimulate bone gain and decrease bone loss (153). Community group exercise programs are beneficial.

Randomized controlled intervention trials showed only limited effectiveness of physical training in patients with cognitive impairment (79).

In an institutional setting, an interdisciplinary and multifactorial prevention program targeting residents, staff, and the environment may reduce falls and femoral fractures (91). The strategies comprised educating staff, modifying the environment, implementing exercise programs, supplying and repairing aids, reviewing drug regimens, providing free hip protectors, having post-fall problem-solving conferences, and guiding staff.

The U.S. Centers for Disease Control and Prevention (CDC) and the STEADI (Stopping Elderly Accidents, Deaths & Injuries) Program advocate a simple "chair rise" exercise for the elderly. Patients should work up to two sets of 10 to 15 repetitions. Lifting is to be done with the legs and hips, without an arm push off. The back is to be kept straight. The return to a seated position is to be done smoothly and carefully without "collapsing" down onto the chair.

A meta-analysis of the effects of exercise on falls and fall-related injuries among individuals with neurodegenerative diseases and frail elderly people (64 studies with a total of 13,241 participants) found that exercise is effective in reducing neurodegenerative disease as well as aging-associated falls and consequent injuries (59). Exercise is effective in reducing falls for frail aging people and participants with neurodegenerative diseases (including dementia, Parkinson disease, and stroke). Exercise also reduced fall-related injuries in subjects with neurodegenerative diseases and decreased fractures and fall-related injuries among frail elderly people. For fall prevention, balance and combined exercise protocols are both effective, and either short-, moderate-, or long-term intervention duration is beneficial. More importantly, exercise has a very low injury rate per participant year (0.007) and has relatively good compliance (75%).

Hip protectors. The effectiveness of anatomically designed external hip protectors in preventing hip fractures has been demonstrated in several biomechanical and clinical studies (112; 98), but previous clinical studies of efficacy were often conflicting, partly because adherence was often low (112; 98; 199; 204; 83). Hip protectors with a more patient-friendly design are now available (including a “fly” for male patients) and may be better tolerated and prove to be of greater utility in preventing hip fractures.

Table 8. Environmental Changes That Can Decrease Falling

Restraints. Strategies that reduce mobility through the use of restraints have been shown to be more harmful than beneficial and should be avoided.

Selected disease-specific considerations

Parkinson disease. Atypical antipsychotic medications increase fracture risk in patients with Parkinson disease (53). Use of quetiapine is associated with a more than 2-fold increased risk of fracture, whereas olanzapine had a 70% higher rate and risperidone a 20% higher rate. The higher-than-expected rate of fractures among those on quetiapine may result from greater sedation, somnolence, dizziness, and orthostatic hypotension.

Although withdrawal of psychotropic medications may significantly reduce the risk of falls in elderly patients on such medications, permanent withdrawal is very difficult to achieve (31).

A 6-week personalized home-based program of exercises and strategies for repeat fallers with Parkinson disease improved functional reach and quality of life and produced a consistent trend toward lower fall rates at both 8 weeks and 6 months and lower rates of injurious falls needing medical attention at 6 months (12).

Special considerations

Pregnancy

Falls during pregnancy occur in about one quarter of women and can cause serious harm to both the mother and the fetus (85). In a systematic review that incorporated 14 studies, identified extrinsic risk factors for falls during pregnancy included slippery floors, cluttered areas, uneven ground, inappropriate shoes, hurrying, walking on stairs, carrying additional loads, poor lighting or obstructed view, sedentary lifestyle, and working in physically demanding jobs. Identified intrinsic factors included age less than 30 years, height more than 160 cm, advanced pregnancy, unintended pregnancy, multiparity, hyperemesis gravidarum, low back pain, gestational diabetes, lower ankle stiffness, and joint laxity. Certainly, many of the identified extrinsic factors are amenable to prenatal counseling.