Secondary Logo

Journal Logo

Brief Report

Multisystem Balance Training Reduces Injurious Fall Risk in Parkinson Disease

A Randomized Trial

Wong-Yu, Irene S.K. PhD; Mak, Margaret K.Y. PhD

Author Information
American Journal of Physical Medicine & Rehabilitation: March 2019 - Volume 98 - Issue 3 - p 239-244
doi: 10.1097/PHM.0000000000001035


Fall-related injuries are common in people with Parkinson disease (PD). In a long-term study, more than 30% of participants reported serious injuries such as fractures and intracranial hemorrhages.1 Increased lengths of hospitalization due to hip fracture were associated with higher healthcare costs and poorer mobility after discharge in people with PD with hip fracture.2 Although axial motor signs are less responsive to dopaminergic medications and deep brain stimulation,3,4 exercise training has been proven to improve balance and reduce falls for people with PD.5–7 Fall prevention in PD commonly targets recurrent fallers but overlooks those with no fall history.8–10 Alarmingly, both PD nonfallers and single fallers (those who had one fall) have worse balance performance than healthy subjects,11 and 21% of PD nonfallers sustain their first fall during the 3-mo follow-up period.12 There is a need for timely treatment to prevent falls at the early stages of PD. Although well-designed exercise programs are effective in reducing falls, there is little evidence that these programs can also prevent injuries caused by falls. Only one study found that a 6-wk home-based exercise program showed the trend of lowering the rates of injurious falls at 6 mos for recurrent fallers.8 Moreover, a recent meta-analysis study reported the effect of exercise on fall risk, not injurious fall risk.7 Therefore, it is important to examine whether exercise training can reduce injurious falls in people with PD, especially for PD nonfallers and single fallers.

For older people, many risk factors for falls and fall-induced injuries seem to be similar and remediable by well-designed balance exercise programs.13 In PD, the fall risk factors include impairment in all six balance domains (biomechanical constraints, verticality and stability limits, anticipatory postural adjustment, reactive postural response, sensory orientation, and dynamic gait stability) as evaluated by the Balance Evaluation Systems Test (BESTest).14 In addition to balance impairment, people with PD show greater deficits in dual-task gait and functional mobility performance.15 This difficulty could be explained by a loss of automatic motor control due to basal ganglia dysfunction and the reliance on cognitive resources to maintain gait stability.16 Compromised dual-task performance has been shown to be associated with increased risk of falling and reduced functional mobility.17 Thus, both balance and dual-task impairments are important fall risk factors that need to be considered in providing effective fall prevention interventions for people with PD.

A published study18 has reported that multisystem balance exercises designed to target the six balance domains19 comprehensively could enhance balance and dual-task functional performance for up to 12 mos for PD nonfallers and single fallers. On the basis of these positive findings, the gains in balance and dual-task performance may also be beneficial in the long term for preventing fall-related injuries. Using a secondary analysis of the data derived from the published trial,18 the primary aim of the current study is to assess the effectiveness of this program in reducing the risk of injurious falls (ratio of noninjurious fallers to injurious fallers) among PD nonfallers and single fallers. The secondary aim is to determine whether the program is effective in reducing the balance and dual-task–related fall risk factors. Our hypotheses assumed that multisystem balance training could reduce the risk of injurious falls for up to 12 mos after training and could lower the risk of falls associated with balance and dual-task functional performance in PD.


Study Design and Participants

The data analyzed in this study were derived from a published single-blinded randomized controlled trial that was designed to evaluate the effectiveness of a multidimensional balance program in improving balance and gait performance in people with PD.18 The eligible participants included individuals with PD diagnosed by neurologists, who were at least 30 yrs of age, had a maximum of one fall in the previous 6 mos, and could walk 30 meters independently. Individuals were excluded if they had any other significant neurological or musculoskeletal disorders or showed cognitive deficits on the Mini-Mental State Examination20 (<24). Before the baseline assessment (Pre), each participant was randomly assigned to either an experimental (EXP, n = 41) or control (CON, n = 43) group using the web-based computer software Research Randomizer.21 Two independent personnel recruited participants from the local PD Association and movement disorder clinics and conducted the randomization procedures. The group assignment was concealed and masked from the principle assessor.

Standard Protocol Approval, Registration, and Participant Consent

The ethics committees of The Hong Kong Polytechnic University approved the use of human participants for this study. Written informed consent was obtained from all participants. The trial was registered on (NCT01799681). The study followed all CONSORT guidelines22 and reported the required information accordingly (see Checklist, Supplemental Digital Content 1,

Outcomes and Sample Size Calculation

The assessor examined each participant in the university's laboratory to collect demographic data and self-reported number of falls and injurious falls for the previous 6 mos at baseline (Pre). The participants were then given diaries to document all information about any falls and fall-related injuries after the commencement of treatment. The total number of fallers was recorded to identify the noninjurious fallers or injurious fallers during the interventions and in the following 12 mos. Injurious falls were defined as falls resulting in moderate or serious symptoms of injuries and classified according to the types of injuries (laceration, abrasion, sprain, contusion, fracture) and the healthcare services required (medical attention, emergency visit, hospitalization).23 The primary outcome of injurious fall risk was determined by the ratio of noninjurious fallers to injurious fallers after completion of the 8-wk treatment (Post) and at 12-mo follow-up (FU12m). A lower ratio indicated a lower injurious fall risk.

The two fall risk outcomes associated with balance in each group, including the ratios of participants in the low fall risk cohorts to participants in the high fall risk cohorts, were determined as the secondary outcomes. As described in the published study,18 dynamic balance performance and dual-task functional performance outcomes were measured using the BESTest total scores and dual-task timed-up-and-go (DTUG) times, respectively, during patients' on-medication state at Pre, Post, and FU12m. It was found that at FU12m, the EXP group had significantly greater gains than the CON group in both outcomes. To examine whether the multisystem balance training could modify fall risks, the participants were classified into either cohort on the basis of the results of two studies examining the utility of BESTest total scores24 and DTUG times25 for identifying fall risk in PD. The cut-off values for PD fallers was 84% for BESTest total score24 (a higher score indicated better balance) and 14.7 secs for DTUG time25 (a shorter time indicated better dual-task functional ability). Participants who had BESTest total scores lower (higher) than 84% or DTUG times longer (shorter) than 14.7 s were classified into the high (low) fall risk cohort. The ratios of the number of participants in the high-risk cohort to the number of participants in the low-risk cohort for each outcome were compared between the EXP and CON groups. A lower percentage of participants in the high-risk cohort indicate a fall risk reduction in that particular group. Lastly, the characteristics of injurious falls were compared between the EXP and CON groups. These characteristics included the types of activities that led to falls, the perceived causes of falls, the ratio of injurious falls to all falls, and the body parts injured. A sample size of 65 was already estimated in the published study18 on the basis of the effect sizes obtained in a pilot trial for BESTest total score and DTUG time.


Physiotherapists provided supervision and treatment to participants in groups of six. The EXP group completed 4 wks each of indoor and outdoor balance training for 3 hrs/wk. The indoor program included flexibility and strength training, balance dance, modified Wing Chun with perturbation-based and upper limb training, and square-stepping exercise. The indoor phase training first equipped the participants with essential balance skills, such as reaching, weight shifting, walking, and stepping responses, in a safe and controlled setting. In the outdoor phase, the participants continued with the learned exercises and progressed to practice advanced fall-prone balance tasks, perturbation-based and dual-task gait, and functional training under a more challenging environment. The exercises performed in both phases targeted all the key balance components under the six domains. The CON group practiced upper limb activities at the same treatment dosage in the sitting position.

Physiotherapists instructed all the participants to practice their exercises at home for 6 mos (for 3 hrs/wk) after completion of the 8 wks of training. The home exercise programs consisted of same types of intervention as the respective 8-wk protocols, with some modification in time allocation for each component. The protocol details can be found in the Appendix (Supplemental Digital Content 2,

Statistical Analysis

In this study, data analyses were performed using the SPSS Statistics software (v. 18.0) and were based on the injurious falls records collected and the two fall risk factors determined for the same group of participants in the published study.18 χ2 tests were used for ratio comparisons because the sample size was greater than 30. Fisher exact tests were used if the expected count was less than 5. The significance levels of both tests were set at P < 0.05. The outcome measures included were the following: (1) injurious fall risk (i.e., the ratio of noninjurious fallers to injurious fallers); (2) two fall risk factors (i.e., the ratios of participants in low fall risk cohorts to participants in the high fall risk cohorts as determined either by the cut-off BESTest total scores or the DTUG times for PD fallers); and (3) the characteristics of injurious falls.


Eighty-three people with PD diagnosed by neurologists using the United Kingdom Brain Bank Criteria26 participated from August 2011 to November 2013, and 80 participants were eligible for intention-to-treat analysis (EXP, n = 41; CON, n = 39). The CONSORT diagram and baseline comparisons of the two groups have been reported.18 There was no significant difference in any of their demographic and fall characteristics. Their mean ± SD age was 61.0 ± 9.0 yrs, and the mean ± SD Hoehn and Yahr staging was 2.5 ± 0.3. The ratios of nonfallers to fallers in the EXP and CON groups for the previous 6 mos were 32:9 and 36:3, respectively. The mean ± SD number of falls for all participants in the previous 6 mos was 0.2 ± 0.8, and there were no fall-related injuries. No adverse events or falls were documented during the physiotherapist-led training or home exercises.

During the 12 mos after the 8-wk training, a total of 17 EXP and 12 CON participants experienced either noninjurious or injurious falls. The proportion of fallers was 41.5% in the EXP group and 30.8% in the CON group. χ2 tests showed no significant between-group difference for the number of fallers; hence, the fall risk was similar for both groups. Five EXP participants (29% in EXP group), as compared with nine CON participants (75% in CON group), sustained injurious falls, with fewer injurious fallers in the EXP group (P = 0.025, risk ratio = 0.392, 95% confidence interval = 1.35–38.33) (Table 1). Hence, the EXP group had a lower risk of injurious falls than the CON group.

Between-group comparisons of fall risk–related and injurious falls outcomes in posttraining and 1-yr follow-up (N = 80)

Regarding the fall risk factors, there was no significant difference between the numbers of participants in the low and high fall risk cohorts as determined by the BESTest total scores and DTUG times at baseline (Table 1). When compared with the CON group at Post, fewer EXP participants had fall risk factors as determined by a BESTest cut-off score of under 84 (32% vs. 67%, P < 0.05) (Table 1) and a DTUG cut-off time of over 14.7 secs (32% vs. 54%, P = 0.045) (Table 1). At FU12m, marginally fewer participants in the EXP group than in the CON group had a BESTest score of less than 84 (51% vs. 72%, P = 0.059) (Table 1). A higher number of EXP group participants (n = 20) than CON group participants (n = 8) used their upper limbs to make the initial landing contact when they fell (48% vs. 20%, P < 0.05).

In addition, no EXP participant experienced multiple injuries (i.e., affecting more than one body part), whereas 15% of the CON participants had sustained such injuries at FU12m (P = 0.055) (Table 2). With regard to the injurious falls, there were no between-group differences in the activities that led to falls, the perceived causes of falls, and the types of healthcare services received after the falls (Table 2). The common causes of injurious falls in both groups were loss of balance and moving hurriedly, whereas only CON group participants reported that falls were caused by external perturbation, loss of concentration, and freezing of gait.

Between-group comparisons of the injurious fall characteristics at FU12m


People with PD were found to have higher rates of fall-related hospital admissions and injuries and longer lengths of hospitalization than people without PD.27 Therefore, it is important to prevent both falls and fall-induced injuries, as well as to minimize the occurrence of serious complications and subsequent deteriorations. The causes of falls in people with PD are multifactorial, with fall history as the strongest nonmodifiable risk factor for recurrent falls.28 Therefore, it is crucial to prevent first falls in PD nonfallers and recurrent falls in single fallers using effective strategies to target the potentially modifiable fall risk factors such as reduced lower limb strength, impaired anticipatory and reactive balance control, and deficits in dual tasking in PD.17,29 Evidence from recently published reviews on randomized controlled trials support the implementation of balance training to reduce fall rates by approximately 60% to 80%; these effects could be maintained for up to 12 mos after treatment ends.6,7 The long-term reduction in fall rate could be attributed to improved balance performance and balance confidence30,31 and the alleviation of motor symptoms.32 In addition to reducing the fall rate of people with PD, lowering the risk of fall-related injuries is essential. Recently, the US Preventive Services Task Force reported that exercise is associated with significant reductions in the risk of experiencing falls and injurious falls in older people.33 To mitigate injurious falls among people with PD, it is important to elucidate the gap between postural imbalance and the underlying injurious fall risk. However, injurious falls data have seldom been reported and have not been analyzed in clinical trials for PD.9

To the best of our knowledge, the present study is the first to focus on PD nonfallers and single fallers and demonstrate that multisystem balance training could potentially reduce their injurious fall risk by 60% in the long term. Our results further indicated that there were fewer participants in the EXP group with lower balance and dual-task–related fall risks at post 8-wk training and a lower balance-related fall risk at FU12m. In the original study, the EXP group was shown to have better balance and dual-task functional performance up to FU12m.18 The BESTest scores and DTUG times are two discrete sensitive outcomes to predict falls among people with PD.24,25 The gains in BESTest score reflect an overall improvement in balance performance. A reduction in DTUG time indicates improved dual-task ability, with less attentional demand and better automaticity to safely perform fall-prone activities. The blend of indoor and outdoor training followed by on-going home exercises enabled the participants to consolidate learned balance skills and to transfer them to daily practice, thus promoting long-term benefits in overall balance performance and modifying important fall risk factors.

It is intriguing that at FU12m, a lower fall risk was associated with balance but not with dual-task functional performance among the EXP participants. It is necessary for exercise interventions to target not only functional activities but also balance abilities in other domains, such as anticipatory postural strategies, attentional capacity, and reactions to perturbations.29,34 Therefore, improved overall balance performance, together with a lower fall risk associated with balance, may contribute to reducing the underlying risks of injurious falls in the long term.

Interestingly, a higher percentage of the EXP group used their outstretched upper limbs to cushion their falls, suggesting that one of the balance-enhancing exercises, modified Wing Chun, could have facilitated protective arm reactions. These learned upper limb rescue strategies might play a role in dissipating the impact energy of any fall across different body parts, thus reducing the likelihood of injury.35

Limitations of the present study do exist. Although the improved balance performance might have reduced injurious falls, it was not associated with a decreased likelihood of falling in the EXP participants. One speculation is that impulsiveness36 and overconfidence may present in certain EXP participants after exercise training. These negative behaviors may exert hazardous effects on their postural stability and result in falls during untrained fall-prone and high-risk activities. Thus, adding psychological or fall risk educational components for fall prevention in PD could be useful, and future research should address the effects of such innovative interventions.29 Despite the EXP exercise program being shown to have significantly decreased the number of injurious fallers at FU12m, the sample size of fallers was small. Future investigations with a larger sample size are necessary to determine the associations between balance impairment and injurious fall rate and fall risk and whether the reduction of balance-related fall risk factors could be responsible for mitigating injurious falls in people with PD.


The main findings of the present study suggest that this multisystem balance training program is a safe, feasible, and potentially effective intervention for reducing injurious fall risk among people with PD for at least 12 mos’ posttraining and that it lowers the known fall risks associated with balance.


The authors thank all the people with PD who participated in this study. The authors also thank the Hong Kong Parkinson’s Disease Foundation and Shun Hing Education and Charity Fund for providing financial support for this study.


1. Auyeung M, Tsoi TH, Mok V, et al.: Ten year survival and outcomes in a prospective cohort of new onset Chinese Parkinson's disease patients. J Neurol Neurosurg Psychiatry 2012;83:607–11
2. Walker RW, Chaplin A, Hancock RL, et al.: Hip fractures in people with idiopathic Parkinson's disease: incidence and outcomes. Mov Disord 2013;28:334–40
3. Curtze C, Nutt JG, Carlson-Kuhta P, et al.: Levodopa is a double-edged sword for balance and gait in people with Parkinson's disease. Mov Disord 2015;30:1361–70
4. Fasano A, Aquino CC, Krauss JK, et al.: Axial disability and deep brain stimulation in patients with Parkinson disease. Nat Rev Neurol 2015;11:98–110
5. Allen NE, Sherrington C, Paul SS, et al.: Balance and falls in Parkinson's disease: a meta-analysis of the effect of exercise and motor training. Mov Disord 2011;26:1605–15
6. Mak MK, Wong-Yu IS, Shen X, et al.: Long-term effects of exercise and physical therapy in people with Parkinson disease. Nat Rev Neurol 2017;13:689
7. Shen X, Wong-Yu IS, Mak MK: Effects of exercise on falls, balance, and gait ability in Parkinson's disease: a meta-analysis. Neurorehabil Neural Repair 2016;30:512–27
8. Ashburn A, Fazakarley L, Ballinger C, et al.: A randomised controlled trial of a home based exercise programme to reduce the risk of falling among people with Parkinson's disease. J Neurol Neurosurg Psychiatry 2007;78:678–84
9. Canning CG, Sherrington C, Lord SR, et al.: Exercise for falls prevention in Parkinson disease: a randomized controlled trial. Neurology 2015;84:304–12
10. Goodwin VA, Richards SH, Henley W, et al.: An exercise intervention to prevent falls in people with Parkinson's disease: a pragmatic randomised controlled trial. J Neurol Neurosurg Psychiatry 2011;82:1232–8
11. Mak MK, Pang MY: Parkinsonian single fallers versus recurrent fallers: different fall characteristics and clinical features. J Neurol 2010;257:1543–51
12. Pickering RM, Grimbergen YA, Rigney U, et al.: A meta-analysis of six prospective studies of falling in Parkinson's disease. Mov Disord 2007;22:1892–900
13. El-Khoury F, Cassou B, Charles MA, et al.: 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
14. Horak FB, Wrisley DM, Frank J: The Balance Evaluation Systems Test (BESTest) to differentiate balance deficits. Phys Ther 2009;89:484–98
15. Campbell CM, Rowse JL, Ciol MA, et al.: The effect of cognitive demand on Timed Up and Go performance in older adults with and without Parkinson disease. J Neurol Phys Ther 2003;27:2–7
16. D'Angelo MC, Milliken B, Jiménez L, et al.: Implementing flexibility in automaticity: evidence from context-specific implicit sequence learning. Conscious Cogn 2013;22:64–81
17. Plotnik M, Giladi N, Dagan Y, et al.: Postural instability and fall risk in Parkinson's disease: impaired dual tasking, pacing, and bilateral coordination of gait during the "ON" medication state. Exp Brain Res 2011;210:529–38
18. Wong-Yu IS, Mak MK: Multi-dimensional balance training programme improves balance and gait performance in people with Parkinson's disease: a pragmatic randomized controlled trial with 12-month follow-up. Parkinsonism Relat Disord 2015;21:615–21
19. King LA, Horak FB: Delaying mobility disability in people with Parkinson disease using a sensorimotor agility exercise program. Phys Ther 2009;89:384–93
20. Folstein MF, Folstein SE, McHugh PR: “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98
21. Urbaniak GC, Plous S: Research Randomizer (version 3.0). Available at: Accessed June 10, 2011
22. Schulz KF, Altman DG, Moher D, et al.: CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. BMC Med 2010;8:18
23. Schwenk M, Lauenroth A, Stock C, et al.: Definitions and methods of measuring and reporting on injurious falls in randomised controlled fall prevention trials: a systematic review. BMC Med Res Methodol 2012;12:50–63
24. Leddy AL, Crowner BE, Earhart GM: Utility of the mini-BESTest, BESTest, and BESTest sections for balance assessments in individuals with Parkinson disease. J NeurolPhys Ther 2011;35:90–7
25. Vance RC, Healy DG, Galvin R, et al.: Dual tasking with the timed “up & go” test improves detection of risk of falls in people with Parkinson disease. Phys Ther 2015;95:95–102
26. Hughes AJ, Daniel SE, Kilford L, et al.: Accuracy of clinical diagnosis of idiopathic Parkinson's disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181–4
27. Paul SS, Harvey L, Canning CG, et al.: Fall-related hospitalization in people with Parkinson's disease. Eur J Neurol 2017;24:523–9
28. Allen NE, Schwarzel AK, Canning CG: Recurrent falls in Parkinson's disease: a systematic review. Parkinsons Dis 2013;2013:906274
29. Canning CG, Paul SS, Nieuwboer A: Prevention of falls in Parkinson's disease: a review of fall risk factors and the role of physical interventions. Neurodegener Dis Manag 2014;4:203–21
30. Shen X, Mak MK: Balance and gait training with augmented feedback improves balance confidence in people with Parkinson's disease: a randomized controlled trial. Neurorehabil Neural Repair 2014;28:524–35
31. Shen X, Mak MK: Technology-assisted balance and gait training reduces falls in patients with Parkinson's disease: a randomized controlled trial with 12-month follow-up. Neurorehabil Neural Repair 2015;29:103–11
32. Morris ME, Menz HB, McGinley JL, et al.: A randomized controlled trial to reduce falls in people with Parkinson's disease. Neurorehabil Neural Repair 2015;29:777–85
33. Guirguis-Blake JM, Michael YL, Perdue LA, et al.: Interventions to prevent falls in older adults: updated evidence report and systematic review for the US Preventive Services Task Force. JAMA 2018;319:1705–16
34. Mansfield A, Peters AL, Liu BA, et al.: Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys Ther 2010;90:476–91
35. Sran MM, Stotz PJ, Normandin SC, et al.: Age differences in energy absorption in the upper extremity during a descent movement: implications for arresting a fall. J Gerontol A Biol Sci Med Sci 2010;65:312–7
36. Smulders K, Esselink RA, Cools R, et al.: Trait impulsivity is associated with the risk of falls in Parkinson's disease. PLoS One 2014;9:e91190

Parkinson Disease; Exercise Therapy; Accidental Falls; Risk Factors

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.