JAMDA xxx (2014) 1e6
JAMDA journal homepage: www.jamda.com
Original Study
Efficacy of Physical Exercise in Preventing Falls in Older Adults With Cognitive Impairment: A Systematic Review and Meta-Analysis Wai Chi Chan MRCPsych a, *, Jerry Wing Fai Yeung PhD b, Corine Sau Man Wong MSocSc a, Linda Chiu Wa Lam MD c, Ka Fai Chung MRCPsych a, James Ka Hay Luk FRCP d, Jenny Shun Wah Lee FRCP e, Andrew Chi Kin Law FRCPC a a
Department of Psychiatry, The University of Hong Kong, Hong Kong School of Chinese Medicine, The University of Hong Kong, Hong Kong Department of Psychiatry, The Chinese University of Hong Kong, Hong Kong d Department of Medicine and Geriatrics, Fung Yiu King Hospital, Hong Kong e Department of Medicine and Geriatrics, Tai Po Hospital, Hong Kong b c
a b s t r a c t Keywords: Cognitive disorders dementia exercise accident prevention accidental falls
Objective: Numerous studies have reported the prevention of falls through exercise among cognitively healthy older people. This study aimed to determine whether the current evidence supports that physical exercise is also efficacious in preventing falls in older adults with cognitive impairment. Methods: Two independent reviewers searched MEDLINE; EMBASE; PsycINFO; the Cumulative Index to Nursing & Allied Health Literature; the Cochrane Central Register of Controlled Trials; the Cochrane Bone, Joint, and Muscle Trauma Group Specialized Register; ClinicalTrials.gov; and the UK Clinical Research Network Study Portfolio up to July 2013 without language restriction. We included randomized controlled trials that examined the efficacy of physical exercise in older adults with cognitive impairment. The methodological qualities of the included trials were appraised according to the criteria developed for the Cochrane review of fall prevention trials. The primary outcome measure was the rate ratio of falls. A meta-analysis was performed to estimate the pooled rate ratio and summarize the results of the trials on fall prevention through physical exercise. Results: Seven randomized controlled trials involving 781 participants were included, 4 of which examined solely older people with cognitive impairment. Subgroup data on persons with cognitive impairment were obtained from the other 3 trials that targeted older populations in general. The metaanalysis showed that physical exercise had a significant effect in preventing falls in older adults with cognitive impairment, with a pooled estimate of rate ratio of 0.68 (95% confidence interval 0.51e0.91). Conclusions: The present analysis suggests that physical exercise has a positive effect on preventing falls in older adults with cognitive impairment. Further studies will be required to determine the modality and frequency of exercise that are optimal for the prevention of falls in this population. Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
The global population is aging at an unprecedented pace. In 2012, 23% of the population in the more developed regions and 9% in the less developed regions were aged 60 years or older.1 It is estimated that by 2050 the proportion of older citizens will increase to 32% and 19% in developed and developing countries, respectively.1 Because the prevalence of dementia rises rapidly after the age of 60 years,2 a dramatic change in population demographics will inevitably increase the number of people affected by this disorder, and make dementia a pressing health care issue. The authors declare no conflicts of interest. * Address correspondence to Wai Chi Chan, Department of Psychiatry, 2/F New Clinical Building, Queen Mary Hospital, 102 Pok Fu Lam Road, Hong Kong. E-mail address: [email protected] (W.C. Chan). http://dx.doi.org/10.1016/j.jamda.2014.08.007 1525-8610/Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
One of the major issues in dementia care is the prevention of falls, for which evidence shows that persons with cognitive impairment are at an elevated risk. The annual incidence of falls among older adults with cognitive impairment is about twice that in the elderly who are cognitively healthy.3,4 Even after taking other known risk factors into account, cognitive impairment remains an independent risk factor for falls.4 Studies also have shown that people with cognitive impairment who fall suffer from poorer outcomes than those who are cognitively healthy. For example, people with dementia have a greater likelihood of needing hospital admission for orthopedic crises than people without dementia, 5 and have a significantly greater mortality rate after sustaining fractures.6,7
2
W.C. Chan et al. / JAMDA xxx (2014) 1e6
The prevention of falls in persons with cognitive impairment is challenging. A number of psychological (eg, behavioral symptoms,4 cognitive impairment,8 poor attention and orientation,9 anxiety,9 and depressive symptoms10), physiological (eg, increased postural sway,9 symptomatic orthostatic hypotension,10 impaired simple reaction time,11 and leaning balance11), and medical risk factors (eg, Lewy body disorder,10 antipsychotic use,4 and antidepressant use9) pertinent to older people with cognitive impairment have been identified. The prevention of falls in persons with cognitive impairment is particularly relevant in long-term care settings, because approximately half of all nursing home residents have dementia.12 It is thus essential to understand which interventions are most efficacious in persons with cognitive impairment before effective fallprevention programs can be devised for long-term care facilities. Fall prevention has been studied extensively in the general older population, and strategies including group and home-based exercise programs, home safety interventions, and multifactorial assessment and intervention programs have been found to reduce the rate of falls and the risk of falling.13 However, the efficacy of these interventions, including the more established types, such as physical exercise, among cognitively impaired individuals remains unclear, because most previous studies on fall prevention either excluded subjects with cognitive impairment or did not conduct subgroup analyses related to cognitive impairment status.14 It is thus not surprising that although several systematic reviews have been conducted to examine the effect of physical exercise on the prevention of falls in the elderly,13,15e17 only 115 included randomized controlled trials (RCTs) conducted specifically on cognitively impaired older people. This particular review,15 for which the search was performed in 2004, found that the evidence on the effect of physical exercise in preventing falls among cognitively impaired populations was inconclusive because only 3 RCTs with conflicting results on the number of falls were available. Subsequently, a meta-analysis that investigated the impact of dementia on the efficacy of fall-prevention programs also found no conclusive evidence to suggest an association between the two.18 However, of the 43 studies included in this meta-analysis, which comprised studies published up to 2005, only 1 examined people with dementia exclusively and another performed a subgroup analysis to compare the efficacy of the preventive programs in participants with or without dementia.18 This partly explains why, based on the studies of older adults with known cognitive impairment published up to 2006, the Panel on the Prevention of Falls in Older Persons of the American Geriatrics Society and British Geriatrics Society concluded that there was insufficient evidence to recommend or oppose single or multifactorial interventions for fall prevention in people with dementia.19 We thus conducted an updated systematic review and meta-analysis to summarize the current evidence on the efficacy of physical exercise, one of the more established fall-prevention interventions,13,16,17 in persons with cognitive impairment. Methods Study Selection The studies included in this review were RCTs that compared the efficacy of physical exercise with routine medical care or other controlled activities in preventing falls in older people with cognitive impairment. Physical exercise referred to planned, structured, and repetitive movement to improve or maintain one or more components of physical health.18 Cognitive impairment needed to be suggested by either a standardized cognitive assessment or a diagnosis of dementia established according to accepted criteria. For trials that did not specifically report information on cognitively impaired or
demented older people, original subgroup data were requested from the authors. To better understand the efficacy of physical exercise in the prevention of falls, we excluded trials in which exercise was only part of a multifactorial program. Literature Search We searched MEDLINE (from 1946); EMBASE (from 1980); PsycINFO (from 1967); the Cumulative Index to Nursing & Allied Health Literature (from 1982); the Cochrane Central Register of Controlled Trials (the latest issue); the Cochrane Bone, Joint, and Muscle Trauma Group Specialized Register; ClinicalTrials.gov; and the UK Clinical Research Network Study Portfolio up to July 2013 using the grouped terms (dementia or alzheimer* or lewy body or lewy bodies or cognitive disorders or mild cognitive impairment or memory disorder or memory disorders) and (exercise or physical activit* or physical training or motor activit* or gait training or balance training or bicycling or swim* or gym* or walk* or danc* or yoga or “tai chi” or “Tai Ji”). No language restriction was applied. We also screened reference lists of relevant reviews.13,15e17 Two reviewers were responsible for searching the databases. Any disagreement was resolved by consensus, discussion, and third-party adjudication (JWFY). We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) in reporting this systematic review. Data Extraction and Quality Evaluation One author (WCC) extracted the data independently and the other (JWFY) checked the accuracy of the extracted data. The following variables were extracted from each study: study design, participant characteristics, treatment and control protocols, and outcome parameters. The primary outcome measure was the rate ratio of falls (ie, the ratio of the total number of falls per unit of person time in any 2 groups in each trial). When the rate ratio was not reported, it was calculated from the ratio of the total number of falls divided by the total length of time during which falls were monitored (person-years) in the 2 comparison groups. A secondary outcome measure was the number of participants who sustained fractures. The methodological qualities of the trials included were appraised according to the 7 domains in Cochrane’s tool for assessing risks of bias: random sequence generation, allocation concealment, blinding of the participants, blinding of the assessors, incomplete outcome data, selective reporting, and other sources of bias. In the domain of other sources of bias, we assessed bias in the recall of falls due to unreliable methods of ascertainment,20 using the criteria developed for the Cochrane review of fall-prevention trials.13 The risk of bias in each domain was rated as low, high, or uncertain. Data Synthesis and Analysis The generic inverse variance method in the Review Manager software (RevMan 5.1, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) was used to group the results of the trials and compile the forest plot. The rate ratio (RaR) was calculated for the rate of falls between 2 groups with a 95% confidence interval (CI). Heterogeneity between the trials was assessed using c2 and I2 statistics. We used a fixed effects model in the absence of statistical heterogeneity (heterogeneity test, P .10) and a randomeffects model when heterogeneity was present (heterogeneity test, P < .10). A sensitivity analysis was carried out by excluding 1 study at a time to explore whether the results were influenced by 1 large study or by a study with an extreme result. The possibility of publication bias would have been assessed using funnel if 10 or more trials had been included in the meta-analysis.21
W.C. Chan et al. / JAMDA xxx (2014) 1e6
Results Identification of Studies The search yielded 16,219 potential studies for review. Among these, 5709 were identical citations and 10,450 were excluded because of irrelevance (eg, studies that were not aimed at fall prevention, studies that used multifactorial interventions). The full texts of 60 articles were retrieved for detailed assessment. Nine of these studies targeted the general elderly population, and original subgroup data relating to participants with cognitive impairment were requested from the authors. The authors of 3 of the studies subsequently provided us with the required data. At the end of the assessment, 53 studies were excluded for various reasons (Figure 1), and 7 trials22e28 met the inclusion criteria for further analysis. Description of the Studies Table 1 summarizes the study populations and design of the 7 trials included. Trials 1 to 4 examined the efficacy of physical exercise in preventing falls in cognitively impaired older persons.22e25 Trials 5 to 7 targeted the general elderly population but subgroup data and information on participants with cognitive impairment were provided by the authors.26e28 The total number of participants in these trials was 781, and the sample size of each trial ranged from 20 to
3
210. Participants in 6 of 7 trials had a mean age of 80 years or older.22,24e28 The mean Mini-Mental State Examination (MMSE) score of participants in 5 of the trials was 16 or higher,22,23,25e27 whereas the mean MMSE score in the trial that recruited nursing home residents was 8.8.24 The remaining trial assessed the cognitive function of participants with the Short Portable Mental Status Questionnaire, in which they made an average of 3.78 errors.28 Six trials had 2 parallel arms and 1 had 3 parallel arms. The intervention groups in all trials underwent exercise in a group setting, and 1 of the studies provided an extra arm for exercise in a home-based setting.23 The 2 exercise groups (group-based and home-based) in this trial were combined as a single intervention group for comparison with the control. The essential elements in the physical exercises were balancing and strength training,22e28 flexibility,22,24,26 walking,23e26,28 and coordination.23,25,26,28 The time period, frequency, and duration of the group exercises were identical in 3 of the trials (twice a week, 60 minutes each, and lasted for 12 months).23,24,26 The frequency of intervention in the other 4 trials varied, and the duration of exercise spanned a shorter period of time (3e6 months).22,25,27,28 The number of falls was reported by caregivers or nursing home staff and was used as an outcome assessment in all trials. Two of the trials also collected information on fractures from registries or caregivers.23,24 In addition, physical performance,22e26,28 functional mobility,22,23,25,26,28 and balance ability22,24,26,28 were compared before and after the intervention in some of the trials.
Fig. 1. Flowchart of systematic review.
60 min Group exercise NA 85.9 n ¼ 54 Moseley et al, 2009* 7
16.0 84.2 Rosendahl et al, 2008* n ¼ 100 6
22.8 81.0 n ¼ 141 Lord et al, 2003* 5
21.7 82.1 n ¼ 122 Zieschang et al, 2013 4
AD, Alzheimer disease; DSM, Diagnostic and Statistical Manual of Mental Disorders; MMSE, Mini-Mental State Examination; NA, not available; SPMSQ, Short Portable Mental Status Questionnaire. *Studies where subgroup data were provided by authors.
45 min
Group exercise
Group exercise
5 times every 29 sessions Activities performed while sitting 2 wk (over 3 mo) (eg, watching films, reading, singing) 2/d 16 wk A tailored program of limited weight-bearing exercises
Either flexibility and relaxation program 12 mo
Motor placebo group training 3 mo
120 min (Intervention 2/wk group); 60 min (Control group) 60 min 2/wk Group exercise
12 mo 2/wk 60 min Group exercise
Patients with AD living in nursing homes Patients with mild to moderate dementia; places of abode not specified Older people with MMSE ranged 20e24 living at self-care apartment villages or intermediate-care hostels Older people diagnosed as dementia by DSM criteria living in residential care facilities Older people diagnosed as cognitive impaired or demented by SPMSQ; places of abode not specified 83.0 n ¼ 134 Rolland et al, 2007 3
18.0 78.0 n ¼ 210 Pitkälä et al, 2013 2
8.8
12 mo 2/wk
Routine medical care, oral and written advice on nutrition and exercise methods Routine medical care
16 wk 2/wk
45 min Demented patients living at home or Group exercise institutions with at least 2 previous falls Home-dwelling patients with AD Tailored home-based 60 min living with spousal caregivers exercise, group exercise 16.3 81.4 n ¼ 20 Toulotte et al, 2003 1
Intervention Duration Intervention Frequency Intervention Time Intervention Mode No. of Mean Mean Study Population Participants Age, y MMSE Trial Authors
Table 1 Description of the Randomized Controlled Trials With Regard to Study Population and Interventions
Daily routine
W.C. Chan et al. / JAMDA xxx (2014) 1e6
Control Group
4
In general, the dropout rates from the exercise interventions were low. The short-term attritions (up to 6 months) ranged from 0%22 to 22.6%,25 and the long-term attritions (up to 12 months) were between 11.3%26 and 35.5%.25 Adherence to the exercise programs by persons with dementia was reported in 2 of the studies.23,24 The overall adherence to exercise programs ranged from moderate to high. In the FINALEX study, 92.9% of the participants took part in at least half of the home-based exercise sessions and 78.6% in at least half of the groupbased exercise sessions.23 In another study that examined exercise interventions among nursing home residents, 19.4% completed more than two-thirds of the sessions and 28.4% completed one-third to twothirds of the sessions.24 Quality Assessment According to the Cochrane Criteria Table 2 summarizes the risk of bias of the trials included that was assessed by applying the Cochrane criteria. Sequence generation and allocation concealment were adequately described in 5 trials.23e25,27,28 Randomization in these trials was performed by people who were not involved in the intervention or the assessment. The remaining trials did not mention the sequencegeneration process or the allocation concealment method and hence had an uncertain risk of selection bias. One trial was described as a double-blind controlled trial,25 in which the participants in the control group performed sham activities, including stretching and other nondemanding motor activities under the supervision of trainers. Participants were considered to be blinded to the intervention because they were not informed of the difference in effectiveness of the 2 training regimens. Two trials blinded the assessors to group allocation but the participants were not blinded.24,28 The remaining trials were either described as nonblind controlled trials23 or did not mention whether participants and assessors were blinded.22,26,27 The risk of attrition bias due to incomplete outcome data was rated as high in 1 trial,25 in which the proportion of dropouts compared with the observed risk of falls was sufficient to induce a clinically important bias. All trials had a low risk of reporting bias due to selective reporting. Falls were recorded prospectively by such methods as a fall diary in 5 trials21e24,26 and a questionnaire reporting the number of falls in the past month in 1 trial.25 Potential recall bias due to the measurement of falls at intervals longer than 1 month was observed in 1 of the trials.28 Efficacy Assessment Figure 2 shows the forest plot of the effect of physical exercise on preventing falls in cognitively impaired older people. All trials provided figures for falls during the intervention period. Four of 7 trials showed a significant effect of exercise on reducing the number of falls. The pooled estimated RaR for falls was 0.68 (95% CI 0.51e0.91, I2 ¼ 79%, P ¼ .01, 7 trials). No significant differences in the number of fractures were reported between the intervention and control group during the intervention period for both Trial 2 (9 versus 4)23 and Trial 3 (5 versus 2).24 The pooled estimated RaR for fracture-related to falls was 1.47 (95% CI 0.56e3.81, I2 ¼ 0%, P ¼ .43, 2 trials). Sensitivity Analyses In a sensitivity analysis that excluded 1 study at a time, the pooled rate ratio was lowest (0.63, 95% CI 0.46e0.85, I2 ¼ 74%) when Trial 3, which recruited patients with poorer cognitive function (mean MMSE score, 8.8), was excluded. The pooled rate ratio was still significant (0.73, 95% CI 0.54e1.00, I2 ¼ 78%, P ¼ .049) when
W.C. Chan et al. / JAMDA xxx (2014) 1e6
5
Table 2 The Risks of Bias of the Included Trials Using Cochrane’s Criteria Trial
Authors (Year)
Random Sequence Generation
Allocation Concealment
Blinding of the Participants
Blinding of the Assessors
Incomplete Outcome Data
Selective Reporting
Method of Ascertaining Falls
1 2 3 4 5 6 7
Toulotte et al (2003) Pitkälä et al (2013) Rolland et al (2007) Zieschang et al (2013) Lord et al (2003) Rosendahl et al (2008) Moseley et al (2009)
U L L L U L L
U L L L U L L
U U U L U U U
L H L L H U L
L L L L H L L
L L L L L L L
L L L L L L H
H, high risk of bias; L, low risk of bias; U, uncertain risk.
Trial 5, which recruited patients with relatively better cognitive function (mean MMSE score, 22.8), was excluded. The overall significant result for fall prevention was not affected by any particular study. Discussion To the best of the authors’ knowledge, this is the first metaanalysis to specifically examine the efficacy of physical exercise in preventing falls in cognitively impaired older people. As previous studies have focused on exercise for the general older population, there has been a paucity of RCTs that examined cognitively impaired individuals. The small number of RCTs investigating these populations caused difficulties in the synthesis of the data. The only systematic review of the effect of physical exercise in cognitively impaired populations, which included 3 RCTs reporting on the number of falls, found that the evidence was inconclusive.15 In this up-to-date review of the available literature, we found that physical exercise was associated with a reduced risk of falls in individuals with cognitive impairment. Seven RCTs involving 781 participants were included in the present study. Physical exercise had a significant effect on preventing falls in people with cognitive impairment, with a pooled RaR of 0.68, which was similar to that achieved by group exercise (0.71) and home-based exercise (0.68) in community-dwelling older people.13 Only 2 of the trials reported the number of fractures during the study period but neither showed a significant difference between the intervention and control groups. The exercise interventions in all of the trials consisted of balancing and strength training. Other exercise components included flexibility, walking, coordination training, and exercise involving executive function. It has been well documented that exercise improves mobility, balance, muscle strength, and physical function in the general older population.13 Recent studies have further suggested that exercise may reduce falls by mechanisms other than improving physiological function. In particular, exercise may have an effect by improving cognitive
performance, such as executive function.29 This mechanism may also be applicable to persons with cognitive impairment, because evidence has shown that exercise training not only improves physical, but also cognitive function in people with dementia.30 Five of the studies delivered exercise interventions twice a week, but more intensive training was also adopted by some of the trials included. The trial that provided the most intensive training (twice daily) resulted in one of the greatest reductions in the risk of falls but a similar magnitude was also achieved by less-intensive exercise. The finding that a greater degree of exercise is more effective in reducing falls in cognitively intact individuals31 has not been validated for persons with cognitive impairment. The optimal regimen of exercise interventions, therefore, requires further exploration. The programs in the trials included were implemented under the supervision of trained personnel, and met the special needs of persons with cognitive impairment, who had a poorer ability to understand and learn new information and often required closer supervision by and repeated explanations from the instructors. A higher level of supervision also allowed optimal progression of the exercise program and the achievement of a greater degree of exercise.28 The availability of close supervision probably accounted, at least in part, for the low attrition rates of the studies included. Other measures were also taken to enhance the participation of persons with cognitive impairment in the exercise programs, including those related to treatment settings (eg, provision of the interventions at the participants’ homes23,26), treatment components (eg, versatile exercises,23 the inclusion of “brain-training” exercises,23 the use of music,24 and a patient-centered approach25), treatment schedules (eg, a gradual increase in the intensity of the exercises24), and logistics (eg, the provision of exercises at no cost,26 and the avoidance of competition with other activities26). Whether single interventions are as effective as multifactorial interventions in preventing falls in older people has been a controversial issue.32,33 The present study offered additional evidence to support the theory that single interventions through physical exercise
Fig. 2. Effects of physical exercises on preventing falls in older adults with cognitive impairment.
6
W.C. Chan et al. / JAMDA xxx (2014) 1e6
have a protective effect and prevent falls in cognitively impaired older people. However, in view of the lack of a head-to-head comparison in persons with cognitive impairment, it is still not clear which approach is more effective in these populations. The current review is limited by the small number of RCTs investigating specifically cognitively impaired older people. We have overcome this limitation partially by obtaining subgroup data from the authors of Trials 5 to 7. Despite our efforts to maximize the number of studies included, the current review could analyze data from only 7 RCTs. Failure to obtain subgroup data on cognitively impaired individuals from the 6 studies that targeted the general elderly population constituted another limitation. In addition, the trials included varied with regard to the severity of cognitive impairment, treatment settings, and dropout rates. Due to their heterogeneity, the results of the current meta-analysis should be interpreted with caution. In summary, the present meta-analysis supported that physical exercise effectively reduces the risk of falls in persons with cognitive impairment. However, because the core components of fall-prevention programs depend on the characteristics of the older adults and the treatment settings, a one-size-fits-all solution is probably not available. For example, the complexity of the exercise program has to be adjusted according to the changing cognitive function. Effective fall-prevention exercise therefore needs to be individually tailored in terms of modality, frequency, and intensity. Further RCTs that investigate older adults with various cognitive and physical abilities will assist in the design of fall-prevention programs that meet their diverse needs. Acknowledgments The authors thank Ms Ruth Wong and Mr Matthew Lam for their assistance in the literature search, and Prof Stephen Lord, Dr Erik Rosendahl, and Dr Anne Moseley for their support in providing us with subgroup data of their respective studies. References 1. United Nations, Department of Economic and Social Affairs. World Population Prospects: The 2012 Revision, Highlights and Advance Tables. 2013. Available at: http://esa.un.org/unpd/wpp/Documentation/pdf/WPP2012_HIGHLIGHTS. pdf. Accessed February 1, 2014. 2. Prince M, Bryce R, Albanese E, et al. The global prevalence of dementia: A systematic review and metaanalysis. Alzheimers Dement 2013;9:63e75.e2. 3. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med 1988;319:1701e1707. 4. van Doorn C, Gruber-Baldini AL, Zimmerman S, et al. Dementia as a risk factor for falls and fall injuries among nursing home residents. J Am Geriatr Soc 2003; 51:1213e1218. 5. Toot S, Devine M, Akporobaro A, Orrell M. Causes of hospital admission for people with dementia: A systematic review and meta-analysis. J Am Med Dir Assoc 2013;14:463e470. 6. Diamantopoulos AP, Hoff M, Hochberg M, Haugeberg G. Predictors of shortand long-term mortality in males and females with hip fractureda prospective observational cohort study. PLoS One 2013;8:e78169. 7. Scandol JP, Toson B, Close JC. Fall-related hip fracture hospitalisations and the prevalence of dementia within older people in New South Wales, Australia: An analysis of linked data. Injury 2013;44:776e783. 8. Nazir A, Mueller C, Perkins A, Arling G. Falls and nursing home residents with cognitive impairment: New insights into quality measures and interventions. J Am Med Dir Assoc 2012;13:819.e1e819.e6.
9. Whitney J, Close JC, Jackson SH, Lord SR. Understanding risk of falls in people with cognitive impairment living in residential care. J Am Med Dir Assoc 2012; 13:535e540. 10. Allan LM, Ballard CG, Rowan EN, Kenny RA. Incidence and prediction of falls in dementia: A prospective study in older people. PLoS One 2009;4:e5521. 11. Taylor ME, Lord SR, Delbaere K, et al. Physiological fall risk factors in cognitively impaired older people: A one-year prospective study. Dement Geriatr Cogn Disord 2012;34:181e189. 12. Harris-Kojetin L, Sengupta M, Park-Lee E, Valverde R. Long-term care services in the United States: 2013 overview. Hyattsville, MD: National Center for Health Statistics. Available at: http://www.cdc.gov/nchs/data/nsltcp/long_ term_care_services_2013.pdf; 2013. Accessed February 1, 2014. 13. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev; 2012:CD007146. 14. Shaw FE. Prevention of falls in older people with dementia. J Neural Transm 2007;114:1259e1264. 15. Hauer K, Becker C, Lindemann U, Beyer N. Effectiveness of physical training on motor performance and fall prevention in cognitively impaired older persons: A systematic review. Am J Phys Med Rehabil 2006;85:847e857. 16. El-Khoury F, Cassou B, Charles MA, Dargent-Molina P. The effect of fall prevention exercise programmes on fall induced injuries in community dwelling older adults: Systematic review and meta-analysis of randomised controlled trials. BMJ 2013;347:f6234. 17. Silva RB, Eslick GD, Duque G. Exercise for falls and fracture prevention in long term care facilities: A systematic review and meta-analysis. J Am Med Dir Assoc 2013;14:685e689.e2. 18. Oliver D, Connelly JB, Victor CR, et al. Strategies to prevent falls and fractures in hospitals and care homes and effect of cognitive impairment: Systematic review and meta-analyses. BMJ 2007;334:382. 19. Summary of the Updated American Geriatrics Society/British Geriatrics Society clinical practice guideline for prevention of falls in older persons. J Am Geriatr Soc 2011;59:148e157. 20. Hannan MT, Gagnon MM, Aneja J, et al. Optimizing the tracking of falls in studies of older participants: Comparison of quarterly telephone recall with monthly falls calendars in the MOBILIZE Boston Study. Am J Epidemiol 2010; 171:1031e1036. 21. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 510 [updated March 2011]. The Cochrane Collaboration; 2011. Available at: www.cochrane-handbook.org. Accessed September 15, 2014. 22. Toulotte C, Fabre C, Dangremont B, et al. Effects of physical training on the physical capacity of frail, demented patients with a history of falling: A randomised controlled trial. Age Ageing 2003;32:67e73. 23. Pitkala KH, Poysti MM, Laakkonen ML, et al. Effects of the Finnish Alzheimer disease exercise trial (FINALEX): A randomized controlled trial. JAMA Intern Med 2013;173:894e901. 24. Rolland Y, Pillard F, Klapouszczak A, et al. Exercise program for nursing home residents with Alzheimer’s disease: A 1-year randomized, controlled trial. J Am Geriatr Soc 2007;55:158e165. 25. Zieschang T, Schwenk M, Oster P, Hauer K. Sustainability of motor training effects in older people with dementia. J Alzheimers Dis 2013;34:191e202. 26. Lord SR, Castell S, Corcoran J, et al. The effect of group exercise on physical functioning and falls in frail older people living in retirement villages: A randomized, controlled trial. J Am Geriatr Soc 2003;51:1685e1692. 27. Rosendahl E, Gustafson Y, Nordin E, et al. A randomized controlled trial of fall prevention by a high-intensity functional exercise program for older people living in residential care facilities. Aging Clin Exp Res 2008;20:67e75. 28. Moseley AM, Sherrington C, Lord SR, et al. Mobility training after hip fracture: A randomised controlled trial. Age Ageing 2009;38:74e80. 29. Liu-Ambrose T, Donaldson MG, Ahamed Y, et al. Otago home-based strength and balance retraining improves executive functioning in older fallers: A randomized controlled trial. J Am Geriatr Soc 2008;56:1821e1830. 30. Heyn P, Abreu BC, Ottenbacher KJ. The effects of exercise training on elderly persons with cognitive impairment and dementia: A meta-analysis. Arch Phys Med Rehabil 2004;85:1694e1704. 31. Sherrington C, Whitney JC, Lord SR, et al. Effective exercise for the prevention of falls: A systematic review and meta-analysis. J Am Geriatr Soc 2008;56: 2234e2243. 32. Campbell AJ, Robertson MC. Fall prevention: Single or multiple interventions? Single interventions for fall prevention. J Am Geriatr Soc 2013; 61:281e284. 33. Day LM. Fall prevention programs for community-dwelling older people should primarily target a multifactorial intervention rather than exercise as a single intervention. J Am Geriatr Soc 2013;61:284e285.
JAMDA journal homepage: www.jamda.com
Original Study
Efficacy of Physical Exercise in Preventing Falls in Older Adults With Cognitive Impairment: A Systematic Review and Meta-Analysis Wai Chi Chan MRCPsych a, *, Jerry Wing Fai Yeung PhD b, Corine Sau Man Wong MSocSc a, Linda Chiu Wa Lam MD c, Ka Fai Chung MRCPsych a, James Ka Hay Luk FRCP d, Jenny Shun Wah Lee FRCP e, Andrew Chi Kin Law FRCPC a a
Department of Psychiatry, The University of Hong Kong, Hong Kong School of Chinese Medicine, The University of Hong Kong, Hong Kong Department of Psychiatry, The Chinese University of Hong Kong, Hong Kong d Department of Medicine and Geriatrics, Fung Yiu King Hospital, Hong Kong e Department of Medicine and Geriatrics, Tai Po Hospital, Hong Kong b c
a b s t r a c t Keywords: Cognitive disorders dementia exercise accident prevention accidental falls
Objective: Numerous studies have reported the prevention of falls through exercise among cognitively healthy older people. This study aimed to determine whether the current evidence supports that physical exercise is also efficacious in preventing falls in older adults with cognitive impairment. Methods: Two independent reviewers searched MEDLINE; EMBASE; PsycINFO; the Cumulative Index to Nursing & Allied Health Literature; the Cochrane Central Register of Controlled Trials; the Cochrane Bone, Joint, and Muscle Trauma Group Specialized Register; ClinicalTrials.gov; and the UK Clinical Research Network Study Portfolio up to July 2013 without language restriction. We included randomized controlled trials that examined the efficacy of physical exercise in older adults with cognitive impairment. The methodological qualities of the included trials were appraised according to the criteria developed for the Cochrane review of fall prevention trials. The primary outcome measure was the rate ratio of falls. A meta-analysis was performed to estimate the pooled rate ratio and summarize the results of the trials on fall prevention through physical exercise. Results: Seven randomized controlled trials involving 781 participants were included, 4 of which examined solely older people with cognitive impairment. Subgroup data on persons with cognitive impairment were obtained from the other 3 trials that targeted older populations in general. The metaanalysis showed that physical exercise had a significant effect in preventing falls in older adults with cognitive impairment, with a pooled estimate of rate ratio of 0.68 (95% confidence interval 0.51e0.91). Conclusions: The present analysis suggests that physical exercise has a positive effect on preventing falls in older adults with cognitive impairment. Further studies will be required to determine the modality and frequency of exercise that are optimal for the prevention of falls in this population. Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
The global population is aging at an unprecedented pace. In 2012, 23% of the population in the more developed regions and 9% in the less developed regions were aged 60 years or older.1 It is estimated that by 2050 the proportion of older citizens will increase to 32% and 19% in developed and developing countries, respectively.1 Because the prevalence of dementia rises rapidly after the age of 60 years,2 a dramatic change in population demographics will inevitably increase the number of people affected by this disorder, and make dementia a pressing health care issue. The authors declare no conflicts of interest. * Address correspondence to Wai Chi Chan, Department of Psychiatry, 2/F New Clinical Building, Queen Mary Hospital, 102 Pok Fu Lam Road, Hong Kong. E-mail address: [email protected] (W.C. Chan). http://dx.doi.org/10.1016/j.jamda.2014.08.007 1525-8610/Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
One of the major issues in dementia care is the prevention of falls, for which evidence shows that persons with cognitive impairment are at an elevated risk. The annual incidence of falls among older adults with cognitive impairment is about twice that in the elderly who are cognitively healthy.3,4 Even after taking other known risk factors into account, cognitive impairment remains an independent risk factor for falls.4 Studies also have shown that people with cognitive impairment who fall suffer from poorer outcomes than those who are cognitively healthy. For example, people with dementia have a greater likelihood of needing hospital admission for orthopedic crises than people without dementia, 5 and have a significantly greater mortality rate after sustaining fractures.6,7
2
W.C. Chan et al. / JAMDA xxx (2014) 1e6
The prevention of falls in persons with cognitive impairment is challenging. A number of psychological (eg, behavioral symptoms,4 cognitive impairment,8 poor attention and orientation,9 anxiety,9 and depressive symptoms10), physiological (eg, increased postural sway,9 symptomatic orthostatic hypotension,10 impaired simple reaction time,11 and leaning balance11), and medical risk factors (eg, Lewy body disorder,10 antipsychotic use,4 and antidepressant use9) pertinent to older people with cognitive impairment have been identified. The prevention of falls in persons with cognitive impairment is particularly relevant in long-term care settings, because approximately half of all nursing home residents have dementia.12 It is thus essential to understand which interventions are most efficacious in persons with cognitive impairment before effective fallprevention programs can be devised for long-term care facilities. Fall prevention has been studied extensively in the general older population, and strategies including group and home-based exercise programs, home safety interventions, and multifactorial assessment and intervention programs have been found to reduce the rate of falls and the risk of falling.13 However, the efficacy of these interventions, including the more established types, such as physical exercise, among cognitively impaired individuals remains unclear, because most previous studies on fall prevention either excluded subjects with cognitive impairment or did not conduct subgroup analyses related to cognitive impairment status.14 It is thus not surprising that although several systematic reviews have been conducted to examine the effect of physical exercise on the prevention of falls in the elderly,13,15e17 only 115 included randomized controlled trials (RCTs) conducted specifically on cognitively impaired older people. This particular review,15 for which the search was performed in 2004, found that the evidence on the effect of physical exercise in preventing falls among cognitively impaired populations was inconclusive because only 3 RCTs with conflicting results on the number of falls were available. Subsequently, a meta-analysis that investigated the impact of dementia on the efficacy of fall-prevention programs also found no conclusive evidence to suggest an association between the two.18 However, of the 43 studies included in this meta-analysis, which comprised studies published up to 2005, only 1 examined people with dementia exclusively and another performed a subgroup analysis to compare the efficacy of the preventive programs in participants with or without dementia.18 This partly explains why, based on the studies of older adults with known cognitive impairment published up to 2006, the Panel on the Prevention of Falls in Older Persons of the American Geriatrics Society and British Geriatrics Society concluded that there was insufficient evidence to recommend or oppose single or multifactorial interventions for fall prevention in people with dementia.19 We thus conducted an updated systematic review and meta-analysis to summarize the current evidence on the efficacy of physical exercise, one of the more established fall-prevention interventions,13,16,17 in persons with cognitive impairment. Methods Study Selection The studies included in this review were RCTs that compared the efficacy of physical exercise with routine medical care or other controlled activities in preventing falls in older people with cognitive impairment. Physical exercise referred to planned, structured, and repetitive movement to improve or maintain one or more components of physical health.18 Cognitive impairment needed to be suggested by either a standardized cognitive assessment or a diagnosis of dementia established according to accepted criteria. For trials that did not specifically report information on cognitively impaired or
demented older people, original subgroup data were requested from the authors. To better understand the efficacy of physical exercise in the prevention of falls, we excluded trials in which exercise was only part of a multifactorial program. Literature Search We searched MEDLINE (from 1946); EMBASE (from 1980); PsycINFO (from 1967); the Cumulative Index to Nursing & Allied Health Literature (from 1982); the Cochrane Central Register of Controlled Trials (the latest issue); the Cochrane Bone, Joint, and Muscle Trauma Group Specialized Register; ClinicalTrials.gov; and the UK Clinical Research Network Study Portfolio up to July 2013 using the grouped terms (dementia or alzheimer* or lewy body or lewy bodies or cognitive disorders or mild cognitive impairment or memory disorder or memory disorders) and (exercise or physical activit* or physical training or motor activit* or gait training or balance training or bicycling or swim* or gym* or walk* or danc* or yoga or “tai chi” or “Tai Ji”). No language restriction was applied. We also screened reference lists of relevant reviews.13,15e17 Two reviewers were responsible for searching the databases. Any disagreement was resolved by consensus, discussion, and third-party adjudication (JWFY). We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) in reporting this systematic review. Data Extraction and Quality Evaluation One author (WCC) extracted the data independently and the other (JWFY) checked the accuracy of the extracted data. The following variables were extracted from each study: study design, participant characteristics, treatment and control protocols, and outcome parameters. The primary outcome measure was the rate ratio of falls (ie, the ratio of the total number of falls per unit of person time in any 2 groups in each trial). When the rate ratio was not reported, it was calculated from the ratio of the total number of falls divided by the total length of time during which falls were monitored (person-years) in the 2 comparison groups. A secondary outcome measure was the number of participants who sustained fractures. The methodological qualities of the trials included were appraised according to the 7 domains in Cochrane’s tool for assessing risks of bias: random sequence generation, allocation concealment, blinding of the participants, blinding of the assessors, incomplete outcome data, selective reporting, and other sources of bias. In the domain of other sources of bias, we assessed bias in the recall of falls due to unreliable methods of ascertainment,20 using the criteria developed for the Cochrane review of fall-prevention trials.13 The risk of bias in each domain was rated as low, high, or uncertain. Data Synthesis and Analysis The generic inverse variance method in the Review Manager software (RevMan 5.1, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) was used to group the results of the trials and compile the forest plot. The rate ratio (RaR) was calculated for the rate of falls between 2 groups with a 95% confidence interval (CI). Heterogeneity between the trials was assessed using c2 and I2 statistics. We used a fixed effects model in the absence of statistical heterogeneity (heterogeneity test, P .10) and a randomeffects model when heterogeneity was present (heterogeneity test, P < .10). A sensitivity analysis was carried out by excluding 1 study at a time to explore whether the results were influenced by 1 large study or by a study with an extreme result. The possibility of publication bias would have been assessed using funnel if 10 or more trials had been included in the meta-analysis.21
W.C. Chan et al. / JAMDA xxx (2014) 1e6
Results Identification of Studies The search yielded 16,219 potential studies for review. Among these, 5709 were identical citations and 10,450 were excluded because of irrelevance (eg, studies that were not aimed at fall prevention, studies that used multifactorial interventions). The full texts of 60 articles were retrieved for detailed assessment. Nine of these studies targeted the general elderly population, and original subgroup data relating to participants with cognitive impairment were requested from the authors. The authors of 3 of the studies subsequently provided us with the required data. At the end of the assessment, 53 studies were excluded for various reasons (Figure 1), and 7 trials22e28 met the inclusion criteria for further analysis. Description of the Studies Table 1 summarizes the study populations and design of the 7 trials included. Trials 1 to 4 examined the efficacy of physical exercise in preventing falls in cognitively impaired older persons.22e25 Trials 5 to 7 targeted the general elderly population but subgroup data and information on participants with cognitive impairment were provided by the authors.26e28 The total number of participants in these trials was 781, and the sample size of each trial ranged from 20 to
3
210. Participants in 6 of 7 trials had a mean age of 80 years or older.22,24e28 The mean Mini-Mental State Examination (MMSE) score of participants in 5 of the trials was 16 or higher,22,23,25e27 whereas the mean MMSE score in the trial that recruited nursing home residents was 8.8.24 The remaining trial assessed the cognitive function of participants with the Short Portable Mental Status Questionnaire, in which they made an average of 3.78 errors.28 Six trials had 2 parallel arms and 1 had 3 parallel arms. The intervention groups in all trials underwent exercise in a group setting, and 1 of the studies provided an extra arm for exercise in a home-based setting.23 The 2 exercise groups (group-based and home-based) in this trial were combined as a single intervention group for comparison with the control. The essential elements in the physical exercises were balancing and strength training,22e28 flexibility,22,24,26 walking,23e26,28 and coordination.23,25,26,28 The time period, frequency, and duration of the group exercises were identical in 3 of the trials (twice a week, 60 minutes each, and lasted for 12 months).23,24,26 The frequency of intervention in the other 4 trials varied, and the duration of exercise spanned a shorter period of time (3e6 months).22,25,27,28 The number of falls was reported by caregivers or nursing home staff and was used as an outcome assessment in all trials. Two of the trials also collected information on fractures from registries or caregivers.23,24 In addition, physical performance,22e26,28 functional mobility,22,23,25,26,28 and balance ability22,24,26,28 were compared before and after the intervention in some of the trials.
Fig. 1. Flowchart of systematic review.
60 min Group exercise NA 85.9 n ¼ 54 Moseley et al, 2009* 7
16.0 84.2 Rosendahl et al, 2008* n ¼ 100 6
22.8 81.0 n ¼ 141 Lord et al, 2003* 5
21.7 82.1 n ¼ 122 Zieschang et al, 2013 4
AD, Alzheimer disease; DSM, Diagnostic and Statistical Manual of Mental Disorders; MMSE, Mini-Mental State Examination; NA, not available; SPMSQ, Short Portable Mental Status Questionnaire. *Studies where subgroup data were provided by authors.
45 min
Group exercise
Group exercise
5 times every 29 sessions Activities performed while sitting 2 wk (over 3 mo) (eg, watching films, reading, singing) 2/d 16 wk A tailored program of limited weight-bearing exercises
Either flexibility and relaxation program 12 mo
Motor placebo group training 3 mo
120 min (Intervention 2/wk group); 60 min (Control group) 60 min 2/wk Group exercise
12 mo 2/wk 60 min Group exercise
Patients with AD living in nursing homes Patients with mild to moderate dementia; places of abode not specified Older people with MMSE ranged 20e24 living at self-care apartment villages or intermediate-care hostels Older people diagnosed as dementia by DSM criteria living in residential care facilities Older people diagnosed as cognitive impaired or demented by SPMSQ; places of abode not specified 83.0 n ¼ 134 Rolland et al, 2007 3
18.0 78.0 n ¼ 210 Pitkälä et al, 2013 2
8.8
12 mo 2/wk
Routine medical care, oral and written advice on nutrition and exercise methods Routine medical care
16 wk 2/wk
45 min Demented patients living at home or Group exercise institutions with at least 2 previous falls Home-dwelling patients with AD Tailored home-based 60 min living with spousal caregivers exercise, group exercise 16.3 81.4 n ¼ 20 Toulotte et al, 2003 1
Intervention Duration Intervention Frequency Intervention Time Intervention Mode No. of Mean Mean Study Population Participants Age, y MMSE Trial Authors
Table 1 Description of the Randomized Controlled Trials With Regard to Study Population and Interventions
Daily routine
W.C. Chan et al. / JAMDA xxx (2014) 1e6
Control Group
4
In general, the dropout rates from the exercise interventions were low. The short-term attritions (up to 6 months) ranged from 0%22 to 22.6%,25 and the long-term attritions (up to 12 months) were between 11.3%26 and 35.5%.25 Adherence to the exercise programs by persons with dementia was reported in 2 of the studies.23,24 The overall adherence to exercise programs ranged from moderate to high. In the FINALEX study, 92.9% of the participants took part in at least half of the home-based exercise sessions and 78.6% in at least half of the groupbased exercise sessions.23 In another study that examined exercise interventions among nursing home residents, 19.4% completed more than two-thirds of the sessions and 28.4% completed one-third to twothirds of the sessions.24 Quality Assessment According to the Cochrane Criteria Table 2 summarizes the risk of bias of the trials included that was assessed by applying the Cochrane criteria. Sequence generation and allocation concealment were adequately described in 5 trials.23e25,27,28 Randomization in these trials was performed by people who were not involved in the intervention or the assessment. The remaining trials did not mention the sequencegeneration process or the allocation concealment method and hence had an uncertain risk of selection bias. One trial was described as a double-blind controlled trial,25 in which the participants in the control group performed sham activities, including stretching and other nondemanding motor activities under the supervision of trainers. Participants were considered to be blinded to the intervention because they were not informed of the difference in effectiveness of the 2 training regimens. Two trials blinded the assessors to group allocation but the participants were not blinded.24,28 The remaining trials were either described as nonblind controlled trials23 or did not mention whether participants and assessors were blinded.22,26,27 The risk of attrition bias due to incomplete outcome data was rated as high in 1 trial,25 in which the proportion of dropouts compared with the observed risk of falls was sufficient to induce a clinically important bias. All trials had a low risk of reporting bias due to selective reporting. Falls were recorded prospectively by such methods as a fall diary in 5 trials21e24,26 and a questionnaire reporting the number of falls in the past month in 1 trial.25 Potential recall bias due to the measurement of falls at intervals longer than 1 month was observed in 1 of the trials.28 Efficacy Assessment Figure 2 shows the forest plot of the effect of physical exercise on preventing falls in cognitively impaired older people. All trials provided figures for falls during the intervention period. Four of 7 trials showed a significant effect of exercise on reducing the number of falls. The pooled estimated RaR for falls was 0.68 (95% CI 0.51e0.91, I2 ¼ 79%, P ¼ .01, 7 trials). No significant differences in the number of fractures were reported between the intervention and control group during the intervention period for both Trial 2 (9 versus 4)23 and Trial 3 (5 versus 2).24 The pooled estimated RaR for fracture-related to falls was 1.47 (95% CI 0.56e3.81, I2 ¼ 0%, P ¼ .43, 2 trials). Sensitivity Analyses In a sensitivity analysis that excluded 1 study at a time, the pooled rate ratio was lowest (0.63, 95% CI 0.46e0.85, I2 ¼ 74%) when Trial 3, which recruited patients with poorer cognitive function (mean MMSE score, 8.8), was excluded. The pooled rate ratio was still significant (0.73, 95% CI 0.54e1.00, I2 ¼ 78%, P ¼ .049) when
W.C. Chan et al. / JAMDA xxx (2014) 1e6
5
Table 2 The Risks of Bias of the Included Trials Using Cochrane’s Criteria Trial
Authors (Year)
Random Sequence Generation
Allocation Concealment
Blinding of the Participants
Blinding of the Assessors
Incomplete Outcome Data
Selective Reporting
Method of Ascertaining Falls
1 2 3 4 5 6 7
Toulotte et al (2003) Pitkälä et al (2013) Rolland et al (2007) Zieschang et al (2013) Lord et al (2003) Rosendahl et al (2008) Moseley et al (2009)
U L L L U L L
U L L L U L L
U U U L U U U
L H L L H U L
L L L L H L L
L L L L L L L
L L L L L L H
H, high risk of bias; L, low risk of bias; U, uncertain risk.
Trial 5, which recruited patients with relatively better cognitive function (mean MMSE score, 22.8), was excluded. The overall significant result for fall prevention was not affected by any particular study. Discussion To the best of the authors’ knowledge, this is the first metaanalysis to specifically examine the efficacy of physical exercise in preventing falls in cognitively impaired older people. As previous studies have focused on exercise for the general older population, there has been a paucity of RCTs that examined cognitively impaired individuals. The small number of RCTs investigating these populations caused difficulties in the synthesis of the data. The only systematic review of the effect of physical exercise in cognitively impaired populations, which included 3 RCTs reporting on the number of falls, found that the evidence was inconclusive.15 In this up-to-date review of the available literature, we found that physical exercise was associated with a reduced risk of falls in individuals with cognitive impairment. Seven RCTs involving 781 participants were included in the present study. Physical exercise had a significant effect on preventing falls in people with cognitive impairment, with a pooled RaR of 0.68, which was similar to that achieved by group exercise (0.71) and home-based exercise (0.68) in community-dwelling older people.13 Only 2 of the trials reported the number of fractures during the study period but neither showed a significant difference between the intervention and control groups. The exercise interventions in all of the trials consisted of balancing and strength training. Other exercise components included flexibility, walking, coordination training, and exercise involving executive function. It has been well documented that exercise improves mobility, balance, muscle strength, and physical function in the general older population.13 Recent studies have further suggested that exercise may reduce falls by mechanisms other than improving physiological function. In particular, exercise may have an effect by improving cognitive
performance, such as executive function.29 This mechanism may also be applicable to persons with cognitive impairment, because evidence has shown that exercise training not only improves physical, but also cognitive function in people with dementia.30 Five of the studies delivered exercise interventions twice a week, but more intensive training was also adopted by some of the trials included. The trial that provided the most intensive training (twice daily) resulted in one of the greatest reductions in the risk of falls but a similar magnitude was also achieved by less-intensive exercise. The finding that a greater degree of exercise is more effective in reducing falls in cognitively intact individuals31 has not been validated for persons with cognitive impairment. The optimal regimen of exercise interventions, therefore, requires further exploration. The programs in the trials included were implemented under the supervision of trained personnel, and met the special needs of persons with cognitive impairment, who had a poorer ability to understand and learn new information and often required closer supervision by and repeated explanations from the instructors. A higher level of supervision also allowed optimal progression of the exercise program and the achievement of a greater degree of exercise.28 The availability of close supervision probably accounted, at least in part, for the low attrition rates of the studies included. Other measures were also taken to enhance the participation of persons with cognitive impairment in the exercise programs, including those related to treatment settings (eg, provision of the interventions at the participants’ homes23,26), treatment components (eg, versatile exercises,23 the inclusion of “brain-training” exercises,23 the use of music,24 and a patient-centered approach25), treatment schedules (eg, a gradual increase in the intensity of the exercises24), and logistics (eg, the provision of exercises at no cost,26 and the avoidance of competition with other activities26). Whether single interventions are as effective as multifactorial interventions in preventing falls in older people has been a controversial issue.32,33 The present study offered additional evidence to support the theory that single interventions through physical exercise
Fig. 2. Effects of physical exercises on preventing falls in older adults with cognitive impairment.
6
W.C. Chan et al. / JAMDA xxx (2014) 1e6
have a protective effect and prevent falls in cognitively impaired older people. However, in view of the lack of a head-to-head comparison in persons with cognitive impairment, it is still not clear which approach is more effective in these populations. The current review is limited by the small number of RCTs investigating specifically cognitively impaired older people. We have overcome this limitation partially by obtaining subgroup data from the authors of Trials 5 to 7. Despite our efforts to maximize the number of studies included, the current review could analyze data from only 7 RCTs. Failure to obtain subgroup data on cognitively impaired individuals from the 6 studies that targeted the general elderly population constituted another limitation. In addition, the trials included varied with regard to the severity of cognitive impairment, treatment settings, and dropout rates. Due to their heterogeneity, the results of the current meta-analysis should be interpreted with caution. In summary, the present meta-analysis supported that physical exercise effectively reduces the risk of falls in persons with cognitive impairment. However, because the core components of fall-prevention programs depend on the characteristics of the older adults and the treatment settings, a one-size-fits-all solution is probably not available. For example, the complexity of the exercise program has to be adjusted according to the changing cognitive function. Effective fall-prevention exercise therefore needs to be individually tailored in terms of modality, frequency, and intensity. Further RCTs that investigate older adults with various cognitive and physical abilities will assist in the design of fall-prevention programs that meet their diverse needs. Acknowledgments The authors thank Ms Ruth Wong and Mr Matthew Lam for their assistance in the literature search, and Prof Stephen Lord, Dr Erik Rosendahl, and Dr Anne Moseley for their support in providing us with subgroup data of their respective studies. References 1. United Nations, Department of Economic and Social Affairs. World Population Prospects: The 2012 Revision, Highlights and Advance Tables. 2013. Available at: http://esa.un.org/unpd/wpp/Documentation/pdf/WPP2012_HIGHLIGHTS. pdf. Accessed February 1, 2014. 2. Prince M, Bryce R, Albanese E, et al. The global prevalence of dementia: A systematic review and metaanalysis. Alzheimers Dement 2013;9:63e75.e2. 3. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med 1988;319:1701e1707. 4. van Doorn C, Gruber-Baldini AL, Zimmerman S, et al. Dementia as a risk factor for falls and fall injuries among nursing home residents. J Am Geriatr Soc 2003; 51:1213e1218. 5. Toot S, Devine M, Akporobaro A, Orrell M. Causes of hospital admission for people with dementia: A systematic review and meta-analysis. J Am Med Dir Assoc 2013;14:463e470. 6. Diamantopoulos AP, Hoff M, Hochberg M, Haugeberg G. Predictors of shortand long-term mortality in males and females with hip fractureda prospective observational cohort study. PLoS One 2013;8:e78169. 7. Scandol JP, Toson B, Close JC. Fall-related hip fracture hospitalisations and the prevalence of dementia within older people in New South Wales, Australia: An analysis of linked data. Injury 2013;44:776e783. 8. Nazir A, Mueller C, Perkins A, Arling G. Falls and nursing home residents with cognitive impairment: New insights into quality measures and interventions. J Am Med Dir Assoc 2012;13:819.e1e819.e6.
9. Whitney J, Close JC, Jackson SH, Lord SR. Understanding risk of falls in people with cognitive impairment living in residential care. J Am Med Dir Assoc 2012; 13:535e540. 10. Allan LM, Ballard CG, Rowan EN, Kenny RA. Incidence and prediction of falls in dementia: A prospective study in older people. PLoS One 2009;4:e5521. 11. Taylor ME, Lord SR, Delbaere K, et al. Physiological fall risk factors in cognitively impaired older people: A one-year prospective study. Dement Geriatr Cogn Disord 2012;34:181e189. 12. Harris-Kojetin L, Sengupta M, Park-Lee E, Valverde R. Long-term care services in the United States: 2013 overview. Hyattsville, MD: National Center for Health Statistics. Available at: http://www.cdc.gov/nchs/data/nsltcp/long_ term_care_services_2013.pdf; 2013. Accessed February 1, 2014. 13. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev; 2012:CD007146. 14. Shaw FE. Prevention of falls in older people with dementia. J Neural Transm 2007;114:1259e1264. 15. Hauer K, Becker C, Lindemann U, Beyer N. Effectiveness of physical training on motor performance and fall prevention in cognitively impaired older persons: A systematic review. Am J Phys Med Rehabil 2006;85:847e857. 16. El-Khoury F, Cassou B, Charles MA, Dargent-Molina P. The effect of fall prevention exercise programmes on fall induced injuries in community dwelling older adults: Systematic review and meta-analysis of randomised controlled trials. BMJ 2013;347:f6234. 17. Silva RB, Eslick GD, Duque G. Exercise for falls and fracture prevention in long term care facilities: A systematic review and meta-analysis. J Am Med Dir Assoc 2013;14:685e689.e2. 18. Oliver D, Connelly JB, Victor CR, et al. Strategies to prevent falls and fractures in hospitals and care homes and effect of cognitive impairment: Systematic review and meta-analyses. BMJ 2007;334:382. 19. Summary of the Updated American Geriatrics Society/British Geriatrics Society clinical practice guideline for prevention of falls in older persons. J Am Geriatr Soc 2011;59:148e157. 20. Hannan MT, Gagnon MM, Aneja J, et al. Optimizing the tracking of falls in studies of older participants: Comparison of quarterly telephone recall with monthly falls calendars in the MOBILIZE Boston Study. Am J Epidemiol 2010; 171:1031e1036. 21. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 510 [updated March 2011]. The Cochrane Collaboration; 2011. Available at: www.cochrane-handbook.org. Accessed September 15, 2014. 22. Toulotte C, Fabre C, Dangremont B, et al. Effects of physical training on the physical capacity of frail, demented patients with a history of falling: A randomised controlled trial. Age Ageing 2003;32:67e73. 23. Pitkala KH, Poysti MM, Laakkonen ML, et al. Effects of the Finnish Alzheimer disease exercise trial (FINALEX): A randomized controlled trial. JAMA Intern Med 2013;173:894e901. 24. Rolland Y, Pillard F, Klapouszczak A, et al. Exercise program for nursing home residents with Alzheimer’s disease: A 1-year randomized, controlled trial. J Am Geriatr Soc 2007;55:158e165. 25. Zieschang T, Schwenk M, Oster P, Hauer K. Sustainability of motor training effects in older people with dementia. J Alzheimers Dis 2013;34:191e202. 26. Lord SR, Castell S, Corcoran J, et al. The effect of group exercise on physical functioning and falls in frail older people living in retirement villages: A randomized, controlled trial. J Am Geriatr Soc 2003;51:1685e1692. 27. Rosendahl E, Gustafson Y, Nordin E, et al. A randomized controlled trial of fall prevention by a high-intensity functional exercise program for older people living in residential care facilities. Aging Clin Exp Res 2008;20:67e75. 28. Moseley AM, Sherrington C, Lord SR, et al. Mobility training after hip fracture: A randomised controlled trial. Age Ageing 2009;38:74e80. 29. Liu-Ambrose T, Donaldson MG, Ahamed Y, et al. Otago home-based strength and balance retraining improves executive functioning in older fallers: A randomized controlled trial. J Am Geriatr Soc 2008;56:1821e1830. 30. Heyn P, Abreu BC, Ottenbacher KJ. The effects of exercise training on elderly persons with cognitive impairment and dementia: A meta-analysis. Arch Phys Med Rehabil 2004;85:1694e1704. 31. Sherrington C, Whitney JC, Lord SR, et al. Effective exercise for the prevention of falls: A systematic review and meta-analysis. J Am Geriatr Soc 2008;56: 2234e2243. 32. Campbell AJ, Robertson MC. Fall prevention: Single or multiple interventions? Single interventions for fall prevention. J Am Geriatr Soc 2013; 61:281e284. 33. Day LM. Fall prevention programs for community-dwelling older people should primarily target a multifactorial intervention rather than exercise as a single intervention. J Am Geriatr Soc 2013;61:284e285.
