Residential Care Summit
Falls Prevention in Residential Care Dr. med. Kilian Rapp, MPH - Geriatric Rehabilitation Clinic, Robert Bosch Hospital Stuttgart - Institute of Epidemiology, Ulm University
Victoria, November 2009
Residential Care Summit
Overview 1. Burden of falls 2. Fall prevention programmes 1. Components for interventions 2. Effectiveness 3. Strategies for implementation
3. New components
Residential Care Summit
1. Burden of falls 1.1 Falls Fall risk >50% per year 2 falls per resident-year 1.2 Consequences Injuries; fear of falling; lower quality of live; higher care need; costs
Serious consequence: (hip) fractures
Residential Care Summit
Hip fracture rates in women in the German population and in residents of nursing homes
Hip fractures / 1,000 person-years
Women 60
50
40
*
30
20
10
* 90+
0 65-69
70-74
75-79
80-84
85-89
90-94
95+
Age German population in 2003 (Icks et al., Osteoporos Int 2008) Residents newly admitted to nursing homes between 2000 and 2005 in Southwest Germany (Rapp et al., JBMR 2008)
Residential Care Summit
Hip fracture rates stratified by the ‘level of care’ at admission
Hip fractures / 1000 py
70
Women Men
60 50 40 30 20 10 0 1
2
Level of care
3 Rapp et al. JBMR 2008
Residential Care Summit
- Falls in residents of nursing homes are frequent and result in a high burden of disease
Effective measures to reduce falls and fallrelated fractures are needed
Residential Care Summit
2. Fall prevention
2.1 Distinction between fall prevention programmes in the community and in nursing homes
Place
Focused on
Community
Individual
Nursing home
Setting
Residential Care Summit
2.2 Components of fall prevention programmes in care homes 2.2.1 Prevention of falls
Staff training
Consequent documentation of falls Fall conferences …
Strength and balance training
Residential Care Summit
Environmental adaptations
Lighting, grips, … New assistive devices sensor mats; specific beds
Medication review
Reduction of psychotropic drugs (Acceptance and motivation by GP important)
Residential Care Summit
2.2.2 Attenuation of the impact of falls
Hip protector
Osteoporosis therapy
Vitamin D Controversial: antiresorptive agents
Adaptation of the underground / floor
Residential Care Summit
2.3 Effectiveness of measures / programmes in nursing homes 2.3.1 Reduction of falls Cameron et al., Cochrane review in preparation (provisional results) Effective?
Comment
Multifactorial intervention
Yes
if delivered by a multidisciplinary team and includes exercise
Supervised exercise
Inconsistent
May even increase fall risk
Vitamin D
Yes
≥ 800 IU
Clinical medication review
Yes
Only 1 trial
Residential Care Summit
2.3.2 Reduction of fractures Effective?
Reference
Comment
Hip protector
Yes
Parker et al. BMJ 2007; Sawka et al. J Clin Epidemiol. 2007
Number of protected falls – new indicator of quality
Vitamin D
Yes
Parikh et al. JAGS 2009
Multifactorial intervention
Yes
Cameron et al. (in preparation)
only 3 small trials
Residential Care Summit
2.4 Ulm fall prevention study
Cluster-randomised trial Time period: 1998-99 Setting: 6 nursing homes in Ulm / Southwest Germany
Intervention
Multifactorial fall prevention program exercise, staff training, environmental adaptations, hip protectors
Results
Reduction of falls by 44% and of fallers by 30%
Becker et al. JAGS 2003; 51:306-13
Residential Care Summit
2.4.1 Benefit in specific populations A) Subgroup analysis stratified by cognition
Intact memory (n = 406)
Impaired memory (n = 319)
HRR: 0.91 (0.68-1.22)
HRR: 0.49 (0.35-0.69)
1. 00
1. 00
0. 75
0. 75
0. 50
0. 50
0. 25
0. 25
0. 00
0. 00
0
50
100
150
200
250
300
350
400
0
50
st _sur v STRATA:
I nt Kont =0
Censor ed I nt Kont =0
100
150
200
250
300
350
400
st _sur v
I nt Kont =1
Censor ed I nt Kont =1
STRATA:
I nt Kont =0
Censor ed I nt Kont =0
I nt Kont =1
Censor ed I nt Kont =1
Rapp et al. JAGS 2008
Residential Care Summit
B) Subgroup analysis stratified by depressive symptoms
No depressive symptoms (n = 407)
Depressive symptoms (n = 318)
HRR: 0.63 (0.46-0.86)
HRR: 0.82 (0.61-1.12)
1. 00
1. 00
0. 75
0. 75
0. 50
0. 50
0. 25
0. 25
0. 00
0. 00 0
50
100
150
200
250
300
350
400
0
50
STRATA:
I nt Kont =0
Censor ed I nt Kont =0
100
150
200
250
300
350
st _sur v
st _sur v I nt Kont =1
Censor ed I nt Kont =1
STRATA:
I nt Kont =0
Censor ed I nt Kont =0
I nt Kont =1
Censor ed I nt Kont =1
Rapp et al. JAGS 2008
400
Residential Care Summit
2.5 Dissemination Does the programme work in daily practice? Main goals: Introduction of a program in > 50 % of all nursing homes in Baden-Württemberg (10.7 million inhabitants; 1200 care homes) – started 2003 Bavaria (12.5 million inhabitants; 1400 care homes) – started 2007
Reduction of hip fractures from > 5 % to < 3 % p.a.
Implemented by a health insurance company Since 2003 more than 1000 nursing homes have been included
Residential Care Summit
2.5.1 Implementation
Change agents: 1-2 per facility (~ 2,500) Standardised falls reporting Exercise: program twice weekly, 50 % cost coverage Exercise instructor: physiotherapist followed by nursing staff (~ 1,800) Manual, back office support, newsletter, website in place Hip protectors: test kit; no reimbursement Vitamin D use and medication review
Residential Care Summit
2.5.2 Participation rates in exercise and availability of hip protectors (subgroup of 49 Bavarian homes, 4300 residents)
50
Participation rates in exercise by care home 40
Availability of a hip-protector (%)
Participation rate (%)
40
30
20
10
Availability of hip protectors by care homes
30
20
10
0
0
Nursing home
Median (Range): 13,5% (3,4-47,8)
Nursing home
Median (Range): 7,5% (0-37,5)
Residential Care Summit
A) Strength and balance training: Predictors of participation
B) Hip protectors: Predictors of availability
Residential Care Summit
2.6 Evaluation of the dissemination
Effect on falls (Pre-Post comparison) Fall reduction: 10 % - 25 % Reduction of hospital admissions: 8 % - 22 %
Effect on hip fractures
Baden-Württemberg 2003-2004 (1,300 homes; 50,000 residents) Fall prevention program
Year of the intervention Hazard rate ratio* (95% CI)
Year after the intervention Hazard rate ratio* (95% CI)
Analysis A No (control homes from Baden-Württemberg) Yes (intervention homes from Baden-Württemberg)
1.00 0.96 (0.83-1.11)
1.00 1.05 (0.90-1.22) Analysis B
No (control homes from Bavaria) Yes (intervention homes from Baden-Württemberg)
1.00 1.00 (0.87-1.16)
1.00 0.98 (0.85-1.14)
* adjusted for age, gender, size of the nursing home, and year of intervention
Rapp et al., JAGS, in press
Residential Care Summit
Evaluation Bavaria 2007 Characteristics of the study population Intervention group
Control group
Nursing homes N Number of beds Median (range)
256
915
96 (22-323)
66 (3-356)
Study population Gender Male, n (%) Female, n (%)
2,887 (21.2) 10,758 (78.8)
6,862 (21.6) 24,946 (78.4)
84.9 (65.3-105.5) 84.3 (7.5)
84.9 (65.2-107.4) 84.2 (7.7)
327
919
Age (years) Median (range) Mean (SD) Femoral fractures, n
Residential Care Summit
Bavaria 2007 Intervention
Controls
2001
Fracture rate † 38.26
Fracture rate † 39.83
Hazard ratio* (95% CI) 0.97 (0.85-1.10)
2002
44.18
42.66
1.04 (0.92-1.18)
2003
41.00
41.12
1.00 (0.98-1.02)
2004
42.41
41.40
1.03 (0.90-1.17)
2005
37.30
40.69
0.92 (0.80-1.05)
2006
40.65
41.23
0.99 (0.86-1.13)
2007
33.45
41.23
0.81 (0.72-0.92)
Year
* adjusted for sex, age and level of care † Fractures/1000 person-years
First intervention year: reduction of hip fractures by nearly 20%
Residential Care Summit
3. New components
Incidence rate of fractures as a function of time since admission to a nursing home All fractures combined and femoral fractures
140
All Femur
120 100
Fractures / 1000 py
Fractures / 1000 py
Fractures of the upper limb and the lower limb (except femur)
80 60 40 20
35
Upper limb Lower limb except femur
30
25
20
15
10
5
0 1
2
3
4
5
6
7
8
9
Months after admission
10
11
12
0 1
2
3
4
5
6
7
8
9
10
11
12
Months after admission
Rapp et al., Osteoporos Int 2009
Residential Care Summit
Incidence rate of fractures as a function of time since admission to a nursing home Stratified by the need of care (level of care)
160
Level of care 1 Level of care 2 Level of care 3
140
Fractures / 1000 py
120 100 80 60 40 20 0 1
2
3
4
5
6
7
8
9
10
11
12
Months after admission
Rapp et al., Osteoporos Int 2009
Residential Care Summit
3.1Potential causes and implications Causes
Implications
New environment Morbidity-related weakness / prior hospitalisation (subacute) delirium Fall prevention major topic immediately after admission Intensive supervision and guidance Hip protectors (Pool) Setting up the bedroom / use of established patterns
Adopted in the Bavaria project Training of change agents Newsletter
Residential Care Summit
Complementary aims of fall prevention
Increase mobility and autonomy Reduce anxiety, fear of falling and social withdrawal Decrease use of restraints Increase quality of life
Decrease care giver burden (costs)
Part of a general concept / policy in the nursing home
Residential Care Summit
4. Summary
The burden of falls and fall-related injuries is particularly high in residents of care homes Multifactorial fall prevention programmes (and not single measures) seem to be most effective To reduce falls To reduce hip fractures
Residents newly admitted to nursing homes should be regarded as a high risk group for falls and fractures
Residential Care Summit
Victoria, November 2009