Preventing Falls in Older Adults: State of the Science
Preventing Falls in Older Adults: State of the Science
Laurence Rubenstein, MD, MPHGreater Los Angeles VA GRECC
Professor of Medicine, UCLA
British Columbia Injury Prevention ConferenceNovember 19-20, 2008
Preventing Falls: What does the evidence show?
Preventing Falls: What does the evidence show?
Background: Epidemiology, costsCauses & risk factorsPrevention approaches--evidence
Studies & meta-analysesAGS/BGS practice guidelines--update
Famous Fallers
Case ReportCase Report
78 year old gentlemanGood general healthGave a 1-hour graduation speech on October 20, 2004…
AP/Wide World Photos
What happened?What happened?
Broke left knee & right armNever fully recoveredCould his fall have been prevented?
Fall Incidence in Older Adults[rate/person/yr] or [rate/bed/yr]
Fall Incidence in Older Adults[rate/person/yr] or [rate/bed/yr]
Home Hospital Nsg Home
Any fall .3 1.5 1.7
Severe fall .03(10%) .3 .35(20%)
Fracture .01 .05 .07
Hip fx .003 .02
Rubenstein LZ, Josephson KR. Clin Geriatr Med. 2002(May);18(2):141-158
Falls MortalityFalls Mortality
Accidents: the 5th leading cause of death in older adultsDeaths from falls: 2/3 of accidental deaths72% of U.S. fall-related deaths occur in the 13% of population age 65+
Rubenstein LZ, Josephson KR. Clin Geriatr Med. 2002(May);18(2):141-158
Costs of FallsCosts of Falls
8% of pop ≥70 visit ERs for falls yearly 1/3 of these are hospitalized5.3% of hosp patients ≥65 are due to fallsU.S. cost est. 2000→$20 B. (2020→$32 B)42% of fallers reduce activity after fall18% restricted activity initiated by fallsPrecipitate NH entry
Rubenstein LZ, Josephson KR. Clin Geriatr Med. 2002(May);18(2):141-158
Falls are #1 cause of NH litigation
Causes of Falls: Summary of 12 StudiesCauses of Falls: Summary of 12 Studies
Accident/environment 31%Gait/balance disorder 17Dizziness/vertigo 13Drop attack 10Confusion 4Postural hypotension 3Vision problem 3Other specified 15Unknown 5
Rubenstein LZ, Josephson KR. Clin Geriatr Med. 2002(May);18(2):141-158
Risk Factors for Falls: 16 Multivariate StudiesRisk Factors for Falls: 16 Multivariate Studies
Factor Signif/All Mean RR Range Weakness 10/11 4.4 1.5 - 10.3 Prior fall 12/13 3.0 1.7 - 7.0 Balance deficit 8/11 2.9 1.6 - 5.4 Gait deficit 10/12 2.9 1.3 - 5.6 Assistive device 8/8 2.6 1.2 – 4.6 Vision deficit 6/12 2.5 1.6 – 3.5 Arthritis 3/7 2.4 1.9 –2.9 ADL deficit 8/9 2.3 1.5 – 3.1 Depression 3/6 2.2 1.7 – 2.3 Cognitive deficit 4/11 1.8 1.0 – 2.3 Age >80 5/8 1.7 1.1 – 2.5
Rubenstein LZ, Josephson KR. Clin Geriatr Med. 2002(May);18(2):141-158
Drugs & Falls: Meta-analysisLeipzig, Cumming, Tinetti, JAGS, 1999
Drugs & Falls: Meta-analysisLeipzig, Cumming, Tinetti, JAGS, 1999
Psychotropics, any: 1.73 (1.52-1.97)Neuroleptics: 1.50 (1.25-1.79)
Sedative/hypnotics: 1.54 (1.40-1.70)
Antidepressants: 1.66 (1.40-1.95)
Benzodiazepines: 1.48 (1.23-1.77)
Diuretics: 1.08 (1.02-1.16)Anti-arrhythmics (Ia) : 1.59 (1.02-2.48)Digoxin: 1.22 (1.05-1.42)
Leipzig RM, Cumming RG, Tinetti ME. J Am Geriatr Soc. 1999(Jan);47(1):40-50
Atypical Anti-psychotics & FallsHien, Cumming, Cameron, et al, JAGS 53:1290, 2005
Atypical Anti-psychotics & FallsHien, Cumming, Cameron, et al, JAGS 53:1290, 2005
Prospective LTC cohort study, AustraliaN=2005, age 65-104 (mean 86)
1.19 (.15)1.90Sedative/anxiolyt1.45 (.01)1.96Antidepressants1.32 (.54)1.0/3.1Risperidone1.74 (.04)2.5/3.3Olanzapine1.35 (.19)1.6/2.6Typical anti-Ψ1.01.0No psychotropicAdj RR (p)RR (alone/+)
Fall Incidence ↑ as Risk Factors ↑(Tinetti, 1988)
00.10.20.30.40.50.60.70.8
0 1 2 3 4+
# risk factors
Risk Factors for Hip FractureRisk Factors for Hip Fracture
05
1015202530
Hip fx/1000 woman yrs
Low3rd
Mid3rd
High3rd
Bone Density
0-23 to 45+
Cummings et al, NEJM 332:767, 1995.
Fall RFs
Environmental Fall Risk FactorsEnvironmental Fall Risk FactorsHome
•low lighting •poor stairs & rails•unstable furniture•rug/carpet hazards•low beds & toilets•no grab bars•slick floors•obstacles•pets•medications
Institution•low lighting •new admission•poor furniture•slick hard floors•low supervision•↓ # of nurses•meal times•no hand rails
Outdoors•bad weather•poor sidewalks•traffic activity•street crossings•uneven steps•distractions•obstacles•↑ activity levels
ri
Intrinsic Risk Factors
•Gait & balance impairment•Peripheral neuropathy•Vestibular dysfunction
•Muscle weakness•Vision impairment•Medical illness•Advanced age•Impaired ADL•Orthostasis•Dementia•Drugs
Extrinsic Risk Factors
•Environmental hazards•Poor footwear•Restraints
Precipitating Causes
•Trips & slips•Drop attack•Syncope•Dizziness
FALL
Fall Injury Risk FactorsFall Injury Risk FactorsRisk Factors Signif/AllGait/balance deficit 3/6Cognitive deficit 3/6Female sex 3/6Vision deficit 2/6Medications 2/6Weakness 2/5ADL deficit 2/6Low body mass 2/6Higher activity 1/6
Rubenstein LZ, Josephson KR. Clin Geriatr Med. 2002(May);18(2):141-158
Fall Risk Assessment MeasuresScott V, et al Age Ageing 2007; 36:130-9
Fall Risk Assessment MeasuresScott V, et al Age Ageing 2007; 36:130-9
Review of 38 validated fall risk measures 23 community tools, 10 NH tools, 8 hospital tools 27 functional mobility tools, 11 multi-factorial tools
Common items includedmental status, fall hx, mobility, other dx, incontinence, drugs, sensory deficits, balance, ADLs, assistive device, weakness, age, gender, acuity, restraint use, functional reach,
Best measures overallHospital: Oliver ‘97, Schmid ‘90, Morse ‘89, Hendrick ‘95, Conley ‘99Outpatient: Shumway ‘00, Tinetti ‘86, Berg ’89, Alessi ’03, Murphy ‘03NH: “universal precautions” (or Morse ‘89, Downton index ‘03)
Fall Risk Assessment Measures: The Reality
Fall Risk Assessment Measures: The Reality
Most can accurately identify patients at higher risk of fallsProbably helpful to sensitize community living elders of their fall risk & what to doImportant for medico-legal purposes in hospitals & NHs: You need to show you’re doing something that is organized and current.
But …virtually all patients in hospital and NHs come out as “high risk.”
Fall Prevention: Growth of RCTsFall Prevention: Growth of RCTs
0
20
40
60
80
100
120
140
'90-91
'92-93
'94-95
'96-97
'98-99
'00-02
'03-04
'05-06
'07-08
BiannualCumulative
Fall Prevention TrialsFall Prevention Trials
Assessment (preventive & post-fall)Exercise & rehabilitation programsEnvironmental modificationsDevicesNursing interventionsCombined interventions
Benefits of a Post-Fall AssessmentResults of a Randomized Controlled Trial in NH
Benefits of a Post-Fall AssessmentResults of a Randomized Controlled Trial in NH
Intervention: 1-2 hr post-fall assessment protocol by GNP (H&P, gait/bal, envir, lab); Feedback to PCP (dx, risk factors, recs)Setting/sample: 700-bed LTC facility, 2/3 F,age x=88, 160 fallers randomized, 2 yr f/u.Results: 3-4 treatable fall risks found per person
↓9% falls, ↓17% mort in treatment group (n.s. trends)↓52% hosp days (p<.01)
Rubenstein et al, Ann Intern Med, 113: 308, 1990
Prevention of Falls in the Elderly Trial (PROFET)
Prevention of Falls in the Elderly Trial (PROFET)
Randomized trial of post-fall assessment of fallers seen in ED & assessed by 7 days.
N=397, ≥65 (mean age 78); LondonAssessment revealed many causes and risk factors and generated many referrals. 12-month follow-up: Intervention group had reduced risk of falls (OR=.39) & hospital admissions (OR=.61). Controls had greater decline in function.
Close J, Ellis M, Hooper R, et al. Lancet. 1999(Jan 9);353(9147):93-97
Clinical Approach to the Faller Clinical Approach to the Faller
Assess & treat any injuryDetermine likely precipitating cause(s)
history, physical , lab (limited)Prevent recurrence
treat underlying cause/illnessidentify & reduce risk factors (e.g., weakness, gait/bal prob, visual prob, polypharmacy)reduce environmental hazardsteach adaptive behavior (e.g., slow rise, cane)
“Falls History”“Falls History”Circumstances & prodrome of fall
sudden LOC, sudden leg weakness, tripped/slipped/hazard, position change, head back, tight collar, cough/urination, palpitations/angina, dizziness/giddiness
Major medical problemsesp. cardiovascular, neurologic
Drugs esp. psychoactive, cardiac, diuretic
Physical Exam: Key AspectsPhysical Exam: Key AspectsVital signs: postural pulse/BP, tempHEENT: vision, hearing, nystagmus Neck: ROM, motion-induced vertigo, bruitCard/Pulm: CHF, arrhythmia, murmurExtrems: arthritis, ROM, deformities, feetNeuro: altered MS, gait/balance deficit, weakness, focal findings, tremor, rigidity, peripheral neuropathy, divided attention
Lab/Diagnostic TestsLab/Diagnostic Tests
CBCBlood glucose, Na+, K+, Ca++, BUNX-ray of injuriesECGHolter monitor &/or CSM (if syncope, arrhythmia, or cardiac cause suspected)Formal gait & balance testing
Tinetti Balance & Gait Scale: Tinetti Balance & Gait Scale:
Sitting balance (1)Rising from chair (4)Standing balance (4)Nudge (2)Eyes closed (1)Turning 360º (2)Sitting down (2)
Initiation (1)Step length (2)Step height (2)Continuity (1)Symmetry (1)Stance/sway (3)Path deviation (2)
Balance (16 pts) Gait (12 pts)
Trueblood PR, Rubenstein LZ. Compr Ther. 1991(Aug);17(8):20-29
Pathologic Gait: CausesPathologic Gait: Causes
Decreased ROM (eg, arthritis, contractures)
Weakness (eg, deconditioning, neuropathy)
Sensory/balance deficit (eg, stroke, neuropathy)
Spasticiy (eg, stroke, cord lesion)
Pain (eg, arthritis, injury)
Impaired central processing (eg, dementia, delirium, stroke, drugs)
The “Aging Home ”Often An Obstacle CourseThe “Aging Home ”
Often An Obstacle Course
Old & rickety furniture & appliancesUnsafe stairs with poor handrailsThrow rugs, frayed carpetsElectrical cords, objects on floorPoor lightingSub-optimal height of bed, toilet, chairsAccumulated clutter of a lifetime
14 dangerous things in this picture
• Stairs without handrail• Deactivated fire alarm• Cloth on space heater• Overloaded outlets• Loose extension cords in pathways• Smoking. Cigarettes unattended• No automatic shut-off on coffee pot• Open bottles of medicine• Old medications in cabinet• Loose rugs• Flip-flop slippers• Clutter on staircase• Newspapers too close to lamp• No handle & no deadbolt on door
From Public Health Canada
The FICSIT Trials:Frailty & Injuries: Coop Studies of Intervention Techniques
The FICSIT Trials:Frailty & Injuries: Coop Studies of Intervention Techniques
7 independent randomized controlled trialsN: Total = 2328, Mean = 333, Range = 100-1323Sites: Atlanta, Boston, Farmington, New Haven, Portland, San Antonio, Seattle
Variety of interventions to reduce falls & frailty, all included exerciseExercise lasted 10-36 weeks, ≥2 year follow-upPooled effects on falls: .90 (95% CI, .81-.99) ▪ Effect for balance exercises: .83 (.70-.98)
Province, MA, Hadley EC, Hornbrook MC, et al. JAMA. 1995; 273(17):1341-1347
Tai Chi and Fall Reduction in Older AdultsLi F et al, J Gerontol Med Sci, 2005
Tai Chi and Fall Reduction in Older AdultsLi F et al, J Gerontol Med Sci, 2005
6-month RCT of 3x/wk Tai-chi vs. stretching in OregonN=256 inactive, home-living elders (age 72-92)6 month results:
Tai-chi StretchingFalls 38 73 p<.01 Fallers 28% 46% p=.01Inj. falls 7% 18% p=.03
Tai-chi group also signif better in: balance, physical performance & fear of falling
Finnish Hip Protector TrialKannus P, N Engl J Med 343: 1506-13, 2000
Finnish Hip Protector TrialKannus P, N Engl J Med 343: 1506-13, 2000
Randomized trial in 22 Finnish geriatric care programs
n=653 subjects, 1148 controls; age 70+; 2/3 NH, 1/3 home care; ≥1 hip fx risk factor
2-yr f/u↓Hip fx at f/u: 21/1000 vs. 46/1000 pers-yr
(RR=.4, 0.2-0.8)↓84% hip fx among fallers wearing protectors Trend toward lower pelvic fx, no effect other fxNNT to prevent one hip fx / yr = 41 (25-115)
Safehip
HipGuard
KPH
CuraMedica
Hip Protectors – Examples
HIPS
Do Hip Protectors Work?Do Hip Protectors Work?
Initial studies, cluster randomized by facility, showed high effectiveness– 50-70% intent to treat– 80-95% among those wearing them
More recent studies, randomized by person, equivocal– Hard to get compliance– Likely contribution from overall program– Patient selection & education crucial
Minns, R. J. et al. Age Ageing 2007 36:140-144; doi:10.1093/ageing/afl186
Hip Protectors are Often Above the Greater TrochanterTop (hard shell): Remploys Caresse, FallGuard, KPH, SafeHip;
Bottom (pads): HipShield, HipSaver, PoseyHipsters, Lyds, Sanavida
Oliver, D. et al. BMJ 2007;334:82
Meta-analysis for hip protectors as a single intervention in care homes--hip fractures
Nursing InterventionsNursing InterventionsRisk assessments (Morse, Hendrich, MDS)Treat identified risksUniversal fall precautions:
call light & assist devices close bed wheels & w/c brakes lockedadequate lightingclean spills immediatelypatient orientation & staff educ
For high-risk patients:move closer to nursing stationincreased observation / sitterprompted toileting
low bedsnon-skid slippers rails & grab barsclutter-free roomsclear signage
bed-chair alarmsspecial careplanship protectors
Care-Steps Fashion Treads
Pillow Paws Walk Alerts
Anti-Slip Footwear – Examples
AirPro AlarmBed & Chair Alarm Chair Sentry
Economy Pad Alarm
Floor Mat Monitor Keep Safe
Bed & Chair Monitors – Examples
QualCare AlarmSafe-T Mate
Alarmed Seatbelt
Locator Alarm
NOCwatch system Patient Down fall detector
Tip Over alarm
Tunstall fall detector
iLife fall detector
Fall Alarms – Accelerometers & Position Sensors
Ambularm
Do Bedrails Prevent Falls? Systematic review. Healy et al, Age Ageing 2008; 37:368-78.
Do Bedrails Prevent Falls? Systematic review. Healy et al, Age Ageing 2008; 37:368-78.
Review of 24 pre-post studies.Most showed some falls after d/c railsSome had small # injuries assoc w/ railsRecs:
Rail use should be individualized Most injuries due to obsolete designs or use in pts likely to climb over themPremature to stop them globally
NOA Floor MatCARE Pad
bedside fall cushion Posey Floor Cushion
Soft Fall bedside matTri-fold bedside mat
Roll-on bedside mat
Bedside Mats – Fall Cushions
Fall Prevention Trials:RAND-CMS Meta-analysisFall Prevention Trials:RAND-CMS Meta-analysis
• Lit review (1980-2002): 830 pubs, 41 RCTsFall risk Monthly fall rate
All RCTs: .88 [.82 - .95] .79 [.71 - .87]
Meta-regression of intervention components:
• Fall eval + f/u .82 [.72 - .94] .63 [.48 - .83] • Exercise .86 [.75 - .99] .84 [.71 - .98]• Environ mod .90 [n.s.] .85 [n.s.]• Education [n.s.] [n.s.]
Exercise Components
Exercise Type
Subjects who fell at least once Mean number of falls
Number of Studies (Arms)
Adjusted Risk Ratio(95% CI)
Number of Studies (Arms)
Adjusted Incident Rate Ratio
(95% CI)
Balance 7 (8) 0.94(074, 1.19)
13 (14) 0.73(0.61, 0.86)
Endurance 7 (7) 0.80(0.66, 0.98)
4 (4) 1.19(0.77, 1.84)
Flexibility 4 (4) 0.72(0.41, 1.25)
5 (5) 0.90(0.60, 1.34)
Strength 8 (9) 0.80(0.54, 1.20)
13 (13) 0.91(0.67, 1.23)
Since the 2003 Meta-analysis, what’s new?
Since the 2003 Meta-analysis, what’s new?
> 35 new published RCTsNew studies of existing models:
Risk assessment + intervention (8), Exercise (14), Multifactorial (8), Hip protectors (3)
New interventionsVisual mods, Vit D + Ca++, Footwear, Vibration
Multifactorial interventions seem bestRF assessment + abatement, exercise, envir modOrganized, consistent, population-based programs
Vitamin D Effect on Falls: Meta-analysisBischoff-Ferrari JAMA 291:1999-06, 2004.
Vitamin D Effect on Falls: Meta-analysisBischoff-Ferrari JAMA 291:1999-06, 2004.
Pooled 5 RCTs, N=1237Vit D reduced OR for falls by 22% (Corrected OR 0.78; 95% CI 0.64-0.96)Effect independent of Ca+ supplement, duration of Rx, sexBaseline Vit D levels not measured
Can Cataract Surgery Reduce Falls? Harwood et al, Br J Ophthalmol 2005:89:53-9
Can Cataract Surgery Reduce Falls? Harwood et al, Br J Ophthalmol 2005:89:53-9
RCT of women age 70+ w/ cataracts randomized to surgery or 12-mo wait listFalls measured by diary + q3mo f/u12 mo results:
34% lower fall rate in surg group (p=.03)3% vs 8% had fractures (p=.03)Surg assoc w/ better activity, anxiety, depression, confidence & visual disability
Tinetti M et al. Effect of evidence dissemination in reducing injuries from falls. N Engl J Med 2008;359:252-261
Connecticut Collaboration for Fall Prevention:Systematic Outreach Education to Physicians, Senior
Centers, Homecare Agencies, & Outpt Rehab Programs
Tinetti M et al. N Engl J Med 2008;359:252-261
Connecticut Collaboration: Adjusted Annual Rates of Serious Fall-Related Injuries and Use of Medical Services per 1000 Persons 70 Years of Age or
Older during the Preintervention, Intervention, and Evaluation Periods
The “Yaktrax” gait stabilizing device – RCT:• ↓58% RR outdoor falls on snow & ice (p<.03)• ↓87% RR injurious falls on snow & ice (p<.02)
• most intervention group falls occurred w/o device
McKiernan FE, JAGS 53:943, 2005
Vibrating Insoles may improve balancePriplata AA, et al. Vibrating insoles & balance in elderly people. Lancet 2003; 362:1123.
1-year double-blind RCT of 70 postmenopausal women showed that brief periods (<20 minutes) of low-level (0.2g, 30 Hz) vibration applied during quiet standing can inhibit bone loss in the spine and femur. Efficacy increased with greater compliance, particularly in subjects with lower body mass (3.45% gain, p<.01).
Rubin C, et al. Prevention of postmenopausal bone loss by low-magnitude, high-frequency mechanical stimuli: J Bone Miner Res 19:343-351, 2004
Vibrating platform improves bone density
Controlled whole body vibration to decrease fall risk in NH residents Bruyere, et al, Arch Phys Med Rehab 86:303,2005
Controlled whole body vibration to decrease fall risk in NH residents Bruyere, et al, Arch Phys Med Rehab 86:303,2005
Oliver, D. et al. BMJ 2007;334:82
Meta-analysis for multifaceted interventions in hospital--falls (random effects model)
Oliver, D. et al. BMJ 2007;334:82
Meta-analysis for multifaceted interventions in care homes for falls (random effects model)
Fall Prevention StrategiesFall Prevention Strategies
COMMUNITY– Ask about falls– Risk-factor screen
& intervention– Post-fall assessment– Exercise program
(strength, balance)– Environmental
inspection & modification
INSTITUTIONOrganized program
– Risk-factor screen– Post-fall assessment– Nurse awareness– Targeted interventions
(e.g., hip pads, sitter, low bed, bed alarms, monitors, prompted toileting)
Evidence Based Guideline for Fall Prevention (AGS-BGS-AAOS Task Force, 2001) SUMMARY
Evidence Based Guideline for Fall Prevention (AGS-BGS-AAOS Task Force, 2001) SUMMARY
Assessment– Inquire about falls, gait, balance at routine visits (at least annually).– Screen persons reporting a problem (e.g., “get up & go” test).– Assess persons failing screen, or w/ >1 fall:
Hx of fall circumstances, meds, chronic illness, mobility levelExamine gait, balance, orthostasis, vision, neuro, cardiovascular
Management of Fallers– Multi-component interventions: assessment & f/u, exercise, gait
training, med review,, treatment (e.g., visual, cardiac, orthostasis)– LTC setting interventions: assessment & f/u, staff education,
gait training & assistive devices, medication review & adjustment – Single interventions: assessment & f/u, exercise (esp balance),
environmental assm’t/mod, medication review & adjustment
Assessment and Management of Falls
AssessmentHistory
MedicationsVision
Gait and balanceLower limb joints
NeurologicalCardiovascular
Multifactorial intervention (as appropriate)
Gait, balance, exercise - programsMedication - modification
Postural hypotension - treatmentEnvironmental hazards - modificationCardiovascular disorders - treatment
Periodic case finding in Primary Care:
Ask all patients about falls in past year
No intervention
No falls
No problems
Gait/balance problems
Patient presents to medical facility
after a fallFall Evaluation*
Check for gait/balance
problem
Single fallRecurrent falls
ConclusionsConclusionsFalls: Common, debilitating, expensivePreventable w/ existing technology
Assessment+f/u, exercise, environment modSystem needed to mobilize evidence-based preventive approachesLikely cost-effective (multiple direct & indirect savings offset program costs)
