- Email: [email protected]
Postural stability and history of falls in cognitively able older adults: The Canton Ticino study
Gait & Posture 36 (2012) 662–666
Contents lists available at SciVerse ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
SIAMOC Best Clinical Paper Award 2011
Postural stability and history of falls in cognitively able older adults: The Canton Ticino study§ Andrea Merlo a,*, Damiano Zemp b, Enrica Zanda b, Sabrina Rocchi c, Fabiano Meroni b, Mauro Tettamanti d, Angela Recchia d, Ugo Lucca d, Pierluigi Quadri b a
LAM – Laboratorio Analisi Movimento, AUSL Reggio Emilia, SOC Neuroriabilitazione, Ospedale di Correggio, Reggio Emilia, Italy Servizio Sottocenerino di Geriatria, Ospedali regionali of Lugano and Mendrisio,Switzerland c Clinica Hildebrand, Centro di Riabilitazione, Brissago, Switzerland d Laboratory of Geriatric Neuropsychiatry, Istituto di Ricerche Farmacologiche ‘‘Mario Negri’’, Milano, Italy b
A R T I C L E I N F O
A B S T R A C T
Keywords: Aging Balance Falls Center of pressure Postural control
Falls are common events in the elderly and represent the main risk factor for fractures and other injuries. Strategies for fall prevention rely on the multifactorial assessment of the risk of falling. The contribution of instrumented balance assessment to the prediction of falls remains unclear in the literature. In this study, we analyzed the association between the fall-history of a wide sample of older people without dementia and the values of a set of posturographic parameters acquired in different visual, proprioceptive and mental conditions. A consecutive sample of 130 cognitively able elderly subjects, age 70 years, was analyzed. Based on their fall-history in the last year, subjects were categorized into non-fallers (NF), fallers (F) and recurrent fallers (RF > 2 falls). Each subject was assessed by measurements of cognition and functional ability. Static posturography tests were performed in five conditions: with eyes open/close (EO/EC) on a firm/compliant (FS/CS) surface and while performing a cognitive task. The center of pressure (COP) mean position referred to the mid-point of the heels, area of the 95% confidence ellipse, sway mean velocities and RMS displacements in the antero-posterior (AP) and medio-lateral (ML) directions were computed and their association with the fall-history was assessed. The mean position of the COP in the AP direction and the confidence ellipse area were associated with the fall-history in the EOFS, ECFS and EOCS conditions (P < 0.05). RMS displacements were also associated with the fall-history in the EOCS condition (P < 0.05). Significant group differences (P < 0.05) were found in the EOCS conditions, which greatly enhanced the differences among NF, F and RF. The ability to control balance while standing with eyes open on a compliant surface showed a high degree of association with the fall-history of older people with no or mild cognitive impairment. ß 2012 Elsevier B.V. All rights reserved.
1. Introduction Falls are common in the elderly and frequently result in injury, disability or hospitalization [1]. Around 30% of people aged 65 or older living in the community and more than 50% of those living in residential care facilities or nursing homes fall every year and about half of those who fall do so repeatedly [2]. Effective fall prevention has the potential to reduce serious fallrelated injuries [3] and related costs. Prevention of falls, however,
§ This paper was not selected through the usual peer review process but through a SIAMOC Reading Committee. * Corresponding author at: LAM – Laboratorio Analisi Movimento, Azienda USL di Reggio Emilia, Via Mandriolo Superiore 11, 42015 Correggio, RE, Italy. Tel.: +39 0522 630391; fax: +39 0522 630332. E-mail address: [email protected] (A. Merlo).
0966-6362/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gaitpost.2012.06.016
is not easy, because they are complex events caused by a combination of intrinsic impairments and disabilities with or without accompanying environmental hazards [1,2]. Thus, multifactorial fall risk assessment has been recommended by recent guidelines for the prevention of falls in older individuals [3,4]. The most commonly used mobility assessments are selfreported capacity in movements and functional rating scales, such as the Tinetti Performance Oriented Mobility Assessment (POMA) [5]. Objective performance indexes are also commonly used [6]. Along with these clinical and functional evaluations, the instrumental assessment of balance by force-platforms is a widely applied tool in assessing standing balance with several parameters computed from the displacement of the Center of Pressure (COP) [7]. Posturographic parameters have been investigated under different testing conditions, such as with open and closed eyes, on firm and compliant surfaces and while performing cognitive tasks [8–10].
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
The role of the instrumented assessment of balance in clinical decision-making has been recently reviewed in [11] and the contribution of COP related measures in the detection of the risk of falling has been recently reported [6,12,13]. However, other prospective studies indicated an overall marginal predictive value for COP related measures [14]. The different levels of cognitive impairment of the samples analyzed in these studies could act as a major confounding factor [1,3], as its effect on both balance and posturographic parameters is known [10]. In this study, we analyzed the association between the fallhistory of a wide sample of older people without dementia and the values of a set of posturographic parameters acquired in different visual, proprioceptive and cognitive conditions. 2. Methods 2.1. Sample The study population consisted of 130 cognitively able individuals (clinical dementia rating, CDR 0.5), 52 M and 78 F, mean age 79 6 years and age range 70–91 years, seen consecutively at the Memory Clinic of the Regional Hospitals of Mendrisio and Lugano, Switzerland. This sample is a sub-group of the Canton Ticino Study [15], an epidemiological database using the CERAD clinical and neuropsychological battery [16] that has included more than 1400 consecutive ambulatory elderly individuals since 1998. 2.2. Clinical evaluation A standard assessment was delivered to all patients, according to the CERAD protocol, including: clinical history (event, yes/no); measurement of cognition with the Mini Mental State Examination (MMSE); a complete battery of neuropsychological tests; measurement of functional ability, such as basic (BADL) and instrumental activities of daily living (IADL) scores and Gait and Balance performance on Tinetti’s Performance Oriented Mobility Assessment. Details on these scales can be found in [16]. 2.3. Fall history History of falls in the previous year was collected within the CERAD assessment. Subjects were categorized as non-fallers (NF), fallers (F, one or two falls) and recurrent fallers (RF, more than two falls).
663
2.5. Experimental procedure Each subject completed a sequence of five 30-s tasks, as follows: standing with eyes open on a firm surface (EOFS); standing with eyes closed on a firm surface (ECFS); standing with eyes open on a firm surface while performing a cognitive task (i.e. counting backward by steps of seven) referred to as a dual task (DT); standing with eyes open on a compliant surface (EOCS) and standing with eyes closed on a compliant surface (ECCS). The compliant surface was obtained by placing a (40 cm 40 cm 3 cm) viscoelastic gel pillow (Elastil II, Laboratores Escarius, La Courneuve, France) over the force plate. The subject sat on a chair while the assessor placed the pillow on the force plate. Subjects were asked to stand barefoot on the force platform in a relaxed posture with arms at their sides. Before each trial, foot position was standardized by means of a custom made removable device. The rear part of this device was larger than the force plate. It was inserted laterally to the force plate into two holes in the floor in a known, fixed position, thus ensuring that heels (and feet) were placed in a known, repeatable position with respect to the origin of the force plate. Moreover, the imposed fixed position of the medial margin of the feet led to minimum variations in the area of the base of support. The distance between the center of the heels ranged from 14 to 16 cm, depending on the subject’s foot size. The angle between each foot and the AP direction was 10 degrees, which corresponds to the natural average of toeing out during stance and gait [18]. The use of the positioning device allowed the placement of the heels in a known position over the force plate and, consequently, the expression of the COP coordinates with respect to the midpoint between the heels. A visual target (a colored circle with a diameter of 6 cm) was placed at a distance of 1.5 m at eyelevel. A single operator performed all acquisitions in both hospitals throughout the years. 2.6. Statistical analysis For each clinical, functional and posturographic parameter, we tested the null hypothesis that the median value was not affected by the fall-history in the last year, by means of a Kruskal–Wallis test. Between-group comparisons were carried out by using the Mann–Whitney test. The non-parametric approach was chosen, as it does not require normality of data and variance equality amongst groups. Comparisons between proportions were carried out by means of the Fisher Exact test. All analyses were conducted with statistical analysis software (SPSS 13.0) using a 5% level of significance.
2.4. Instrumentation 3. Results Posturographic data were acquired by means of one piezoelectric force plate (Kistler 9281C, Kistler, Winterthur, Switzerland) in Mendrisio and one strain-gage based force plate (AMTI OR-6, Watertown, MA, USA) in Lugano. In both cases, the acquisition was performed by using Bioware software (Kistler, Winterthur, Switzerland) at the sampling frequency of 100 Hz. The calibration of both force plates was checked before the beginning of the study as in [17]. Data were off-line filtered at 20 Hz by means of a zero-lag Butterwoth filter and the global COPrelated parameters were computed, as in [7], by means of a custom made Matlab software. Among the parameters that can be derived from static posturography tests, we focused on those that emerged from a systematic review on their use as predictors of falls among older people [14], including the antero-posterior (AP) and the medio-lateral (ML) mean COP position referred to the mid-point of the heels, the sway mean velocity, the AP and ML mean velocity, the AP and ML RMS displacement and the area of 95% confidence ellipse.
Participants were separated into fall-status entry groups: 67 non-fallers, 45 fallers and 18 recurrent fallers. Demographic, clinical and functional data of the sample are reported in Table 1. No age differences were found among non-fallers, fallers and recurrent fallers (Kruskal–Wallis, P = 0.29). Functional scales were affected by the fall-history (Kruskal–Wallis, P < 0.01). In detail, the median value of both POMA and IADL scores from F and RF subjects were significantly different than those from NF (Mann–Whitney, P < 0.05). No significant differences were found between fallers and recurrent fallers. Values of posturographic parameters (mean SD) are reported in Table 2 for all tasks. The presence of a statistically significant effect of fall-history on posturographic parameters (Kruskal–Wallis, P < 0.05) is outlined by a light-grey background of the cell. Between group statistically significant differences are indicated by symbols. In the EOFS task, the mean AP COP position and the area of the confidence ellipse were associated with the history of falls. The
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
664
Table 1 Demographic, clinical and functional characteristics of the sample. Descriptive measure
Non fallers (NF) (N = 67)
1–2 times fallers (F) (N = 45)
Recurrent fallers (RF) (N = 18)
Group effect P
Age (y) Sex (% men) Height (m) Weight (kg) Body mass index (kg/m2) Clinical dementia rating CDR = 0 (%); CDR = 0.5 (%) Mini mental state evaluation IADL (percentage of dependence) POMA, gait score (median; 25th–75th percentile) POMA, balance score (median; 25th–75th percentile)
79 5 43% 1.62 0.10 69 14 26.3 4.9 25%; 75% 26 3 8 16 11; 11–12 14; 14–15
79 6 42% 1.63 0.08 64 14* 23.9 3.8* 20%; 80% 25 3 12 11* 11; 9–12* 13; 12–15*
81 6 22% 1.58 0.05y 70 16y 28.2 7.2y 11%; 89% 25 3 21 23* 10; 9–11* 14; 13–14*
0.29 0.15 0.08 0.06 <0.05 0.37 0.13 <0.01 <0.01 <0.01
Note: Values are mean SD or as otherwise indicated. Kruskal–Wallis test was used for group effect computation. Mann–Whitney test was used for between-groups comparisons. The Fisher’s Exact test was use to compare proportions. * Significantly different from non fallers, P < 0.05. y Multiple fallers significantly different from single fallers, P < 0.05.
COP position was more anterior in the F group than in the NF group (P < 0.01), while in the RF group, it remained posterior and similar to that of the NF group. The confidence ellipse area increased significantly in the RF group with respect to the NF one (P < 0.05), while no significant differences were found between the F group and the NF group. In the ECFS conditions, COP-related parameters showed the same association with the fall-history as the EOFS ones. In the EOCS task, five out of eight parameters were associated with the fall-history. The mean position of the COP in the AP direction confirmed the previously described trend with a significant increase for F (P < 0.05) and a significant backward variation for RF (P < 0.05). The confidence ellipse area increased significantly in the RF group (P < 0.01). The amount of increase in the RF group was about 70% with respect to the NF one (from 720 to
1220 mm2), while it was limited to about 20% in the tasks on a firm surface. The RMS of sway in both the AP and ML directions was significantly different between RF and both NF (P < 0.01) and F (P < 0.05) subjects. The mean velocity in the ML direction was significantly different between RF and NF (P < 0.05). In the ECCS task, none of the computed parameters was influenced by the fall-history. Similarly, no associations were found for the DT condition. 4. Discussion 4.1. Clinical and functional assessment This study investigated the association between posturographic parameters and fall-history in a consecutive sample of older
Table 2 Values of posturographic parameters obtained from older non-fallers (NF, N = 67), single fallers (F, N = 45) and recurrent fallers (RF, N = 18) during 30-s trials of quiet standing in different measuring conditions. Parameter
COP mean AP position (mm from heels)
COP Mean ML position (mm from heels midpoint)
Mean velocity (mm/s)
95% conf. ellipse area (mm2)
AP RMS (mm)
ML RMS (mm)
Mean AP velocity (mm/s)
Mean ML velocity (mm/s)
Fall history
Measuring condition Eyes open firm surface
Eyes closed firm surface
NF F RF NF
101 18 110 15** 98 16y 3 10
103 20 112 16* 99 17y 2 11
F RF NF F RF NF F RF NF F RF NF F RF NF F RF NF F RF
5 4 6 12 17 8 20 14 20 8 260 190 250 160 310 140* 51 52 62 31 31 31 15 7 18 13 18 8 62 63 72
3 14 6 12 25 14 27 21 29 15 310 280 290 510 390 200* 62 62 62 32 32 31 22 13 24 20 26 14 74 86 94
Note: Grey cells indicate significative effect group, Kruskal–Wallis test (p < 0.05). * Statistically different from non fallers (p < 0.05). ** Statistically different from non fallers (p < 0.01). y Statistically different from single fallers (p < 0.05). yy Statistically different from single fallers (p < 0.01).
Eyes open compliant surface 109 18 117 15* 108 17y 3 11 3 14 2 13 27 12 29 16 31 13 720 390 820 520 1220 590**,yy 83 93 11 3**,y 52 52 7 2**,yy 24 12 16 15 27 12 84 94 11 4*
Eyes closed compliant surface
Dual task
111 19 119 18 105 20 3 12
108 17 114 16 106 15 4 12
2 14 2 15 42 18 40 22 45 23 1110 620 1210 780 1330 900 11 3 11 4 11 4 52 62 62 38 16 36 20 40 21 12 6 12 7 14 6
6 13 5 10 21 10 25 15 26 12 360 270 350 220 340 180 52 21 51 32 41 31 18 9 22 14 23 11 84 84 84
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
individuals examined at the Memory Clinics of our institution and found cognitively intact (CDR staging 0) or affected by at most a mild cognitive impairment (CDR staging 0.5) without dementia. This last condition is characterized by either physiological aging or an early process of cognitive decline that does not affect balance [19,20]. Elderly adults with dementia have been excluded from this study, as dementia is a known risk factor sufficient to consider a subject at a high risk of falling without the need of further screening [21]. An age of at least 70 years was chosen to minimize the confounding effect of age on COP parameters, which can be found by including subjects from the age of 65 [6,22]. In our study, the distribution of age was homogenous in the three groups of fallers, thus controlling for the effect on results of age, which is a known factor of risk [1–3,21]. Both the POMA and the IADL scores presented an association with the fall history in accordance with literature [21,23]. 4.2. Instrumented assessment The values of parameters obtained from our group of nonfallers were consistent with normative data reported in the literature for older people [7,24]. The minor differences can be accounted for by the differences in cut-off frequencies used for the low-pass filter (5 Hz or 10 Hz). In the EOFS, ECFS and EOCS tasks, the mean AP position of the COP moved forward in single fallers with respect to non-fallers but not in recurrent fallers. The possibility of using the mean AP COP to detect fallers from non-fallers among cognitively able older people is new to the literature. Forward displacement of the COP mean position during quiet standing has been reported for patients with Parkinson’s disease only [25]. Our finding could rely on the evolution of kyphosis, a development that leads to forward displacement of the center of mass with a consequent forward displacement of the mean COP position [26]. This leads subsequently to compensatory mechanisms such as neck and knee flexion [26] to keep the line of sight horizontal, thus aligning the trunk backward over the base of support and bringing the COP mean position backward again. The reduction in the COP forward position could also become necessary as a consequence of the weakness of the plantar flexor muscles to limit their effort in the control of the body forward leaning. However, as we do not have direct measurements of joint angles and muscle strength, we can only speculate on these possible causes. The 95% confidence ellipse area increased significantly in recurrent fallers in the EOFS, ECFS and EOCS tasks, in accordance with results in [12,13] and the percentage increment in the ellipse area was much greater in the task on a compliant surface. It is interesting to note that a similar statistically significant increase was not found for the two unidirectional RMS values. RF subjects in the EOCS condition showed one or more occurrences of imbalance rather than a regular oscillation. Such occurrences are not adequately measured by the RMS, which takes a mean over the whole time duration of the trial. Thus, in further studies, local COPbased indicators should be considered. In our study, the effect of fall-history on the ML sway was found in the EOCS condition only, while in the literature, performance in the ML direction has been associated with future falls [14]. This may be due to differences in feet position. A more natural distance of about 15 cm between heels was used in our protocol, while most published papers acquired the quiet stance with narrow-based foot positions (e.g. feet together) [14]. The use of a non-challenging foot position could explain the lack of additional information provided by the ECFS condition in our study. Because of a sufficiently wide base of support, subjects did not rely on visual contribution to maintain balance.
665
Measurement of balance in the Eyes Open on a Compliant Surface condition was associated with the fall-history in cognitively able elderly subjects. The EOCS task enhanced the differences in most of the COP-based parameters between recurrent fallers and nonfallers. To the authors’ knowledge, this result is new to the literature. Standing on a foam surface is probably the most commonly used method to increase specificity in posturographic examinations [9], as it is an effective way to challenge postural control and produces a substantial, multi-directional balance perturbation [9]. The major sensitivity of the EOCS condition to the fall-history could be accounted for by the well-documented age-associated increase in sensory detection thresholds, the decrease in nerve conduction velocity and the decrease in central processing capacity [1,3,12]. Moreover, this condition elicits the demand for fast inversion of ankle joint moments as a region of foam gives way under body weight [27]. Consistent with our results, the ability to react quickly with the entire body during challenging conditions was found in the literature [21]. The assessment of balance in the EOCS task could explore these same abilities through a quick and safe task and provide indicators with different median values for non-fallers, single fallers or recurrent fallers. Based on our results, the EOCS task could be used to improve the multifactorial assessment of the risk of fall [6,28]. In the ECCS conditions, the parameter mean values of nonfallers were generally as high as those of multiple fallers with no sensitivity to the fall-history. This task was probably too challenging for the study population. The fact that the DT assessment did not reflect the history of fall in our sample can be explained by the exclusion of any subjects with dementia. A deterioration of COP-related parameters in DT is characteristic of elderly demented individuals, while it does not take place in age-matched controls [10]. 4.3. Limitations of this study A limit of this study is the acquisition of a single 30-s trial for each condition instead of three repetitions [29,30]. However, recent findings indicate a high to very high reliability of the variables investigated in our study in older people [30]. Despite this limitation, statistically significant differences between groups arose due to the joint effect of the EOCS condition and the sufficiently wide sample size. Another limitation is that the retrospective design of the study provides weaker evidence than prospective studies since older adults may under-report falls or not recall minor fall events. 5. Summary and conclusions Our results have shown a high degree of association between the fall-history of older people with no or mild cognitive impairment and their ability to control balance while standing with eyes open on a compliant surface. Results suggest that this evaluation can be included in the multifactorial analysis of the risk of fall for these individuals to improve the performance of risk-offall classifiers based on both clinical and instrumented parameters. Acknowledgments The authors would like to acknowledge Isabella Campanini for her precious contribution to the refinement of the award-winning presentation of this study at the XII congress of the Italian Society of Clinical Movement Analysis. This research was supported by the Fondazione per lo Studio delle Malattie Neurodegenerative delle Persone Adulte e dell’Anziano [Foundation for the Study of Neurodegenerative Diseases in Adults and the Elderly], Lugano, Switzerland.
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
666
Conflict of interest statement There are no conflicts of interest. References [1] Berry SD, Miller RR. Falls: epidemiology, pathophysiology, and relationship to fracture. Current Osteoporosis Report 2008;6(December (4)):149–54. [2] Kannus P, Sieva¨nen H, Palvanen M, Ja¨rvinen T, Parkkari J. Prevention of falls and consequent injuries in elderly people. Lancet 2005;366:1885–93. [3] American Geriatric Society, British Geriatric Society Panel on Falls Preventions. Guidelines for the prevention of falls in older persons. Journal of the American Geriatrics Society 2011;59:148–57. [4] Gillespie LD, Gillespie WJ, Robertson MC, et al. Interventions for preventing falls in elderly people. Cochrane Database System Review 2003;4:CD000340. [5] Tinetti ME, Williams TF, Mayewski R. Fall risk index for elderly patients based on number of chronic disabilities. American Journal of Medicine 1986;80: 429–524. [6] Panzer VP, Wakefield DB, Hall CB, Wolfson LI. Mobility assessment: sensitivity and specificity of measurement sets in older adults. Archives of Physical Medicine and Rehabilitation 2011;92:905–12. [7] Prieto TE, Myklebust JB, Hoffmann RG, Lovett EG, Myklebust BM. Measures of postural steadiness: differences between healthy young and elderly adults. IEEE Transactions on Biomedical Engineering 1996;43:956–66. [8] Nashner LM, Black FO, Wall C. Adaption to altered support and visual conditions during stance: patients with vestibular deficits. Journal of Neuroscience 1982;2:536–44. [9] Patel M, Fransson PA, Lush D, Gomez S. The effect of foam surface properties on postural stability assessment while standing. Gait & Posture 2008;28: 649–56. [10] Manckoundia P, Pfitzenmeyer P, d’Athis P, Dubost V, Mourey F. Impact of cognitive tasks on the posture of elderly subjects with Alzheimer’s disease compared to healthy elderly subjects. Movement Disorders 2006;21:236–73. [11] Nardone A, Schieppati M. The role of instrumental assessment of balance in clinical decision making. European Journal of Rehabilitation Medicine 2010; 46:221–37. [12] Melzer I, Kurz I, Oddsson IE. A retrospective analysis of balance control parameters in elderly fallers and non-fallers. Clinical Biomechanical 2005;984–8. [13] Norris JA, Marsh AP, Smith IJ, Kohut RI, Miller ME. Ability of static and statistical mechanics posturographic measures to distinguish between age and fall risk. Journal of Biomechanics 2005;38(June (6)):1263–72. [14] Piirtola M, Era P. Force platform measurements as predictors of falls among older people – a review. Gerontology 2006;52:1–16. [15] Lucca U, Tettamanti M. The Italian version of Consortium to Establish a Registry of Alzheimer Disease (CERAD). Alzheimer’s & Dementia 2008;4:310.
[16] Fillenbaum GG, van Belle G, Mirris JC, Mohs RC, Mirra SS, Davis PC, et al. Consortium to Establish a Registry of Alzheimer Disease (CERAD): the first twenty years. Alzheimer’s & Dementia 2008;4:96–109. [17] Cedraro A, Cappello A, Chiari L. A portable system for in situ re-calibration of force platforms: experimental validation. Gait & Posture 2009;29(April (3)):449–53. [18] Kirtley C. Clinical gait analysis. Theory and practice. New York: Churchill Livingstone; 2005. [19] Petersen RC, Smith GE, Waring SC, et al. Mild cognitive impairment. Clinical characterization and outcome. Archives of Neurology 1999;56:303–8. [20] Quadri P, Merlo A, Zanda E, Rocchi S, Meroni F, Zemp D, et al. Balance, gait, postural stability, and falls in mild cognitive impairment: the Canton Ticino Study. Alzheimer’s & Dementia 2011;7:S537. [21] Chu LW, Chiu AYY, Chi I. Impact of falls on the balance, gait, and activities of daily living functioning in community-dwelling Chinese older adults. Journal of Gerontology Series A Biological Sciences and Medical Sciences 2006;61(4): 399–404. [22] Tucker MG, Kavanagh JJ, Morrison S, Barrett RS. What are the relations between voluntary postural sway measures and falls-history status in community-dwelling older adults? Archives of Physical Medicine and Rehabilitation 2010;91:750–8. [23] Raıˆche M, He´bert R, Prince F, Corriveau H. Screening older adults at risk of falling with the Tinetti balance scale. Lancet 2000;356(September (9234)): 1001–2. [24] Demura S, Kitabayashi T, Aoki H. Body-sway characteristics during a static upright posture in the elderly. Geriatrics & Gerontology International 2008;8: 188–97. [25] Schieppati M, Nardone A. Free and supported stance in Parkinson’s disease. Brain 1991;(114):1227–44. [26] Ryan SD, Fried LP. The impact of kyphosis on daily functioning. JAGS 1997;(45):1479–86. [27] LaRoche DP, Cremin KA, Greenleaf B, Croce RV. Rapid torque development in older female fallers and nonfallers: a comparison across lower-extremity muscles. Journal of Electromyography Kinesiology 2010;20(3):482–8. [28] Gates S, Lamb SE, Fisher JD, Cook MW, Carter YH. Multifactorial assessment and targeted intervention for preventing falls and injuries among older people in community and emergency care settings: systematic review and metaanalysis. BMJ 2008;336:130. [29] Benvenuti F, Mecacci R, Gineprari I, Bandinelli S, Benvenuti E, Ferrucci L, et al. Kinematic characteristics of standing disequilibrium: reliability and validity of a posturographic protocol. Archives of Physical Medicine and Rehabilitation 1999;80:278–87. [30] Moghadam M, Ashayeri H, Salavati M, Sarafzadeh J, Taghipoor KD, Saeedi A, et al. Reliability of center of pressure measures of postural stability in healthy older adults: effects of postural task difficulty and cognitive load. Gait & Posture 2011;33:651–5.
Contents lists available at SciVerse ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
SIAMOC Best Clinical Paper Award 2011
Postural stability and history of falls in cognitively able older adults: The Canton Ticino study§ Andrea Merlo a,*, Damiano Zemp b, Enrica Zanda b, Sabrina Rocchi c, Fabiano Meroni b, Mauro Tettamanti d, Angela Recchia d, Ugo Lucca d, Pierluigi Quadri b a
LAM – Laboratorio Analisi Movimento, AUSL Reggio Emilia, SOC Neuroriabilitazione, Ospedale di Correggio, Reggio Emilia, Italy Servizio Sottocenerino di Geriatria, Ospedali regionali of Lugano and Mendrisio,Switzerland c Clinica Hildebrand, Centro di Riabilitazione, Brissago, Switzerland d Laboratory of Geriatric Neuropsychiatry, Istituto di Ricerche Farmacologiche ‘‘Mario Negri’’, Milano, Italy b
A R T I C L E I N F O
A B S T R A C T
Keywords: Aging Balance Falls Center of pressure Postural control
Falls are common events in the elderly and represent the main risk factor for fractures and other injuries. Strategies for fall prevention rely on the multifactorial assessment of the risk of falling. The contribution of instrumented balance assessment to the prediction of falls remains unclear in the literature. In this study, we analyzed the association between the fall-history of a wide sample of older people without dementia and the values of a set of posturographic parameters acquired in different visual, proprioceptive and mental conditions. A consecutive sample of 130 cognitively able elderly subjects, age 70 years, was analyzed. Based on their fall-history in the last year, subjects were categorized into non-fallers (NF), fallers (F) and recurrent fallers (RF > 2 falls). Each subject was assessed by measurements of cognition and functional ability. Static posturography tests were performed in five conditions: with eyes open/close (EO/EC) on a firm/compliant (FS/CS) surface and while performing a cognitive task. The center of pressure (COP) mean position referred to the mid-point of the heels, area of the 95% confidence ellipse, sway mean velocities and RMS displacements in the antero-posterior (AP) and medio-lateral (ML) directions were computed and their association with the fall-history was assessed. The mean position of the COP in the AP direction and the confidence ellipse area were associated with the fall-history in the EOFS, ECFS and EOCS conditions (P < 0.05). RMS displacements were also associated with the fall-history in the EOCS condition (P < 0.05). Significant group differences (P < 0.05) were found in the EOCS conditions, which greatly enhanced the differences among NF, F and RF. The ability to control balance while standing with eyes open on a compliant surface showed a high degree of association with the fall-history of older people with no or mild cognitive impairment. ß 2012 Elsevier B.V. All rights reserved.
1. Introduction Falls are common in the elderly and frequently result in injury, disability or hospitalization [1]. Around 30% of people aged 65 or older living in the community and more than 50% of those living in residential care facilities or nursing homes fall every year and about half of those who fall do so repeatedly [2]. Effective fall prevention has the potential to reduce serious fallrelated injuries [3] and related costs. Prevention of falls, however,
§ This paper was not selected through the usual peer review process but through a SIAMOC Reading Committee. * Corresponding author at: LAM – Laboratorio Analisi Movimento, Azienda USL di Reggio Emilia, Via Mandriolo Superiore 11, 42015 Correggio, RE, Italy. Tel.: +39 0522 630391; fax: +39 0522 630332. E-mail address: [email protected] (A. Merlo).
0966-6362/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gaitpost.2012.06.016
is not easy, because they are complex events caused by a combination of intrinsic impairments and disabilities with or without accompanying environmental hazards [1,2]. Thus, multifactorial fall risk assessment has been recommended by recent guidelines for the prevention of falls in older individuals [3,4]. The most commonly used mobility assessments are selfreported capacity in movements and functional rating scales, such as the Tinetti Performance Oriented Mobility Assessment (POMA) [5]. Objective performance indexes are also commonly used [6]. Along with these clinical and functional evaluations, the instrumental assessment of balance by force-platforms is a widely applied tool in assessing standing balance with several parameters computed from the displacement of the Center of Pressure (COP) [7]. Posturographic parameters have been investigated under different testing conditions, such as with open and closed eyes, on firm and compliant surfaces and while performing cognitive tasks [8–10].
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
The role of the instrumented assessment of balance in clinical decision-making has been recently reviewed in [11] and the contribution of COP related measures in the detection of the risk of falling has been recently reported [6,12,13]. However, other prospective studies indicated an overall marginal predictive value for COP related measures [14]. The different levels of cognitive impairment of the samples analyzed in these studies could act as a major confounding factor [1,3], as its effect on both balance and posturographic parameters is known [10]. In this study, we analyzed the association between the fallhistory of a wide sample of older people without dementia and the values of a set of posturographic parameters acquired in different visual, proprioceptive and cognitive conditions. 2. Methods 2.1. Sample The study population consisted of 130 cognitively able individuals (clinical dementia rating, CDR 0.5), 52 M and 78 F, mean age 79 6 years and age range 70–91 years, seen consecutively at the Memory Clinic of the Regional Hospitals of Mendrisio and Lugano, Switzerland. This sample is a sub-group of the Canton Ticino Study [15], an epidemiological database using the CERAD clinical and neuropsychological battery [16] that has included more than 1400 consecutive ambulatory elderly individuals since 1998. 2.2. Clinical evaluation A standard assessment was delivered to all patients, according to the CERAD protocol, including: clinical history (event, yes/no); measurement of cognition with the Mini Mental State Examination (MMSE); a complete battery of neuropsychological tests; measurement of functional ability, such as basic (BADL) and instrumental activities of daily living (IADL) scores and Gait and Balance performance on Tinetti’s Performance Oriented Mobility Assessment. Details on these scales can be found in [16]. 2.3. Fall history History of falls in the previous year was collected within the CERAD assessment. Subjects were categorized as non-fallers (NF), fallers (F, one or two falls) and recurrent fallers (RF, more than two falls).
663
2.5. Experimental procedure Each subject completed a sequence of five 30-s tasks, as follows: standing with eyes open on a firm surface (EOFS); standing with eyes closed on a firm surface (ECFS); standing with eyes open on a firm surface while performing a cognitive task (i.e. counting backward by steps of seven) referred to as a dual task (DT); standing with eyes open on a compliant surface (EOCS) and standing with eyes closed on a compliant surface (ECCS). The compliant surface was obtained by placing a (40 cm 40 cm 3 cm) viscoelastic gel pillow (Elastil II, Laboratores Escarius, La Courneuve, France) over the force plate. The subject sat on a chair while the assessor placed the pillow on the force plate. Subjects were asked to stand barefoot on the force platform in a relaxed posture with arms at their sides. Before each trial, foot position was standardized by means of a custom made removable device. The rear part of this device was larger than the force plate. It was inserted laterally to the force plate into two holes in the floor in a known, fixed position, thus ensuring that heels (and feet) were placed in a known, repeatable position with respect to the origin of the force plate. Moreover, the imposed fixed position of the medial margin of the feet led to minimum variations in the area of the base of support. The distance between the center of the heels ranged from 14 to 16 cm, depending on the subject’s foot size. The angle between each foot and the AP direction was 10 degrees, which corresponds to the natural average of toeing out during stance and gait [18]. The use of the positioning device allowed the placement of the heels in a known position over the force plate and, consequently, the expression of the COP coordinates with respect to the midpoint between the heels. A visual target (a colored circle with a diameter of 6 cm) was placed at a distance of 1.5 m at eyelevel. A single operator performed all acquisitions in both hospitals throughout the years. 2.6. Statistical analysis For each clinical, functional and posturographic parameter, we tested the null hypothesis that the median value was not affected by the fall-history in the last year, by means of a Kruskal–Wallis test. Between-group comparisons were carried out by using the Mann–Whitney test. The non-parametric approach was chosen, as it does not require normality of data and variance equality amongst groups. Comparisons between proportions were carried out by means of the Fisher Exact test. All analyses were conducted with statistical analysis software (SPSS 13.0) using a 5% level of significance.
2.4. Instrumentation 3. Results Posturographic data were acquired by means of one piezoelectric force plate (Kistler 9281C, Kistler, Winterthur, Switzerland) in Mendrisio and one strain-gage based force plate (AMTI OR-6, Watertown, MA, USA) in Lugano. In both cases, the acquisition was performed by using Bioware software (Kistler, Winterthur, Switzerland) at the sampling frequency of 100 Hz. The calibration of both force plates was checked before the beginning of the study as in [17]. Data were off-line filtered at 20 Hz by means of a zero-lag Butterwoth filter and the global COPrelated parameters were computed, as in [7], by means of a custom made Matlab software. Among the parameters that can be derived from static posturography tests, we focused on those that emerged from a systematic review on their use as predictors of falls among older people [14], including the antero-posterior (AP) and the medio-lateral (ML) mean COP position referred to the mid-point of the heels, the sway mean velocity, the AP and ML mean velocity, the AP and ML RMS displacement and the area of 95% confidence ellipse.
Participants were separated into fall-status entry groups: 67 non-fallers, 45 fallers and 18 recurrent fallers. Demographic, clinical and functional data of the sample are reported in Table 1. No age differences were found among non-fallers, fallers and recurrent fallers (Kruskal–Wallis, P = 0.29). Functional scales were affected by the fall-history (Kruskal–Wallis, P < 0.01). In detail, the median value of both POMA and IADL scores from F and RF subjects were significantly different than those from NF (Mann–Whitney, P < 0.05). No significant differences were found between fallers and recurrent fallers. Values of posturographic parameters (mean SD) are reported in Table 2 for all tasks. The presence of a statistically significant effect of fall-history on posturographic parameters (Kruskal–Wallis, P < 0.05) is outlined by a light-grey background of the cell. Between group statistically significant differences are indicated by symbols. In the EOFS task, the mean AP COP position and the area of the confidence ellipse were associated with the history of falls. The
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
664
Table 1 Demographic, clinical and functional characteristics of the sample. Descriptive measure
Non fallers (NF) (N = 67)
1–2 times fallers (F) (N = 45)
Recurrent fallers (RF) (N = 18)
Group effect P
Age (y) Sex (% men) Height (m) Weight (kg) Body mass index (kg/m2) Clinical dementia rating CDR = 0 (%); CDR = 0.5 (%) Mini mental state evaluation IADL (percentage of dependence) POMA, gait score (median; 25th–75th percentile) POMA, balance score (median; 25th–75th percentile)
79 5 43% 1.62 0.10 69 14 26.3 4.9 25%; 75% 26 3 8 16 11; 11–12 14; 14–15
79 6 42% 1.63 0.08 64 14* 23.9 3.8* 20%; 80% 25 3 12 11* 11; 9–12* 13; 12–15*
81 6 22% 1.58 0.05y 70 16y 28.2 7.2y 11%; 89% 25 3 21 23* 10; 9–11* 14; 13–14*
0.29 0.15 0.08 0.06 <0.05 0.37 0.13 <0.01 <0.01 <0.01
Note: Values are mean SD or as otherwise indicated. Kruskal–Wallis test was used for group effect computation. Mann–Whitney test was used for between-groups comparisons. The Fisher’s Exact test was use to compare proportions. * Significantly different from non fallers, P < 0.05. y Multiple fallers significantly different from single fallers, P < 0.05.
COP position was more anterior in the F group than in the NF group (P < 0.01), while in the RF group, it remained posterior and similar to that of the NF group. The confidence ellipse area increased significantly in the RF group with respect to the NF one (P < 0.05), while no significant differences were found between the F group and the NF group. In the ECFS conditions, COP-related parameters showed the same association with the fall-history as the EOFS ones. In the EOCS task, five out of eight parameters were associated with the fall-history. The mean position of the COP in the AP direction confirmed the previously described trend with a significant increase for F (P < 0.05) and a significant backward variation for RF (P < 0.05). The confidence ellipse area increased significantly in the RF group (P < 0.01). The amount of increase in the RF group was about 70% with respect to the NF one (from 720 to
1220 mm2), while it was limited to about 20% in the tasks on a firm surface. The RMS of sway in both the AP and ML directions was significantly different between RF and both NF (P < 0.01) and F (P < 0.05) subjects. The mean velocity in the ML direction was significantly different between RF and NF (P < 0.05). In the ECCS task, none of the computed parameters was influenced by the fall-history. Similarly, no associations were found for the DT condition. 4. Discussion 4.1. Clinical and functional assessment This study investigated the association between posturographic parameters and fall-history in a consecutive sample of older
Table 2 Values of posturographic parameters obtained from older non-fallers (NF, N = 67), single fallers (F, N = 45) and recurrent fallers (RF, N = 18) during 30-s trials of quiet standing in different measuring conditions. Parameter
COP mean AP position (mm from heels)
COP Mean ML position (mm from heels midpoint)
Mean velocity (mm/s)
95% conf. ellipse area (mm2)
AP RMS (mm)
ML RMS (mm)
Mean AP velocity (mm/s)
Mean ML velocity (mm/s)
Fall history
Measuring condition Eyes open firm surface
Eyes closed firm surface
NF F RF NF
101 18 110 15** 98 16y 3 10
103 20 112 16* 99 17y 2 11
F RF NF F RF NF F RF NF F RF NF F RF NF F RF NF F RF
5 4 6 12 17 8 20 14 20 8 260 190 250 160 310 140* 51 52 62 31 31 31 15 7 18 13 18 8 62 63 72
3 14 6 12 25 14 27 21 29 15 310 280 290 510 390 200* 62 62 62 32 32 31 22 13 24 20 26 14 74 86 94
Note: Grey cells indicate significative effect group, Kruskal–Wallis test (p < 0.05). * Statistically different from non fallers (p < 0.05). ** Statistically different from non fallers (p < 0.01). y Statistically different from single fallers (p < 0.05). yy Statistically different from single fallers (p < 0.01).
Eyes open compliant surface 109 18 117 15* 108 17y 3 11 3 14 2 13 27 12 29 16 31 13 720 390 820 520 1220 590**,yy 83 93 11 3**,y 52 52 7 2**,yy 24 12 16 15 27 12 84 94 11 4*
Eyes closed compliant surface
Dual task
111 19 119 18 105 20 3 12
108 17 114 16 106 15 4 12
2 14 2 15 42 18 40 22 45 23 1110 620 1210 780 1330 900 11 3 11 4 11 4 52 62 62 38 16 36 20 40 21 12 6 12 7 14 6
6 13 5 10 21 10 25 15 26 12 360 270 350 220 340 180 52 21 51 32 41 31 18 9 22 14 23 11 84 84 84
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
individuals examined at the Memory Clinics of our institution and found cognitively intact (CDR staging 0) or affected by at most a mild cognitive impairment (CDR staging 0.5) without dementia. This last condition is characterized by either physiological aging or an early process of cognitive decline that does not affect balance [19,20]. Elderly adults with dementia have been excluded from this study, as dementia is a known risk factor sufficient to consider a subject at a high risk of falling without the need of further screening [21]. An age of at least 70 years was chosen to minimize the confounding effect of age on COP parameters, which can be found by including subjects from the age of 65 [6,22]. In our study, the distribution of age was homogenous in the three groups of fallers, thus controlling for the effect on results of age, which is a known factor of risk [1–3,21]. Both the POMA and the IADL scores presented an association with the fall history in accordance with literature [21,23]. 4.2. Instrumented assessment The values of parameters obtained from our group of nonfallers were consistent with normative data reported in the literature for older people [7,24]. The minor differences can be accounted for by the differences in cut-off frequencies used for the low-pass filter (5 Hz or 10 Hz). In the EOFS, ECFS and EOCS tasks, the mean AP position of the COP moved forward in single fallers with respect to non-fallers but not in recurrent fallers. The possibility of using the mean AP COP to detect fallers from non-fallers among cognitively able older people is new to the literature. Forward displacement of the COP mean position during quiet standing has been reported for patients with Parkinson’s disease only [25]. Our finding could rely on the evolution of kyphosis, a development that leads to forward displacement of the center of mass with a consequent forward displacement of the mean COP position [26]. This leads subsequently to compensatory mechanisms such as neck and knee flexion [26] to keep the line of sight horizontal, thus aligning the trunk backward over the base of support and bringing the COP mean position backward again. The reduction in the COP forward position could also become necessary as a consequence of the weakness of the plantar flexor muscles to limit their effort in the control of the body forward leaning. However, as we do not have direct measurements of joint angles and muscle strength, we can only speculate on these possible causes. The 95% confidence ellipse area increased significantly in recurrent fallers in the EOFS, ECFS and EOCS tasks, in accordance with results in [12,13] and the percentage increment in the ellipse area was much greater in the task on a compliant surface. It is interesting to note that a similar statistically significant increase was not found for the two unidirectional RMS values. RF subjects in the EOCS condition showed one or more occurrences of imbalance rather than a regular oscillation. Such occurrences are not adequately measured by the RMS, which takes a mean over the whole time duration of the trial. Thus, in further studies, local COPbased indicators should be considered. In our study, the effect of fall-history on the ML sway was found in the EOCS condition only, while in the literature, performance in the ML direction has been associated with future falls [14]. This may be due to differences in feet position. A more natural distance of about 15 cm between heels was used in our protocol, while most published papers acquired the quiet stance with narrow-based foot positions (e.g. feet together) [14]. The use of a non-challenging foot position could explain the lack of additional information provided by the ECFS condition in our study. Because of a sufficiently wide base of support, subjects did not rely on visual contribution to maintain balance.
665
Measurement of balance in the Eyes Open on a Compliant Surface condition was associated with the fall-history in cognitively able elderly subjects. The EOCS task enhanced the differences in most of the COP-based parameters between recurrent fallers and nonfallers. To the authors’ knowledge, this result is new to the literature. Standing on a foam surface is probably the most commonly used method to increase specificity in posturographic examinations [9], as it is an effective way to challenge postural control and produces a substantial, multi-directional balance perturbation [9]. The major sensitivity of the EOCS condition to the fall-history could be accounted for by the well-documented age-associated increase in sensory detection thresholds, the decrease in nerve conduction velocity and the decrease in central processing capacity [1,3,12]. Moreover, this condition elicits the demand for fast inversion of ankle joint moments as a region of foam gives way under body weight [27]. Consistent with our results, the ability to react quickly with the entire body during challenging conditions was found in the literature [21]. The assessment of balance in the EOCS task could explore these same abilities through a quick and safe task and provide indicators with different median values for non-fallers, single fallers or recurrent fallers. Based on our results, the EOCS task could be used to improve the multifactorial assessment of the risk of fall [6,28]. In the ECCS conditions, the parameter mean values of nonfallers were generally as high as those of multiple fallers with no sensitivity to the fall-history. This task was probably too challenging for the study population. The fact that the DT assessment did not reflect the history of fall in our sample can be explained by the exclusion of any subjects with dementia. A deterioration of COP-related parameters in DT is characteristic of elderly demented individuals, while it does not take place in age-matched controls [10]. 4.3. Limitations of this study A limit of this study is the acquisition of a single 30-s trial for each condition instead of three repetitions [29,30]. However, recent findings indicate a high to very high reliability of the variables investigated in our study in older people [30]. Despite this limitation, statistically significant differences between groups arose due to the joint effect of the EOCS condition and the sufficiently wide sample size. Another limitation is that the retrospective design of the study provides weaker evidence than prospective studies since older adults may under-report falls or not recall minor fall events. 5. Summary and conclusions Our results have shown a high degree of association between the fall-history of older people with no or mild cognitive impairment and their ability to control balance while standing with eyes open on a compliant surface. Results suggest that this evaluation can be included in the multifactorial analysis of the risk of fall for these individuals to improve the performance of risk-offall classifiers based on both clinical and instrumented parameters. Acknowledgments The authors would like to acknowledge Isabella Campanini for her precious contribution to the refinement of the award-winning presentation of this study at the XII congress of the Italian Society of Clinical Movement Analysis. This research was supported by the Fondazione per lo Studio delle Malattie Neurodegenerative delle Persone Adulte e dell’Anziano [Foundation for the Study of Neurodegenerative Diseases in Adults and the Elderly], Lugano, Switzerland.
A. Merlo et al. / Gait & Posture 36 (2012) 662–666
666
Conflict of interest statement There are no conflicts of interest. References [1] Berry SD, Miller RR. Falls: epidemiology, pathophysiology, and relationship to fracture. Current Osteoporosis Report 2008;6(December (4)):149–54. [2] Kannus P, Sieva¨nen H, Palvanen M, Ja¨rvinen T, Parkkari J. Prevention of falls and consequent injuries in elderly people. Lancet 2005;366:1885–93. [3] American Geriatric Society, British Geriatric Society Panel on Falls Preventions. Guidelines for the prevention of falls in older persons. Journal of the American Geriatrics Society 2011;59:148–57. [4] Gillespie LD, Gillespie WJ, Robertson MC, et al. Interventions for preventing falls in elderly people. Cochrane Database System Review 2003;4:CD000340. [5] Tinetti ME, Williams TF, Mayewski R. Fall risk index for elderly patients based on number of chronic disabilities. American Journal of Medicine 1986;80: 429–524. [6] Panzer VP, Wakefield DB, Hall CB, Wolfson LI. Mobility assessment: sensitivity and specificity of measurement sets in older adults. Archives of Physical Medicine and Rehabilitation 2011;92:905–12. [7] Prieto TE, Myklebust JB, Hoffmann RG, Lovett EG, Myklebust BM. Measures of postural steadiness: differences between healthy young and elderly adults. IEEE Transactions on Biomedical Engineering 1996;43:956–66. [8] Nashner LM, Black FO, Wall C. Adaption to altered support and visual conditions during stance: patients with vestibular deficits. Journal of Neuroscience 1982;2:536–44. [9] Patel M, Fransson PA, Lush D, Gomez S. The effect of foam surface properties on postural stability assessment while standing. Gait & Posture 2008;28: 649–56. [10] Manckoundia P, Pfitzenmeyer P, d’Athis P, Dubost V, Mourey F. Impact of cognitive tasks on the posture of elderly subjects with Alzheimer’s disease compared to healthy elderly subjects. Movement Disorders 2006;21:236–73. [11] Nardone A, Schieppati M. The role of instrumental assessment of balance in clinical decision making. European Journal of Rehabilitation Medicine 2010; 46:221–37. [12] Melzer I, Kurz I, Oddsson IE. A retrospective analysis of balance control parameters in elderly fallers and non-fallers. Clinical Biomechanical 2005;984–8. [13] Norris JA, Marsh AP, Smith IJ, Kohut RI, Miller ME. Ability of static and statistical mechanics posturographic measures to distinguish between age and fall risk. Journal of Biomechanics 2005;38(June (6)):1263–72. [14] Piirtola M, Era P. Force platform measurements as predictors of falls among older people – a review. Gerontology 2006;52:1–16. [15] Lucca U, Tettamanti M. The Italian version of Consortium to Establish a Registry of Alzheimer Disease (CERAD). Alzheimer’s & Dementia 2008;4:310.
[16] Fillenbaum GG, van Belle G, Mirris JC, Mohs RC, Mirra SS, Davis PC, et al. Consortium to Establish a Registry of Alzheimer Disease (CERAD): the first twenty years. Alzheimer’s & Dementia 2008;4:96–109. [17] Cedraro A, Cappello A, Chiari L. A portable system for in situ re-calibration of force platforms: experimental validation. Gait & Posture 2009;29(April (3)):449–53. [18] Kirtley C. Clinical gait analysis. Theory and practice. New York: Churchill Livingstone; 2005. [19] Petersen RC, Smith GE, Waring SC, et al. Mild cognitive impairment. Clinical characterization and outcome. Archives of Neurology 1999;56:303–8. [20] Quadri P, Merlo A, Zanda E, Rocchi S, Meroni F, Zemp D, et al. Balance, gait, postural stability, and falls in mild cognitive impairment: the Canton Ticino Study. Alzheimer’s & Dementia 2011;7:S537. [21] Chu LW, Chiu AYY, Chi I. Impact of falls on the balance, gait, and activities of daily living functioning in community-dwelling Chinese older adults. Journal of Gerontology Series A Biological Sciences and Medical Sciences 2006;61(4): 399–404. [22] Tucker MG, Kavanagh JJ, Morrison S, Barrett RS. What are the relations between voluntary postural sway measures and falls-history status in community-dwelling older adults? Archives of Physical Medicine and Rehabilitation 2010;91:750–8. [23] Raıˆche M, He´bert R, Prince F, Corriveau H. Screening older adults at risk of falling with the Tinetti balance scale. Lancet 2000;356(September (9234)): 1001–2. [24] Demura S, Kitabayashi T, Aoki H. Body-sway characteristics during a static upright posture in the elderly. Geriatrics & Gerontology International 2008;8: 188–97. [25] Schieppati M, Nardone A. Free and supported stance in Parkinson’s disease. Brain 1991;(114):1227–44. [26] Ryan SD, Fried LP. The impact of kyphosis on daily functioning. JAGS 1997;(45):1479–86. [27] LaRoche DP, Cremin KA, Greenleaf B, Croce RV. Rapid torque development in older female fallers and nonfallers: a comparison across lower-extremity muscles. Journal of Electromyography Kinesiology 2010;20(3):482–8. [28] Gates S, Lamb SE, Fisher JD, Cook MW, Carter YH. Multifactorial assessment and targeted intervention for preventing falls and injuries among older people in community and emergency care settings: systematic review and metaanalysis. BMJ 2008;336:130. [29] Benvenuti F, Mecacci R, Gineprari I, Bandinelli S, Benvenuti E, Ferrucci L, et al. Kinematic characteristics of standing disequilibrium: reliability and validity of a posturographic protocol. Archives of Physical Medicine and Rehabilitation 1999;80:278–87. [30] Moghadam M, Ashayeri H, Salavati M, Sarafzadeh J, Taghipoor KD, Saeedi A, et al. Reliability of center of pressure measures of postural stability in healthy older adults: effects of postural task difficulty and cognitive load. Gait & Posture 2011;33:651–5.