Falls
Epidemiology of injurious falls
Evaluation of the elderly individual who falls
INTRODUCTION
The incidence of falls and related injuries will continue to pose significant challenges to healthcare systems globally as the world’s population ages. This chapter describes the epidemiology, aetiologies, consequences and prevention of falls in the community, acute and long-term care settings.
DEFINITION OF FALLS
Earlier falls research employed inconsistent definitions of falls in the older age group, but one of the most accepted is that by Tinetti et al.1 which defined a fall in the non-hospitalized geriatric patient as ‘an event which results in a person coming to rest unintentionally on the ground or lower level, not as a result of a major intrinsic event (such as a stroke) or overwhelming hazard’. A similar definition is also used for the in-patient and long-term setting. The International Classification of Diseases, Tenth Revision defines several codes for falls, each with broad descriptions reflecting the place of occurrence and activity, such as a fall on the same level from slipping, tripping and stumbling (w01), to other fall on the same level including a fall from bumping against an object and from or off a toilet (w18).
EPIDEMIOLOGY OF FALLS
A myriad of studies on falls over the last three decades have reported different incidences. Varied definitions of falls used, marked heterogeneity of studied populations, differences in data collection methods, poor recall by participants particularly in retrospective studies and under-reporting of falls in cases without associated injuries are contributing factors.
Community setting
Approximately one in three adults above the age of 65 fall at least once in a year, increasing to more than one in two in the 85+ age group. Of those who fall in one year, half to two-thirds experience a repeat fall in the subsequent year. The majority of falls occur indoors and during the day, in a person’s usual place of residence and in frequently used rooms such as bedrooms, kitchens and dining rooms. Those who fall indoors tend to be older, are female and have various indicators of poor health and frailty, while outdoor fallers are younger, male and are relatively physically active and healthy. US studies suggest similar rates of falls between different races, but a recent systematic review reported consistently lower fall prevalence rates for East Asians compared to Western populations.
Long-term care setting
Falls incidence in nursing care facilities are about three times those in the community, with more than half of residents experiencing at least one fall annually. Falls frequently occur in the residents’ rooms or in bathrooms (75%), during transfers (41%) and when walking (36%). Most falls are observed between 10 am and midday and between 2 pm and 8 pm. Men fall more often than women and falls are less common in people requiring the least and highest levels of care.
Acute care setting
Although falls incidence varies with different wards and different healthcare systems, annual fall incidence rates in studies of hospitalized elderly patients reported double the rates of community dwelling populations. Nearly half of patients in stroke rehabilitation wards report at least one fall during an admission, and high rates of falls also occur in psychogeriatric and geriatric rehabilitation units.
The falls incidence of in-patients with any recent clinical fracture is also high. Within 3 months post-fracture, the falls rate is 15%, rising to 3.5 falls per patient-year. After hip fractures, up to one in two people fall again at least once within 2–12 months after subacute rehabilitation and 28% fall more than once, resulting in a new fracture in 12% of cases and a second hip fracture in 5% of cases.
EPIDEMIOLOGY OF INJURIOUS FALLS
In the community setting, 40–60% of falls lead to injuries: 30–50% result in minor injuries, 5–6% in major injuries other than fractures and 5% in fractures. In 2011, unintentional falls was the eighth leading cause of all deaths in the 65+ age group, but the most common cause of injury death and the most common cause of non-fatal injuries treated in US hospital emergency departments.
Numerous risk factors are associated with injurious falls, with the most important being lower-extremity muscle weakness, peripheral neuropathy, low pulmonary capacity, difficulties in gait, long acting benzodiazepines, cardiovascular medications and cognitive impairment (see Tables 12.1 and 12.2).
In the setting of long-term care, about 4% of falls (range, 1–10%) result in fractures, while other serious injuries such as head trauma, soft-tissue injuries and severe lacerations occur in about 11% of falls (range, 1–36%). Risk factors for these injurious falls are the same as those for falls in general, however, some risk factors show strong links, such as female sex, functional independence, the number of falls and use of mechanical restraints.
In the acute care setting, 30% of the hospitalized patients who fall suffer injuries, of which 4–6% are severe, including fractures, subdural haematomas, bleeding and even death.
CAUSES/RISK FACTORS
Community setting
A myriad of risk factors for falls have been identified and are classified broadly as intrinsic and extrinsic factors. Intrinsic factors generally include individuals’ age related decline in systems involved in balance performance that may precipitate a fall, while extrinsic factors include medications, environmental hazards and hazardous activities. Risk factors have also been described in terms of possible causative categories, such as environment, medication, medical conditions, age related physiological changes, nutrition and lack of physical activity.
In the community setting early pivotal research of adults over 65 by Tinetti identified a number of risk factors: use of any sedative hypnotics or benzodiazepine, polypharmacy (four or more medications), postural hypotension, environmental hazards, muscular strength and range of motion impairments. Furthermore, half of the falls occurred in the presence of environmental hazards and only 5% and 10% during hazardous activity and acute illness, respectively. However more importantly, the risk of falling increased linearly with the number of risk factors, from 8% with none to 78% with four or more combination of risk factors.
Subsequent population studies since have identified a number of other risk factors (Table 12.1), however each risk factor on its own may not independently increase the risk of falling. The strongest associations were history of falls, gait problems, walking aids use, vertigo, Parkinson’s disease and antiepileptic drug use. The others were less associated and some traditional factors such as orthostatic hypotension were not associated (Table 12.2).
Some of the more established risk factors are discussed in more detail below.
GAIT AND BALANCE
Gait disturbance is prevalent in older people. With age a decline in body-orienting reflexes, muscle strength and tone, step length and height produce a gait pattern that is stiffer and less coordinated with poorer posture control. The ability to shift weight or execute reach-to-grasp reactions rapidly or adopt an effective stepping reaction to maintain balance or recover equilibrium quickly after a perturbation is also impaired. Walking speed is slower with shorter stride length, greater propensity of landing flat-footed, less lateral sway, smaller ankle plantar flexion and hip extension during push-off and greater tendency to misstep.
Slower gait speed is the most significant gait parameter associated with increased risk of falls and is independent of other factors such as cognitive impairment and disability. Poorer performance on swing, double-support phase, swing time variability and stride length variability are also predictive of falls. Not surprisingly, both neurological gait pathologies (e.g. hemiparetic, frontal, Parkinsonian, unsteady, neuropathic and spastic) and ‘higher-level gait disorder’ are associated with increased risk of falls. Unsteady and neuropathic gait in particular are two gait subtypes that predict risk of falls independent of disability and cognitive status.
Overall, balance impairment alone confers only a moderate increase in falls risk in community dwelling older adults. However, differences in fall outcome, length of follow-up and balance measurement tools may underestimate the magnitude of association.
COGNITION
Global measures of cognition are linked with fall related injuries, and a diagnosis of dementia, in both community and institution dwelling older adults, confers a high risk for any fall and recurrent falls. However, even those with mild cognitive impairment are twice as likely to experience a fall as those with normal cognition.
Table 12.1 Risk factors for falls reported in epidemiological studies
Four cognitive domains are thought to influence fall prevalence: attention (especially dual tasking), executive function, information processing and reaction time. Among the subtypes of attention, divided attention, or dual tasking, has been found to be most related to balance, gait and fall risk. In situations which require both attention to a task and gait, subjects with limited attention capacity show a decline in one task or both. This deficit is evident in elderly fallers and patients with neurological disease, such as stroke, Alzheimer’s disease or Parkinson’s disease compared with healthy older adults.
In addition, systematic reviews have also demonstrated consistent evidence that executive function and dual-task performance are highly associated with falls or falls risk independently of their association with declines in gait speed.2
PHYSIOLOGICAL FACTORS
The maintenance of balance depends on the interaction of various sensory and motor inputs and integration by the central nervous system. With age there is a decline in the functioning of each sensori-motor component, and impairments in each function are associated with increased risk of falling. Impaired depth perception, reduced visual contrast sensitivity, reduced proprioception and poorer tactile sensitivity in the lower limbs, reduced quadriceps strength, slow reaction time, and impaired static and dynamic balance predict falls and multiple falls. Muscle strength, particularly of the lower extremities, decreases with age and is associated with falls. The falls incidence in frail older people is three times that of their more active and vigorous counterparts. Having a walking speed in the range stroll/very slow/non-ambulant compared with normal/brisk/fast doubles the risk of falls. In fact, slow gait velocity (<0.7 m/s) at baseline is associated with future adverse events, hospitalizations, new falls and requirement for a caregiver.
Table 12.2 Risk factors for falls in the community setting
MEDICATIONS
Several commonly used medications have shown association with falls, but each class confers different levels of risk (Table 12.3).3 Benzodiazepines, antipsychotics (both atypical and typical) and antidepressants, in particular selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants (TCAs), significantly increase the risk of falls. Psychotropic medications, which include those used in dementia, increase the risk of falling by 47% in older adults living in the community. This risk is further increased when two or more psychotropic drugs are combined. Benzodiazepine use in the older population is prevalent and adverse outcomes often involve prolonged exposure, but the mechanisms affecting the risk of falls may differ across benzodiazepines. For example, the cumulative duration of its use in the previous 30 days for alprazolam, 2 previous weeks’ exposure for flurazepam and temazepam use has a statistically significant association with the risk of fall related injuries only when the effects of both the cumulative use and current dose are jointly taken into account. The risk of fall related injuries associated with cumulative use of temazepam may be driven mostly by a withdrawal effect, especially among the elderly who have been using it over a prolonged period.
Table 12.3 Drug class and risk of falls
Falls are associated with other medications as well, such as antihypertensive agents. They are associated with a 24% increased odds of falling but the relationship between initiating different antihypertensive medications and the occurrence of falls and fractures is variable. Insulin treated patients are at increased risk compared to non-insulin treated patients and non-diabetics. Metformin and secretagogues are not linked to falls but patients on thiazolidinediones are more prone to having fractures after a fall. Surprisingly, β-blockers and opiates are not independently associated with increased risk of falls unless orthostatic hypotension is present. Other cardiovascular medications, however, are implicated in fall risk, including digoxin, type 1a anti-arrhythmics and diuretics. With diuretics there is an increased risk in falling the day following a change in diuretics medication or an increase in the dose.
In addition to the effects of individual drugs or drug class, the interactions between multiple drugs, that is, polypharmacy, and in particular multiple drugs that have associations with falls, that is, falls risk increasing drugs (FRIDs), are important factors to consider. The prevalence of falls increases with overall number of drugs prescribed, certain drug combinations (e.g. a TCA plus any hypotension producing drug) and multiple FRIDs. Polypharmacy and in particular in the context of its association with falls has been defined as four or more medications. However the optimal discriminating number of concomitant medications associated with not only falls but also frailty, disability and mortality is five or more. The number of medications on hospital discharge and FRIDs on discharge in one study also was significantly associated with recurrent falls. More importantly, recurrent falls were most likely to occur with only 1.5 FRIDs in the frail and 2.5 FRIDs in the robust. In this study antidepressants and anxiolytics were the most frequently dispensed FRIDs.
VERTIGO AND DIZZINESS
Vestibular dysfunction is common in older people and often results in impairments in posture and gait, characterized by postural instability and a broad-based, staggering gait pattern with unsteady turns placing the older adult at an increased risk of recurrent falls. However, there is no clear causal relationship between age-associated changes in vestibular function and falls when visual and peripheral sensations are intact.
VISION IMPAIRMENT
Intact vision is integral to enable planning and coordination of movements in response to environmental hazards as well as assisting with the maintenance of balance. Inputs from the visual, vestibular and somatosensory systems are integrated centrally and instructions are sent to the motor system to maintain balance. Optical flow provides information about anteroposterior body sway and information from eye movements provides information about lateral body sway. Both the central and peripheral visual system assesses optical flow and thus postural control. It is not surprising that standing postural control is poorer in individuals with cataract, age related macular degeneration or glaucoma.
Vision is also used to scan the travel pathway for obstacles and changes in terrain, typically one to two steps ahead to enable safe ambulation through the environment and when negotiating steps and stairs. The lower peripheral visual fields in particular provide exproprioception (position of the lower limbs relative to the environment) information, which is used to fine tune the gait. In this respect vision plays a major role in successful stair negotiation.
Visual impairment increases with age affecting 3.1% of 65–74 year olds, 11.6% of 75–84 year olds and 35.5% of 85+ year old individuals. The most prevalent age related causes are cataract, macular degeneration, glaucoma and presbyopia. Glaucoma sufferers have limited perception due to restricted visual fields and loss of peripheral vision. Individuals with age related macular degeneration have poorer vision, slower visual reaction times, and visuomotor and balance deficits, and are prone to clumsiness and increased risk of falls.
Aside from a poor level of visual acuity, visual field deficits, impaired depth perception, low contrast sensitivity, stereoacuity and changes in visual acuity demonstrate even stronger links with falls. Surprisingly, the effect of vision improvement through cataract surgery may in fact increase falls rate. This effect on falls rate is similar to that seen in patients with new spectacles. New spectacles may be associated with changes in magnification, optical centres, lens type (e.g. progressive addition lens [PAL] rather than single vision lens) and position of bifocals/PALs, which could adversely affect falls risk. A change from a distance single vision lens to PALs or bifocals distorts the peripheral visual field in PALs and provides a blurred and magnified view of the lower visual field beyond near working distance in both PALs and bifocals. This affects the peripheral optic flow information used for postural control and makes it difficult to judge the position of obstacles in the lower visual field, including obstacles, step and stair edges and/or foot placements relative to such environmental obstacles. Thus the recommendation is that changes to refractive corrections in older people should be conservative, and PALs or bifocals should never be prescribed to patients who are used to wearing single vision glasses and who could be categorized as at high risk for falls.
COMORBIDITIES
A number of chronic diseases are independently associated with falls (Table 12.1) and both number and type of chronic conditions play a role in falls. Fall prevalence increases linearly with the number of chronic conditions, suggesting an additive effect of chronic disease on fall risk, irrespective of the specific condition. However not all clusters of chronic conditions are associated with falls. Patients with a number of chronic conditions, which include hypertension or chronic obstructive pulmonary disease (COPD), tend to be at higher risk of falls. With hypertension both the condition itself and treatment side effects known to induce orthostatic hypotension may be the underlying mechanisms. People with uncontrolled hypertension and orthostatic hypotension are 2.5 times more likely to have recurrent falls than those with uncontrolled hypertension and no orthostatic hypotension.
Similarly, people with COPD have both increased prevalence of falls as well as dysfunction in physiological risk factors for falls, such as impaired postural control. Skeletal muscle dysfunction and cerebral hypoxemia have been postulated as contributing factors.
Acute and subacute care setting
Risk factors for falls in these settings include advanced age, agitation, confusion or disorientation, generalized muscle and/or leg weakness, unstable gait, urinary incontinence, a history of previous falls, visual deficit or the use of certain medications (hypnotics, sedatives, etc). The hospital environment itself, such as the presence/absence of bed rails, the height and stability of any type of seat (including the toilet) or furniture and equipment may present as new obstacles. Also hospitalization itself may make older people become more disoriented or agitated, or suffer a decline in function and thus be at increased risk of falls.
Status at discharge such as a decline in mobility, use of assistive device, cognitive impairment and self-report of confusion after hospital discharge are also major risk factors for falls. More significantly, patients who were functionally dependent and needed professional help after discharge had the highest rate of falls (20.2%).
A recent meta-analysis by Deandrea et al.4 showed a number of risk factors resembling those in the community setting (Table 12.4). Age was less associated and gender did not confer increased risk compared to community setting.
Patients with recent hip fractures present a higher risk group. Recurrent falls affect those who are less mobile and less active, have a higher number of chronic diseases and medications, have higher pre-fracture disability, have chronic heart failure, have lower vitamin D levels, have lower handgrip strength and have reduced quality of life.
Similar to patients without hip fracture in the community, the use of a rollator frame (rolling walker) and nocturnal urinary incontinence are also associated with falls. The measures of impairment before hip fracture (disability, poor vitamin D status and a more sedentary lifestyle), combined with strength and balance impairments persistent after surgery, identified the most vulnerable individuals who would go on to suffer injurious falls again. Among these risk factors, hip abductor weakness postoperatively had the strongest relationship with fall related injuries risk.
Notably, the other risk factors associated with falls after hip fractures resemble those in the population without hip fracture: age, female sex, difficulties in activities of daily living (ADL), orthostatic hypotension and polypharmacy, more handicap, lower Activities-specific Balance Confidence (ABC) Scale and falls efficacy score, worry over further falls, previous falls and poor performance with the 5-metre Timed Up and Go (TUG) test, timed 10-metre walk and the Turn180 test.
Long-term care setting
Prospective studies have shown hip weakness, poor balance and number of prescribed medications to be factors most strongly associated with falling among institutionalized subjects. Other factors reported in the literature include increased age, male sex, higher care classification, incontinence, psychoactive medication use, previous falls and slow reaction times. Use of walking aids, presence of moderate disability, wandering tendencies, Parkinson’s disease and dizziness were other independent risk factors reported in an earlier meta-analysis.4 However in contrast to the community dwelling setting, age, gender, vision impairment, depression, stroke and incontinence were not associated with an increased risk of falls.
Table 12.4 Risk factors for falls in in-patient settings
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
