Afl35(s2).book(afl084.fm)

Age and Ageing 2006; 35-S2: ii37–ii41
doi:10.1093/ageing/afl084
CLINICAL RISK ASSESSMENT, INTERVENTIONS AND SERVICES
Falls in older people: epidemiology, risk factors
and strategies for prevention
UCLA School of Medicine and Geriatric Research Education and Clinical Center (GRECC), VA Medical Center, Sepulveda, CA 91343, USA Address correspondence to: Laurence Z. Rubenstein. E-mail: lzrubens@ucla.edu or laurence.rubenstein@med.va.gov Abstract
Falls are a common and often devastating problem among older people, causing a tremendous amount of morbidity, mortalityand use of health care services including premature nursing home admissions. Most of these falls are associated with one ormore identifiable risk factors (e.g. weakness, unsteady gait, confusion and certain medications), and research has shown thatattention to these risk factors can significantly reduce rates of falling. Considerable evidence now documents that the mosteffective (and cost-effective) fall reduction programmes have involved systematic fall risk assessment and targeted inter-ventions, exercise programmes and environmental-inspection and hazard-reduction programmes. These findings have beensubstantiated by careful meta-analysis of large numbers of controlled clinical trials and by consensus panels of experts whohave developed evidence-based practice guidelines for fall prevention and management. Medical assessment of fall risks andprovision of appropriate interventions are challenging because of the complex nature of falls. Optimal approaches involveinterdisciplinary collaboration in assessment and interventions, particularly exercise, attention to co-existing medical condi-tions and environmental inspection and hazard abatement.
Keywords: geriatrics, fall prevention Background and epidemiology
or require hospitalisation. Moreover, the rates of falls and theirassociated complications rise steadily with age and are about Falls and unstable balance rank high among serious clinical twice these figures for persons aged >75 years. Persons living problems faced by older adults. They are a cause of substan- in long-term care institutions have much higher rates (0.6–3.6 tial rates of mortality and morbidity as well as major con- per bed annually, mean 1.7). Falls among those in institutions tributors to immobility and premature nursing home also tend to result in more serious complications, with 10– placement. Unintentional injuries are the fifth leading cause 25% of such falls resulting in fracture or laceration.
of death in older adults (after cardiovascular disease, cancer, The way in which a person falls often determines the stroke and pulmonary disorders), and falls constitute type of injury sustained—wrist fractures usually result from two-thirds of these deaths. In the United States, about forward or backward falls onto an outstretched hand and three-fourths of deaths due to falls occur in the 13% of hip fractures typically from falls to the side, whereas back- the population age ≥65, indicative of primarily a geriatric ward falls directly onto the buttocks have much lower rates syndrome. About 40% of this age group living at home will of associated fractures [1]. Wrist fractures are more com- fall at least once each year, and about 1 in 40 of them will be mon than hip fractures between ages 65 and 75, whereas hip hospitalised. Of those admitted to hospital after a fall, only fractures predominate in ages after that, probably reflecting about half will be alive a year later. Repeated falls and insta- slowed reflexes and loss of ability to protect the hip by bility are very common precipitators of nursing home ‘breaking the fall’ with one’s wrist after age 75.
The problem of falls in the elderly population is clearly Many population-based studies have described the epide- more than simply a high incidence, because young children miology of falls for older people in different settings, and rates and athletes certainly have higher incidences of falls than all vary considerably. Lowest rates (0.3–1.6 per person annually, but the frailest elderly groups. Rather, it is a combination of weighted mean 0.65) occur among community-living, gener- a high incidence together with a high susceptibility to injury, ally healthy elderly people (age ≥65). Although most of these because of a high prevalence of clinical diseases (e.g.
falls result in no serious injury, about 5% do induce a fracture osteoporosis) and age-related physiological changes (e.g.
L. Z. Rubenstein
slowed protective reflexes) that make even a relatively mild all decline with ageing and impair ability to avoid a fall after fall particularly dangerous. In addition, recovery from fall an unexpected trip or slip. In old age, the ‘strategy’ for injury is often delayed in older persons, which in turn maintaining balance after a slip shifts from the rapid cor- increases risk of subsequent falls through deconditioning.
recting ‘hip strategy’ (fall avoidance through weight shifts at Another complication is the post-fall anxiety syndrome, in the hip) to the ‘step strategy’ (fall avoidance via a rapid step) which an individual down-regulates activity in a perhaps to total loss of ability to correct in time to prevent a fall.
overcautious fear of falling; this in turn further contributes Age-associated impairments of vision, hearing and memory to deconditioning, weakness and abnormal gait and in the also tend to increase the number of trips and stumbles.
long run may actually increase risk of falls.
The broad category of gait problems and weakness is the The National Health Interview Survey indicates that falls next commonest specific precipitating cause for falls are the largest single cause of restricted activity days among (10–25% in most series). The ability to walk normally older adults, accounting for 18% of restricted days. Moreo- depends on several bio-mechanical components, including ver, fall-related injuries recently accounted for 6% of all free mobility of joints, particularly in the legs; appropriate medical expenditures for persons aged >65. The U.S. Public timing of muscle action; appropriate intensity of muscle Health Service has estimated that two-thirds of deaths due action; and normal sensory input, including vision, proprio- to falls are potentially preventable, based on a retrospective ception and vestibular system. Gait and balance problems analysis of causes and circumstances of serious falls. Identi- have many aetiologies, and many therapeutic approaches fying and eliminating environmental risks in homes or can be effective. Readily identifiable gait problems adversely institutions could prevent many falls due primarily to envi- affect function in 20–40% of people aged >65 (and 40–50% ronmental causes, and adequate medical evaluation and of those aged >85), and about half of these problems are treatment for underlying medical risk factors, such as unsta- severe. In a large longitudinal study of persons aged ≥75, ble gait and disabling medical conditions, could prevent 10% needed assistance to walk across the room, 20% were unable to climb a flight of stairs without help and 40% wereunable to walk half a mile. Gait problems can stem from Causes and risk factors for falls
simple age-related changes in gait and balance as well asfrom specific dysfunctions of the nervous, muscular, skele- There are many distinct causes for falls in old people, as tal, circulatory and respiratory systems or from simple listed in Table 1, which summarises data from 12 of the deconditioning following a period of inactivity.
largest retrospective studies of falls among older persons The next major reported cause of falls is dizziness, living in a variety of settings. ‘Accidental’ or environment- which is an extremely common symptom among older per- related is the most frequently cited, accounting for 30–50% sons. However, it is a non-specific symptom and may reflect in most series. However, many falls attributed to accidents problems as diverse as cardiovascular disorders, hyperventi- really stem from the interaction between identifiable envi- lation, orthostasis, drug side-effect, anxiety or depression.
ronmental hazards and increased individual susceptibility to The related problem of orthostatic hypotension, defined as hazards from accumulated effects of age and disease. Older a drop of over 20 mmHg of systolic blood pressure between people have stiffer, less co-ordinated and more dangerous lying and standing, has a 10–30% prevalence among ‘nor- gaits than do younger people. Posture control, body-orienting mal’ elderly people living at home. It can stem from several reflexes, muscle strength and tone, and height of stepping factors, including autonomic dysfunction (frequently relatedto age, diabetes or brain damage), hypovolaemia, lowcardiac output, Parkinsonism, metabolic and endocrine dis- Table 1. Causes of falls in elderly adults: summary of 12 orders, and medications (particularly sedatives, antihyper- studiesa that carefully evaluated elderly persons after a fall tensives and antidepressants). The orthostatic drop may be more pronounced in the morning, because the baroreceptor response is diminished after prolonged recumbency. How- ever, it is a less common cause of falls than its prevalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
would indicate, probably reflecting the fact that most per- sons with the syndrome become accustomed to it and are able to find a seat or adjust before falling.
Drop attacks are defined as sudden falls without loss of consciousness or dizziness and have in the past been impli- cated in between 1 and 10% of falls. Patients typically experience abrupt leg weakness, sometimes precipitated by sudden head movement. The weakness is usually transient but can persist for hours. This syndrome has been attrib-uted to transient vertebrobasilar insufficiency, although it probably stems from diverse mechanisms, including leg Mean percentage calculated from the 3,628 falls in the 12 studies.
weakness and knee instability. Drop attacks are today Ranges indicate the percentage reported in each of the 12 studies.
dThis category includes arthritis, acute illness, drugs, alcohol, pain, epilepsy and reported much less often—probably reflecting better diag- nostic precision. In the past, the drop attack category was Falls in older people—an overview
often used as a ‘waste basket’ category for otherwise unex- psychoactive medications, have also been identified in a plained falls. In reality, true drop attacks are quite uncommon.
number of studies as risk factors for falls, although their rel- Syncope, or sudden loss of consciousness, usually results ative risk has generally been in the 1.5–1.7 range, just below from decreased cerebral blood flow or metabolic factors. It that of the other factors in the list.
has been the attributable cause of between 2 and 10% of Most of the factors on the list are amenable to improve- falls in several series but has been excluded from many ment, implying ways that many falls can potentially be pre- other series either by definition (because syncope is not a vented; moreover, the effectiveness of these preventive typical type of fall) or because many elderly patients with strategies has been documented in a number of studies.
syncope are acutely hospitalised and are treated differently.
(Among the most widespread of the risk factor reduction Other specific causes of falls include disorders of the strategies involve regular exercises to improve strength, gait central nervous system, cognitive deficits, poor vision, and balance, and results have been promising.) drug side-effects, alcohol intake, anaemia, hypothyroidism,unstable joints, foot problems, severe osteoporosis with Evaluating the fall patient
spontaneous fracture and acute illness. Because most elderlyindividuals have multiple identifiable risk factors predisposing When assessing a patient who has fallen, obtaining a full to falls, the exact cause can often be difficult to determine.
report of the circumstances and symptoms surrounding the Because a single specific cause for falling often cannot fall is crucial [2]. Reports from witnesses are important, as be identified, and because falls are usually multifactorial in the patient may have poor recollection of the event. Fall origin, many investigators have performed both prospective circumstances that can point to a specific aetiology or narrow- and retrospective epidemiological studies to identify specific down the differential diagnosis include sudden rise from a risk factors that place individuals at increased likelihood of lying or sitting position (orthostatic hypotension), trip or falling. In many ways, identifying risk factors for falls is slip (gait, balance or vision disturbance or an environmental much more useful than trying to classify specific precipitat- hazard), drop attack (vertebrobasilar insufficiency), looking ing causes retrospectively. Not only are prospective data up or sideways (arterial or carotid sinus compression) and likely to be more accurate than data derived from chart loss of consciousness (syncope or seizure). Symptoms review after the event, but by identifying risk factors early, experienced near the time of falling may also point to a the most effective preventive strategies can be devised and potential cause—dizziness or giddiness (orthostatic hypo- instituted. Table 2 lists the major fall risk factors, and their tension, vestibular problem, hypoglycaemia, arrhythmia and relative importance, pooled from a large number of such drug-side effect), palpitations (arrhythmia), incontinence or studies [2–6]. The most important of these risk factors are tongue biting (seizure), asymmetric weakness (cerebrovas- muscle weakness and problems with gait and balance.
cular disease) or chest pain (myocardial infarction or coro- Muscle weakness is an extremely common finding nary insufficiency). Medications and concomitant medical among the aged population when looked for, mostly stem- problems may be important contributors.
ming from disease and inactivity rather than ageing per se.
On the post-fall physical examination, it is especially Case–control studies demonstrate substantially increased pertinent to look for particular findings that may have risk of falls and fractures among individuals with gait and directly contributed to the fall, as well to note other fall risk muscle dysfunctions. A simple screening test of gait and bal- factors. Important to look for are orthostatic changes in ance function, such as the ‘timed up and go’ or Tinetti’s gait pulse and blood pressure, presence of arrhythmias, carotid and balance test [7], is often useful in identifying risk and bruits, nystagmus, focal neurological signs, weakness and documenting need for treatment. Medications, specifically other musculoskeletal abnormalities, visual loss, gait distur-bances and cognitive dysfunction. It is often useful toattempt (under carefully monitored conditions) to repro- Table 2. Important individual risk factors for falls: sum- duce the circumstances that might have precipitated the fall, e.g. positional changes, head turning or carotid pressure.
Gait and stability should be assessed by close observation of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
how the patient rises from a chair, stands with eyes open and closed, walks, turns and sits down. One should take particular note of gait velocity and rhythm, stride length, double support time (the time spent with both feet on the floor), height of stepping, use of assistive devices and degree of sway. Use of a formal gait assessment screening protocol, such as the Tinetti balance and gait instrument [ 7 ], can be very helpful. More detailed gait evaluation canbe useful among persons who fail the screen.
Laboratory tests are seldom very useful, although a full bNumber of studies with significant association/total number of studies look- blood count, serum electrolytes and ECG often disclose contributory abnormalities. More costly tests (e.g. Holter Relative risks (prospective studies) and odds ratios (retrospective studies).
dNumber in parenthesis indicated the number of studies that reported relative monitoring and gait laboratory evaluation) should be reserved for persons with suggestive signs or symptoms.
L. Z. Rubenstein
Therapeutic and preventive approaches
ment, risk factor reduction, exercise, environmental modifica-tion and education) have been tested, and recent meta- Once the cause(s) and/or risk factor(s) of falling are deter- analyses have documented the effectiveness of several mined, appropriate specific therapy can be instituted [2].
approaches [9]. Effective approaches include multidimen- The following are among the more obvious examples: (i) sional risk factor assessment tied to targeted interventions, cardiac dysrhythmias clearly related to a fall should be exercise programmes (which include balance, strength and treated with antiarrhythmics or a pacemaker, or both; (ii) endurance training), and environmental assessment and mod- hypovolaemia due to haemorrhage or dehydration calls for ification. Programmes combining all of these approaches treatment directed towards restoring haemodynamic stabil- seem to have had the strongest effects. Recent clinical prac- ity; (iii) Parkinsonism usually responds to specific therapy, at tice guidelines from the AGS/BGS/AAOS panel and other least transiently; however, in advanced cases, safe ambula- organisations have strongly advocated preventive approaches tion can require extensive assistance. Discontinuing medica- using these three components [2]. Post-fall assessments, as tion that causes postural hypotension or undue sedation is outlined above, have been shown to reveal many otherwise important whenever possible. For patients with gait and bal- undetected treatable conditions and risk factors, as well as to ance disturbances, specific assistive devices (e.g. walkers, significantly prevent falls and reduce hospitalisations [2, 10, canes and shoe modifications) are often helpful. Also help- 11]. Exercise programmes can clearly improve strength, ful can be a programme of gait training under supervision of endurance and body mechanics, and several controlled trials a physical therapist, individualised to deal with the specific have shown significant reduction in falls [2, 12–15].
underlying cause(s) (e.g. weakness, imbalance and arthritis).
Several European trials of hip protector pads mainly in Several techniques may benefit patients with persistent nursing home settings have reported dramatic reductions in orthostatic hypotension due to autonomic dysfunction.
hip fractures [16]. Compliance has been an issue but appears These include sleeping in a bed with the head raised to min- to be surmountable, especially with more comfortable imise sudden drop in blood pressure on rising, wearing elas- newer models. Some preliminary new data indicate possible tic stockings to minimise venous pooling in the legs, rising mild benefits from vitamin D on balance and fall reduction.
slowly or sitting on the side of the bed for several minutes In summary, newest study data confirm the clear effec- before standing and avoiding heavy meals and vigorous tiveness of a number of interventions in preventing falls, activity in hot weather. If conservative mechanical measures including fall risk assessments tied to interventions, exercise, are ineffective, blood volume can be increased by liberalis- environmental inspection and modification, and combined ing dietary salt. If this is ineffective, mineralocorticoid ther- interventions. The future looks bright in this area, as in so apy (fludrocortisone increasing gradually from 0.1 mg/day) many areas of geriatrics. Systematic attention to fall preven- or an α-1 agonist, such as midodrine (beginning at 2.5 mg tid), tion is a vital part of comprehensive care of the older adult.
can help to maintain blood pressure, as long as associatedmedical conditions do not preclude these agents, and duecare is taken to avoid side-effects such as supine hyperten-sion and fluid accumulation.
Key points
More difficult is managing and preventing recurrent falls • Falls occur in 30–60% of older adults each year, and among patients for whom a specific cause cannot be identi- 10–20% of these result in injury, hospitalisation and/or fied or who have multiple or irreversible causes. A careful search for, and correction of, other risk factors that pre- • Most falls are associated with identifiable risk factors dispose to falling (such as visual and hearing deficits) is (e.g. weakness, unsteady gait, confusion and psychoac- essential. For disabilities that do not properly resolve with treatment of the underlying medical disorder (e.g. hemipare- • Research shows that detection and amelioration of risk sis, ataxia, persistent weakness or joint deformities), a trial factors can significantly reduce the rate of future falls.
of short-term rehabilitation may improve safety and dimin- • Other evidence-based fall reduction methods include ish long-term disability. When irreversible problems exist, systematic exercise programmes and environmental residual limitations should be explained and coping meth- inspection and improvement programmes.
• Recent international groups have developed useful clini- Physicians should caution patients to eliminate home cal guidelines for reducing the risk of falls.
hazards such as loose or frayed rugs, trailing electrical cordsand unstable furniture. Patients and their families should beadvised of the importance of specific environmentalimprovements—adequate lighting, bathroom grab rails and References
raised toilet seat, secure stairway banisters, raising or lowering 1. Nevitt MC, Cummings SR. Type of fall and risk of hip and
bed and an easily accessible alarm system are possibilities. A wrist fractures: the study of osteoporotic fractures. The Study visiting nurse or any experienced person can perform a of Osteoporotic Fractures Research Group. J Am Geriatr Soc home evaluation to suggest modifications. Checklists to aid in this process are available [2,8].
2. AGS/BGS/AAOS Panel on Falls Prevention. Guideline for
Fall-prevention has been an area of active research over the prevention of falls in older persons. J Am Geriatr Soc the past 10–15 years. A number of programmes (e.g. assess- Falls in older people—an overview
3. Robbins AS, Rubenstein LZ, Josephson KR et al. Predictors of
11. Rubenstein LZ, Robbins AS, Josephson KR et al. The value of
falls among elderly people: results of two population-based assessing falls in an elderly population: a randomised clinical studies. Arch Intern Med 1989; 149: 1628–33.
trial. Ann Intern Med 1990; 113: 308–16 4. Rubenstein LZ, Josephson KR. The epidemiology of falls and
12. Tinetti ME, Baker DJ, McAvay G et al. A multifactorial
syncope. Clin Geriatr Med 2002; 18: 141–58.
intervention to reduce the risk of falling among elderly 5. Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nurs-
people living in the community. N Engl J Med 1994; 331: ing home. Ann Intern Med 1994; 121: 442–51.
6. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in
13. RubensteinLZ, Josephson KR, Trueblood PRet al. Effects of a
older people: a systematic review and meta-analysis. J Am group exercise program on strength, mobility, and falls among fall-prone elderly men.J Gerontol A Biol Sci Med Sci 2000; 7. Tinetti ME, Williams TF, Mayewski R. Fall risk index for eld-
erly patients based on number of chronic disabilities. Am J 14. Province MA, Hadley EC, Hornbrook MC et al. The
effects of exercise on falls in elderly patients: a preplanned 8. United States Consumer Product Safety Commission: Home Safety
meta-analysis of the FICSIT trials. JAMA 1995; 273: Checklist For Older Consumers, Washington, DC USCPSC, 1985.
9. Shekelle P, Maglione M, Chang J et al. Falls Prevention Interven-
15. Jensen J, Nyberg L, Gustafson Y, Lundin-Olsson L. Fall &
tions in the Medicare population. RAND-HCFA Evidence Report injury prevention in residential care—effects in residents with Monograph, HCFA Publication #HCFA-500-98–0281, Balti- higher & lower levels of cognition. J Am Geriatr Soc 2003; 51: more, MD, 2002 [Also Chang et al., BMJ 2004; 328: 680].
10. Close J, Ellis M, Hooper R et al. Prevention of falls in the eld-
16. Kannus P, Parkkari, Neimi S et al. Prevention of hip fracture in
erly trial (PROFET): a randomised controlled trial. Lancet elderly people with use of a hip protector. N Engl J Med 2000;

Source: ftp://ftp.prip.tuwien.ac.at/pub/outgoing/zamba/fallpapers/Rubenstein06.pdf