Transcript Slip, Sliding Away: New Research in Falls Prevention

Current Research on Falls
Prevention
Jane Mahoney, MD
University of Wisconsin Medical School
Dec 15, 2004
Scope of the Problem
• In 1999, accidents were the 8th leading
cause of death for adults age 65 and older in
the US, and the leading cause of accidental
deaths was falls.
• Fractures accounted for 531,000
hospitalizations in the over-65 age group.
Falls in Wisconsin
• In 2002, there were 22,500 hospitalizations
in Wisconsin for fall-related injuries.
• The state’s death rate due to falls has
increased 20% from 1992 to 2002
• The state’s death rate due to falls is almost
twice the national average.
Fall-Injury Rates Are Increasing
Over Time
2000
1600
1200
800
400
80
19
85
19
90
19
95
19
5
19
7
70
0
19
number per 100,000
• Kannus et al, Lancet
1997
• Finnish data – national
hospital discharge
register
• Age-adjusted incidence
of fall-related injury
for ages 60 and over
Women
Men
Purpose
• Overview of current guidelines for fall prevention
• Intervention research: multifactorial trials,
exercise, group cognitive-behavioral classes
• Prevention after hospital discharge
• Preliminary data, Kenosha County Falls
Prevention Study
• Dane County SAFE Study: evaluating research
findings in a community setting
Definition of Accidental Fall
An accidental fall is an event which results in
a person coming to rest inadvertently on the
ground or other lower level not due to
obvious loss of consciousness, stroke,
seizure or sustaining a violent blow.
Components of Postural Control
Sensory
Input
Central
Processing
Visual
Environment
Vestibular
Proprioceptive
Cognition
CNS pathways
Medications
Effector
Output
Musculoskeletal
Strength
Biomechanical
Risk Factors For Falls from
Epidemiologic Studies
• Previous hx of falls
• Balance or gait
impairment
• Dementia
• Visual deficit
• Neuropathy
• Muscle weakness
• Psychotropic
medications
• Depression
• Arthritis, Parkinson’s,
stroke
Risk Factors are Additive
Tinetti, NEJM, 1988
80
70
60
50
40
30
20
10
0
0 rf
1
2
% falling
3
4+
2001 Guidelines
American Geriatrics Society, British Geriatric
Society, American Academy of Orthopedic Surgeons
• All older adults should be asked at least once a
year about falls.
• All older adults who report a single fall should be
observed rising from a chair and walking.
• Older adults with 2 or more falls in the past year, 1
fall with injury, or 1 fall with gait and balance
problems should receive a fall evaluation followed
by multifactorial intervention.
2001 Guidelines
Multifactorial Intervention
• Gait training including advice on assistive devices
• Review/modify medications, especially
psychotropics
• Individualized, progressive exercise programs
with balance training
• Treat postural hypotension
• Modify environmental hazards
• Treat cardiovascular disorders including
arrythmias
Randomized Trials of
Multifactorial Interventions
Study
Outcome
• Tinetti, NEJM 1994
Rate
31%
• Wagner, AJPH, 1994
• Close, Lancet, 1999
• Day, BMJ, 2002
Risk
Risk
Rate
9%
61%
33%
Benefit of Exercise in Reducing
Falls
• Previous studies have shown that patients
with a history of multiple previous falls will
benefit from individualized physical therapy
• Physical therapy should be progressive, last
several months, and should include balance
exercises
Randomized Trials of Group
Exercise
Outcome
Study
Wolf, JAGS, 1996
Tai Chi
Lord, JAGS, 2003
standing
Barnett, Age Ageing, 2003 standing
Day, BMJ, 2002
standing
Wolf, JAGS, 2003
Tai Chi
Risk
47%
Rate
Rate
Rate
Risk
22%
40%
18%
25% NS
Group Exercise for Falls
Prevention
• Include standing exercises that challenge
balance
– Stepping, Tai Chi, change of direction, dance
steps
• Complexity and speed of exercises increase
• Classes held 1-2 times per week, typically
also with home exercises
• Exercises are individualized as needed
Group classes: cognitivebehavioral learning
• 7-week classes plus 1 home OT visit to
improve self-efficacy, encourage behavioral
change, reduce falls
• Focus on improving balance and strength,
improving home and community
environamental and behavioral safety,
encouraging vision screen and med review
• Results = 31% reduction in falls
Post-hospital falls prevention rationale
Environment
Sensory
CNS
Musculoskeletal
Output
Delirium
Systemic Effects of Illness
Acute Changes in Postural Control
New Medications
Environment
Sensory
CNS
changes
Musculoskeletal
Output
Bedrest, Deconditioning
Effects of Bedrest
•
•
•
•
•
•
Loss of muscle mass and strength
Orthostasis, volume contraction
Increased body sway
Slower gait speed
Visual-spatial abnormalities
Impaired coordination
Risk of Falls after Hospitalization
Mahoney, JAGS, 1994
• Older adults discharged from St. Mary’s
Hospital after acute illness - 14% fell in the
month after hospital discharge.
• Risk was higher among those receiving home
nursing compared to those not (20% vs 8%
fell, p=.01)
Risk factors by home nursing use
Not receiving home nursing
Vision impairment
Self-report of confusion
Receiving home nursing
Mobility imp pre-hosp
Decline in mobility by
discharge
Use of anticholinergics or
antihistamines
Self-report of confusion
Falls After Hospital Discharge
Mahoney, Arch Int Med, 2000
- 311 older adults receiving home nursing after discharge
7
1000 person-days
6
5
4
3
2
1
0
2
4
6
8
10
12
Weeks After Hospital Discharge
14
Rehospitalizations Due to Fall
Injuries
• 15% of all re-hospitalizations in the first
month were due to fall injuries.
Risk Factors for Falling:
Pre-Hospital
Pre-Hospital:
• Prior dependence in
ADLs
• Used standard walker
• > 2 falls in yr prior
• # hospitalizations
in year prior
Odds Ratio
2.3
3.2
1.7
1.1
Risk Factors Potentially Related to
Hospitalization and Acute Illness
Post-Hospital:
Admit for GI dx
First generation tricyclic
Uses cane indoors
Middle tertile balance
Lowest tertile balance
Probable delirium
Odds Ratio
2.5
3.2
0.3
2.2
3.3
6.7
Post-Hospital Falls Prevention :
Nikolaus, Bach: JAGS, 2003
• Home visit during hospitalization followed
by 1+ visits after discharge
• Typically OT and other member of interdisc
team (RN, PT or SW)
• Evaluate and modify home hazards, teach
safe behaviors including use of mobility and
functional aids
Results
• 30% decrease in falls in 1-year follow-up
compared to no home visits
• Most effective in those with 2+ falls in year
prior: IRR = 0.63
• Both groups got comprehensive geriatric
assessment prior to discharge
Post-Hospital Fall Prevention:
Cumming et al, JAGS 1999
• 1+ home OT visits, and 1 phone call 2
weeks post-first visit
• Assess and modify home hazards, teach safe
behaviors, evaluate and recommend safe
footwear
Results
• 19% reduction in fallers (p=.050)
• 36% reduction in fallers among those with
prior hx of falls (p=.001)
Approach to post-hospital falls
prevention
• Minimize bedrest during hospitalization
• Observe patient doing functional tasks
– walking, transferring, reaching, dressing
• Educate older patients about post-hospital risk
– Use mobility aid, caution with maneuvers
– Eyeglasses, sturdy footwear, home safety check
• Stratify post-hospital falls risk:
– 2+ falls in year prior
– significant decline in mobility with hosp
For high risk patients
• Reduce psychotropics
• Refer to home health for home OT (if qualifies)
–
–
–
–
Evaluate transfers and ADL
Assess need for home functional aids
Assess and modify home hazards
Teach safe behaviors
• Obtain PT in-hospital
– Evaluate for home assistive device
– Evaluate need for home PT
– Provide balance, strengthening exercises for home
Applying Multifactorial
Interventions in the Community
• Multifactorial falls prevention strategies have been
successful in research studies
– utilized specific exercise programs or physical
therapists
– utilized multiple specialists
• It is unknown if a multifactorial intervention
utilizing existing medical systems will decrease
falls.
Randomized Trial of CommunityBased Multifactorial Intervention
• Kenosha County Falls Prevention Study
– Funded by Wisc Resource Center Prevention
Grant
– Algorithm for falls assessment,
recommendations, and monthly follow-up.
– Recommendations to physician, referral to PT
followed by exercise, other referrals as needed.
Methods
•
•
•
•
Inclusion Criteria:
- Residing in Kenosha County, WI, age >65.
- Two or more falls in past year, or one fall in past 1 to
2 years with injury or gait and balance problems
Exclusion Criteria:
- Residence in Nursing home or CBRF
- Diagnosis of dementia, no related caregiver in home.
Baseline information collected regarding: demographics,
health status, mobility, function, cognition, depression,
medications, vision, and health behaviors.
Followed monthly for falls for 1 year
Enrollment Characteristics
616 Referred
418 Eligible (68%)
349 Enrolled (83% of eligible)
Baseline Characteristics (n=349)
MEASUREMENT
DOMAIN
Age
BASELINE
80.0 ±7.5
Demographics
Female
78.5%
2.4 ± 2.5
Falls
No. falls in past year
Health status
Emergency Room visit(s) past 4 months
30.7%
Mobility
Assistive device use indoors
35.9%
Barthel Index
Function
No. of independent Instrumental Activities of Daily
Living out of 7, (IADLs)
Cognition
Mini-Mental State Exam (max 30)
Meds
No. of prescription medications
Any alcohol intake
Health
Behaviors
Frequency of exercise (days per week) , (%)
88.1 ±16.6
4.8 ± 2.2
27.1 ± 4.4
5.7 ± 3.3
37.3%
<1
34.7%
1-3
21.2%
4-7
44.1%
Differences in 2+ fallers versus
single fallers
Kenosha County Falls Prevention Study
funded by the Wisconsin Department of
Health and Human Services
Differences in recurrent fallers
versus single fallers
•
The AGS recommends that older adults who
have had 2+ falls in the past year, 1 fall with
injury, or 1 fall with gait or balance problems
receive a multifactorial falls evaluation.
•
Purpose: to examine baseline characteristics
of those who have had 2+ falls in the past 12
months, compared to those with 1 fall in past 1-2
years. If there are differences, this could have
implications for treatment.
Enrollment by Falls History
200
180
160
140
120
100
80
60
40
20
0
2+falls past 12 mos,
n=189
1 fall past 12 mos. With
injury, n=64
1 fall past 12 mos. with
gait/balance problems,
n=51
fall past 12-24 mos. With
injury, n=30
fall past 12-24 mos. With
gait/balance problems,
n=15
•Comparison: 2+ falls past 12 mos. (n=189) vs. 1 fall in past 24 mos. (n=160)
•Two-sample t-tests for continuous variables and Pearson’s chi-square tests for
categorical variables.
Comparison of Baseline Characteristics
DOMAIN
MEASUREMENT
2+ FALLS
PAST YEAR
1 FALL PAST 1-2
YEARS
N=160
P-VALUE
N=189
Demographics
Falls
Health status
Age
79.9
80.0
0.94
Female
73.5%
84.4%
0.014
No. falls in past yr
3.7
0.8
<0.0001
Hx of hip fx , %
11.2%
7.6%
0.25
Hx of CVA , %
31.2%
18.8%
0.008
Health rated fair/poor ,
%
38.1%
21.3%
0.007
ER visits in past 4 mos,
%
38.1%
21.9%
0.001
Assistive device use
indoors , %
42.3%
28.1%
0.006
Without
help
60.9%
83.8%
Some
help
26.5%
11.9%
Unable
12.7%
4.4%
Mobility
Walk
outside, %
<0.0001
Comparison of Baseline Characteristics
DOMAIN
MEASUREMENT
2+ FALLS
PAST 12 MOS
P-VALUE
N=189
1 FALL PAST 24 MOS.
WITH INJURY OR
GAIT/BALANCE
PROBLEMS
N=160
No. IADLs
4.3
5.4
<0.0001
Barthel Index score
85.1
91.6
0.0002
Cognition
MMSE score
26.6
27.6
0.028
Depression
GDS scpre
3.4
2.5
0.004
6.2
5.
0.0007
Medication
No. prescription
medications
No. Psychotropics
0.3
0.1
0.018
Able to watch TV, %
91.5%
96.9%
0.037
Any intake alcohol , %
34.4%
40.6%
0.29
Exercise program , %
18%
18.1%
0.97
<1
36%
33.1%
1-3
19.5%
23.1%
4-7
44.4%
43.8$
Function
Vison
Health
Behaviors
Frequency of
exercise, times
per week , %
0.70
Barthel Comparison
SELECTED
BARTHEL
ACTIVITY
lower score
indicating more
impairment
MEAN BARTHEL SCORE
2+ FALLS PAST
YEAR
1 FALL PAST 1-2
YEARS
P-VALUE
Bathing Self
3.6
4.4
0.0002
Dressing
8.8
9.3
0.036
Toileting
9.6
9.9
0.019
Transferrring
14.1
14.7
0.014
Walking on level
surface
11.7
13.3
0.001
Climbing stairs
7.1
8.5
0.0002
Conclusion
• There are multiple significant differences in
domains of: health status, mobility, function,
cognition, depression, medications, and vision,
comparing recurrent fallers and single fallers.
Recurrent fallers are more likely to have risk
factors in multiple domains.
• The propensity for positive exercise behavior was
similar in both groups.
Implications
• Given the greater number of risk factors and
impairments in the recurrent faller group,
we may need to consider focusing a
multifactorial approach toward this group.
• Our data on exercise behavior suggests
recurrent fallers may be equally likely to
adhere to an exercise intervention as single
fallers
Limitations
• The sample was self selected by those
interested in a falls prevention trial and may
not be representative of all fallers.
• This was primarily a white, middle-class
population and may not be generalizable to
other populations.
Dane County SAFE Study
• Three-year RCT funded by CDC
• Will randomize 420 older adults at high risk for
falls to multifactorial intervention and follow-up
or health information booklets.
• Intervention similar to Kenosha County study.
• But, supplemented by educational initiatives to
increase physician and physical therapy utilization
of recommendations.
Grant Overview
 Two components
 Goal 1: In-home multifactorial assessment randomized
trial for high-risk older adults
 Goal 2: Education of primary health care providers in
Dane County.
Goal 1: Multifactorial intervention trial
• Target group:
– Community-residing adults age 65 and older at high
risk for falls – AGS criteria
• 2+ falls in the past year
• 1 fall with injury
• 1 fall with abnormal gait or balance
– Exclusion criteria:
• residence in NH or CBRF
• Unable to give informed consent and no related caregiver in
home.
Randomization
High risk older adults
(n=420)
informed consent
baseline assessment
In-home multifactorial intervention
(n=210)
Educational booklets
(n=210)
Outcomes
• Primary outcome = falls
– Hypothesized 40% reduction in rate of falls over 1 year
compared to control group
– Falls obtained via monthly calendar
• Secondary outcomes
– # hospitalizations and hospital days
– # nursing home admissions and NH days
– Change in function, mood, vision, medications, fear of
falling, and physical performance at 12 months
compared to baseline.
Multifactorial assessment
• Follows principles of AGS guidelines
• Can be performed by PT or RN with crosstraining
• Requires about 2 hours to perform
• Is performed in-home preferably with
caregiver present
Algorithm assesses:
– History of falls, comorbidities, risk related to IADLs
and ADLs, fear of falling, risky behaviors, footwear
– depression, cognition
– medications, alcohol intake
– Exam: Orthostatics, vision, visual fields, vibration,
Romberg
– Gait and balance: Sensory integration, reactive
balance, Berg balance, Tinetti Gait, Attention,
Foot/ankle alignment
Outcome of assessment
• Algorithm generates recommendations to
patient and physician, and referrals to PT,
opthalmology, podiatry, OT, and other
health provider and community resources
• Assessor returns to the home within 2
weeks to provide recommendations and
referrals
Intervention continues for 1 year
• Monthly phone call from assessor to encourage
and assess compliance, help with problem-solving,
etc.
• For most participants, the algorithm generates a
referral to physical therapy. Physical therapy is
followed by an ongoing, individualized exercise
plan for community or home exercise, with an
exercise “buddy” if needed.
Goal 2: Provider Education
• Target Groups
– Primary care physicians
– Physician Assistants and NPs
– Physical Therapists, Paramedics
• Purpose: Educate for falls prevention
• Outcomes: Compare change in rate of
hospitalizations for fall-related injuries in
Dane County to other counties
Strategies for Recruitment
• Direct to seniors
• Community groups
• Professional providers
• Enrolled to date: 337
• Enrollment will continue through April 05
Referrals by Referral Type (N=729)
7%
2%
1%
13%
3%
7%
67%
DCAAA
Post Hosp/ER
Home Care
Primary MD/clinic
Therapy OT/PT
Other Indirect
Self Referred
Participants by Referral Type
(N=290)
6%
1%
1%
14%
2%
10%
66%
DCAAA
Post-Hosp/ER
home care
Primary MD/clinics
Therapy (PT/OT)
Other Indirect
Self Referred
Thank-you





Wisconsin Dept. of Health and Family Services
Terry Shea, PT, Co-Principal Investigator
Bob Przybelski, MD, Co-Investigator
Ron Gangnon, Mari Palta, Biostatistics
Nurses and physical therapists with the Dane
County SAFE Study
 Sheila Guilfoyle, Coordinator
 Community agencies, health care providers