Managing Falls in Older Patients With Cognitive Impairment

Morag E Taylor; Kim Delbaere; Jacqueline CT Close; Stephen R Lord

Aging Health. 2012;8(6):573-588.

Future Perspective & Conclusion

In this population of vulnerable older people, with double the risk of falls, more work is needed to enable identification of effective fall prevention strategies. We have a limited understanding of fall risk in what is a heterogeneous group of older individuals. As our understanding of risk in people with cognitive impairment and dementia increases, we need large RCTs testing interventions, particularly in the community. Interventions may include: exercise, involving primarily balance and coordination activities, assessment of mood and autonomic function, medication review and bone health management. The core components of an intervention designed to prevent falls will depend on the setting, for example RACFs versus the community and need to be tested rigorously in the selected population. Once robust evidence is available, the challenge will be in translating research into cost-effective clinical practice.

1 2 3 4 5 6

Next Section

Acknowledgements
SR Lord is currently a NHMRC Senior Principal Research Fellow and K Delbaere is a NHMRC Career Development Fellow.

Abstract
Epidemiology of Falls & Cognitive Impairment & Dementia
Fall Risk Factors in Cognitive Impairment & Dementia
Assessment of Fall Risk in Cognitive Impairment/Dementia
Fall Prevention in Cognitively Impaired Older People
Future Perspective & Conclusion
References
Sidebar

Table 1. Risk factors for falls in people with cognitive impairment and dementia ^†.
Risk factor	Association
*Psychosocial & demographic factors*
Advanced age	*
Male gender	*
History of falls	***
Walks with an aid	**
Inactivity	*
ADL limitations	***
Living alone	ns
*Medical factors*
Dementia duration	*
Dementia severity	**
Parkinson's disease/parkinsonism	**
Arthritis/musculoskeletal complaint	**
Symptomatic orthostatic hypotension	*
Peripheral neuropathy	*
Autonomic neuropathy	*
Cardiac arrhythmia	*
Cataracts	*
Impaired vision	*
*Medication factors*
Psychoactive medications	**
Antidepressants	**
Cardiovascular medications	*
NSAIDs	–
Polypharmacy	**
*Balance & mobility factors*
Impaired stability when standing	***
Impaired stability when leaning and reaching	*
Impaired gait and mobility	***
*Sensory & neuromuscular factors*
Poor visual contrast sensitivity	–
Muscle weakness	*
Slow reaction time	*
Impaired proprioception	–
*Neuropsychological factors*
Wandering/behavioral factors	**
Attention and orientation	*
Poor memory	*
Depression/depressive symptoms	**
Impaired executive function	**
Anxiety	*
Fear of falling	ns
*Environmental factors*
Home hazards	–

†Studies including samples of people with cognitive impairment and/or dementia. In determining the rating, we used reported results from both univariate and multivariate analyses.
***Good evidence as a risk factor for falls, multiple studies have reported as a risk factor, consistent findings.
**Some evidence as a risk factor for falls, a few studies have reported as a risk factor, generally consistent findings.
*Limited evidence as a risk factor for falls, one study or conflicting evidence as a fall risk factor. – Little or no evidence of association (not found in published studies despite research to examine the issue). ADL: Activities of daily living; ns: Not studied.

Table 2. Intervention studies specifically targeting cognitively impaired older people or presenting subgroup analysis for this group.
Study (year)	Cohort	Study design	Intervention	Outcome	Ref.
*Community*
Shaw et al. ^† (2003)	n = 274 MMSE <24 ≥65 years of age presenting to ED with a fall 22% community dwellers	RCT Prospective 1 year follow-up	I: multifactorial assessment and treatment (including medical and medication assessment and management, cardiovascular assessment and management, supervised exercise for 3 months, self-directed exercise for walking aid prescription, footwear, home hazard modification) C: usual care	No significant difference between control and intervention for proportion of fallers (RR: 0.92 [0.81–1.05]), number of falls (estimated mean difference: -0.02 [-0.32–0.09]) or time to first fall (log rank test: p = 0.459)	[87]
*Hospital*
Stenvall et al. (2007)	n = 64 with dementia Patients ≥70 years of age with femoral neck fracture Geriatric and orthopedic wards	Stratified RCT Follow-up = hospitalized period	I: geriatric unit specializing in geriatric orthopedic patients, comprehensive geriatric assessment, management and rehabilitation C: usual care, orthopedic ward	Fall incidence rate lower among people with dementia (IRR: 0.07 [0.01–0.57]). In those with dementia, one faller in intervention group, compared with 11 in control (p = 0.006)	[109]
Haines et al. (2011)	n = 300 with CI 'Older adults' Acute, subacute and rehabilitation wards	3-group RCT Follow-up = hospitalized period	I: patient education 1. complete program (written, video, 1:1 with physiotherapist) 2. materials only (written and video) C: usual care	Injurious falls increased in the complete program in CI participants (7.49 injurious falls/1000 patient days vs 2.89 in controls) Proportion of CI fallers was similar (complete program 26%, control 24%)	[111]
*Residential care*
Jensen et al. (2003)	n = 170, MMSE <19 n = 171, MMSE ≥19 ≥65 years Nine residential facilities in Sweden	Cluster RCT nonblinded 34 week follow-up	I: multifactorial fall prevention program (including staff education, environmental adjustments, exercise, medication review, aids, hip protectors and post fall conferences) C: usual care	Lower cognition group: no significant difference time to first fall (p = 0.420), no significant difference in incidence rate of falls (IRR: 1.05 [0.84–1.30]) and significant reduction in femoral fractures in the intervention group (I = 0, C = 10; p = 0.006) Higher cognition group: significantly longer time to first fall (p = 0.001), reduced incidence rate of falls (IRR: 0.61 [0.48–0.78])	[100]
Shaw et al. ^† (2003)	n = 274 MMSE <24 ≥65 years of age presenting to ED with a fall 78% residents of care facilities	RCT Prospective 1-year follow-up	I: multifactorial assessment and treatment (including medical and medication assessment and management, cardiovascular assessment and management, supervised exercise for 3 months, self directed exercise for 1 year, walking aid prescription, footwear and home hazard modification) C: usual care	No significant difference between control and intervention for proportion of fallers (RR: 0.92 [0.81–1.05]), number of falls (estimated mean difference: -0.02 [-0.32–0.09]) or time to first fall (log rank test: p = 0.459)	[87]
Detweiler et al. (2005)	n = 8 Residents with the 'highest fall incidence' 60 bed dementia special care unit	Pre/post 4 months	I: trained certified nursing assistant; 1 day/night shift provided care to the eight residents C: usual care	Decrease in total number of falls for the intervention period (p = 0.024)	[106]
Rapp et al. (2008)	n = 148 with cognitive impairment ≥60 years of age, all residents Six nursing homes in Germany	Cluster RCT 12-month follow-up	I: multifactorial (including staff and resident education, environmental, hip protectors, exercise twice/week for 12 months) C: usual care	Increased time to first fall (HR: 0.49 [0.35–0.60]) and reduced incidence rate of falls (IRR: 0.43 [0.28–0.66]) in the cognitively impaired group	[97]
Neyens et al. (2009)	n = 518 Psychogeriatric nursing home patients 12 nursing homes in the Netherlands, psychogeriatric wards MMSE (C: 7 ± 7; I: 9 ± 8)	Cluster RCT 12-month trial Mean duration of inclusion 0.5 years	I: general medical assessment, multidisciplinary fall prevention team, specific fall risk evaluation tool, general and individual fall prevention activities (e.g., assistive and protective aids, exercise and medication review) C: usual care	Intervention group had lower adjusted fall incidence rate (IRR: 0.64 [0.43–0.96]). Subgroup analysis showed that the longer subjects participated in the intervention the lower the fall rate (≥0.7 years; IRR: 0.43 [0.19–0.94])	[104]

^†78% residents of care facilities, 22% community dwellers.
C: Control; CI: Cognitive impairment/cognitively impaired; ED: Emergency department; HR: Hazard ratio; I: Intervention; IRR: Incident rate ratio; MMSE: Mini-Mental State Examination; RCT: Randomized controlled trial; RR: Relative risk.

Sidebar

Executive Summary

Epidemiology of Falls & Cognitive Impairment/Dementia

Dementia affects 5–7% of the population over 60 years of age and 60% of this population fall each year.
Older people with dementia have high rates of fall-related injury with consequences including institutionalization and death.

Fall Risk Factors in Cognitive Impairment & Dementia

Neurodegenerative processes associated with dementia may lead to a different 'fall risk profile' when compared with cognitively intact individuals; however, some risk factors may be shared.
Previous falls and walking aid use are associated with future falls; the association with age and gender is inconsistent.
The following factors increase fall risk: Lewy body and Parkinson's disease dementia, psychotropic and multiple mediation use, activity of daily living limitations, poor balance and impaired gait.
Sensorimotor factors have not been studied sufficiently to draw definitive conclusions, although slow reaction time appears to be associated with falls.
Depression, depressive symptoms and behavioral difficulties are associated with falls.
Environmental factors have not been sufficiently studied; available evidence suggests no clear association between home hazards and falls.
Screening and assessment tools have been validated in cognitively impaired older people and help to identify those at high risk of falls.

Fall Prevention in Cognitively Impaired Older People

In the community, there is little evidence for fall prevention strategies. Few studies have defined the cognitive level of their included samples adequately and even fewer have conducted subgroup analyses on cognitively impaired participants.
In residential care, multifaceted programs may prevent falls and fractures, particularly if they involve exercise, are driven by individual comprehensive geriatric assessment and include vitamin D supplementation in residents who are deficient.
In hospitals, multifaceted programs may prevent falls, particularly if they involve strategies such as specialist geriatric postoperative management. Multimedia patient education with/without allied health follow-up has not been demonstrated to prevent falls.

Other Potential Interventions

Supervised exercise, regular medication review, assessment and management of bone health, depression and orthostatic hypotension, as well as regular vision checks with prescription of appropriate lenses, may prevent falls and/or fall-related injuries in cognitively impaired older people.

Future Perspective & Conclusion

More good quality studies elucidating fall risk in people with dementia are needed.
With improved knowledge of fall risk factors, targeted interventions require evaluation in large randomized controlled trials.
Future challenges will involve translating successful interventions into cost-effective clinical practice.