Dance Exercise for Older Adults: A Pilot StudyInvestigating Standing Balance Following a SingleLesson of Danzon

Azucena Guzman-Garcıa • Leif Johannsen •

Alan M. Wing

� American Dance Therapy Association 2011

Abstract Despite evidence that long-term dance exercise results in improved

functional balance in older adults, it remains unclear whether people must first

achieve a minimum level of skill before they can benefit from the exercise in terms

of body balance. In order to evaluate whether or not exercise during early learning

of dance is beneficial, we asked dance-inexperienced older adults to perform in pairs

a 30-min lesson of danzon. Motion capture and ground reaction force recordings

were used to determine the variability of body sway during quiet, upright standing

before and immediately after the lesson. Significant reductions in variability of sway

normalized by body height were found following the dance exercise. Interestingly,

the taller individual who was instructed to take the ‘‘leader’’ role showed greater

gains. Our findings indicate that learning to dance danzon at a beginner’s level may

be beneficial for balance in the elderly, despite the increased cognitive load in the

early stages of skill acquisition.

Keywords Dance interventions � Danzon � Aging � Posture � Balance � Body sway

Introduction

Short-term improvement in the stability of body balance following various forms of

physical exercise is a relatively robust outcome for older adults (Howe, Rochester,

A. Guzman-Garcıa

Institute for Ageing and Health, Newcastle University, Campus for Ageing and Vitality,

Newcastle upon Tyne NE4 5PL, UK

e-mail: Azucena.Guzman@ncl.ac.uk

L. Johannsen (&) � A. M. Wing

Behavioural Brain Sciences Centre, School of Psychology, University of Birmingham,

Birmingham B15 2TT, UK

e-mail: l.johannsen@bham.ac.uk

123

Am J Dance Ther

DOI 10.1007/s10465-011-9114-6

Jackson, Banks, & Blair, 2007). For example, water-based exercises have been

shown to improve standing and leaning balance and superior shoulder flexibility at

retest in community-dwelling older people (Lord et al., 2006). Soft style martial arts

with slow motion routines such as Tai Chi Chuan have been recommended for

exercising body balance as well (Wolfson, 1996; Judge, 2003). Furthermore, single-

leg strength, balance with eyes open and closed, locomotion and agility were

improved following dance-based aerobic exercise in older women (Shigematsu

et al., 2002). Finally, dance exercises have been reported to result in improved

functional balance both when performed by isolated individuals (Zhang, Ishikawa-

Takata, Yamazaki, Morita, & Ohta, 2006; Buchneret al., 1997) or in pairs and

groups (Sofianidis, Hatzitaki, Douka, & Grouios, 2009; McKinley et al., 2008;

Federici, Bellagamba, & Rocchi, 2005).

In a comparison of older regular social dancers with experience of ballroom, line

and swing dancing against older non-dancers, the social dancers showed not only

better standing balance but also faster gait, longer stride length, and relatively more

time spent in the swing phase per gait cycle (Verghese, 2006). These results were

interpreted as evidence of more stable balance while walking in older social

dancers. More generally, it has been inferred from such findings that engaging

senior populations in a particularly enjoyable form of physical exercise, such as

dance, results in improved physical ability as well as reduced risk for fall-related

injuries (Keogh, Kilding, Pidgeon, Ashley, & Gillis, 2009).

On the other hand, learning to dance is likely to involve significant cognitive

resources, such as attention, executive function, and memory, especially early in

skill acquisition, before the sequencing of movements becomes automatic (Fitts &

Posner, 1967). It might therefore be assumed that learning to dance as a coordinated

pair would show some costs, and not lead immediately to benefits in dynamic

balance stability. Indeed, the ‘‘learning to dance’’ exercise may be experienced as

quite demanding and lead to physical fatigue rather than improvement in balance. It

could have important implications for elderly dance training if dance activities only

begin to show benefits after a specific level of skill is reached. It could even imply

increased risk in initial stages of learning. An important question therefore, is

whether older adults without any previous dance exposure might fail to show any

beneficial gains or even exhibit reductions in dynamic body stability during early

phases of learning to dance.

In order to assess this question, our pilot study focused on a group of

inexperienced older adults who learned the choreography of an unfamiliar pair

dance. For our study, we chose the danzon, a popular Latin ballroom dance enjoyed

by middle-aged and older adults in Mexico. The danzon technique is performed at a

slow-to-moderate speed that allows for the assembling and disassembling of steps,

and for performing small steps in which the feet slide or drag and stop on each beat

along with gentle hip and body movements. These movements contrast with tango

or foxtrot, for example, both of which involve stylized walking, with strong lifted

marked steps between beats (Flores y Escalante, 2006). The danzon style appears to

afford lower physical demands, while still providing moderate cardiovascular

exercise.

Am J Dance Ther

123

Method

Participants

Ten healthy older adults (mean = 65.1 years, SD = 4.2; 9 females, 1 male) were

recruited in five pairs (4 same-gender, 1 mixed-gender) in a community setting.

Subject inclusion criteria were: 60 years of age or older; able to walk independently

without walking aid; must not have experienced any falls in the previous 12 months.

Participants were not involved in any dance exercise on a regular basis prior to this

study. In same-gender pairs, the taller individual was assigned the ‘‘leader role’’

during the dance (average ‘‘leader’’ vertical C7 vertebra’s height = 1424 mm,

SD = 39; average ‘‘follower’’ vertical C7 vertebra’s height = 1361 mm, SD = 24;

t(8) = 3.02, p = 0.02). Written informed consent was obtained from all participants

and the study was approved by the University of Birmingham Ethics Committee.

Materials and Procedure

After familiarization with the music, every pair attended a 30-min danzon lesson,

instructed by Guzman-Garcıa. In this pilot study, the choreography followed by

each pair consisted of a set of four steps: (1) ‘‘danzon entrance’’ pattern (dancers

stand next to each other and touch each other’s hand at the index, middle and ring

fingers) performed at the start, middle and end of a song; (2) ‘‘box step;’’ (3)

‘‘lateral’’ and (4) ‘‘twirl dancing backwards/forwards’’ (in which the leader helps the

follower turn). Each dance sequence lasted about 3 min and ten sequences were

performed. After each sequence, the pair received brief, personalized feedback

about their synchronization, overall body posture, interpersonal coordination and

stepping accuracy.

Immediately before and after the lesson, body sway during quiet upright

standing, with and without vision while facing straight ahead, was recorded

simultaneously for both individuals in a pair. Both participants were standing side-

by-side about one meter apart. The person taking on the ‘‘leader’’ role was always

standing to the left of the ‘‘follower,’’ which coincided with the ‘‘danzon entrance’’

positions at the start and end of each dance sequence. Three standing trials of 15 s

duration were recorded for each vision condition in a random sequence. Each

participant wore comfortable shoes and stood with a 5 cm gap between their feet on

two separate force platforms (model 4060H by Bertec, Columbus, Ohio), used to

record ground reaction forces and moments for determining the anterior–posterior

(AP) and medial–lateral (ML) components of centre-of-pressure (CoP) fluctuations.

CoP fluctuations reflect the generation of ankle torque as part of the central nervous

system’s control of the location of the body’s centre-of-mass (CoM). Body balance

can be considered stable if the horizontal position of CoM is kept within the base of

support defined by the placement of the feet (Winter, 2005).

Participants’ body sway kinematics were captured with an optoelectronic motion

tracking system of twelve cameras (Oqus 300 cameras and Qualisys software,

Gothenburg, Sweden). Participants had reflective markers attached to the seventh

cervical vertebra (C7); right and left shoulders; sternum; elbow; left and right wrist;

Am J Dance Ther

123

back and front of pelvic bones, left and right ankles; back of both shoes as well as

left and right toes. Actual sway of the trunk, in terms of either the C7 or the sternum

markers, was used for approximating CoM motion. Kinematics were recorded at

200 Hz, while data from the force platforms were sampled at 1,000 Hz. Kinematic

and kinetic data were digitally low-pass filtered at 10 Hz and differentiated when

velocity measures were required (dC7, dCoP).

Variability of body sway in terms of the standard deviation (SD) was computed

from each complete trial for the trunk and CoP velocity measures. The resulting

variables (SD dC7, SD dCoP) were normalized by dividing by the participant’s

height and multiplying by 1000 (mm/s 9 1/mm = s-1) and were averaged for each

vision condition. Normalization of body sway by height was performed to control

for differences in height between the ‘‘leader’’ and ‘‘follower’’ roles. As we searched

for exercise effects on two variables, a mixed, repeated-measures, multivariate

analysis of variance (MANOVA) with exercise, sway direction and vision condition

as within-subject factors and dance role as between-subject factor was computed.

Significance levels were set to 0.05. We report conservative Greenhouse-Geisser

corrected p-values, in order to protect our statistical conclusions against violations

of the prerequisites for parametric statistical tests such as the MANOVA.

Results

All significant effects and interactions that resulted from the analysis are listed in

Table 1. On both measures, sway was more variable with closed eyes (dC7:

mean = 6.07 s-1, SD = 1.81; dCoP: mean = 12.70 s-1, SD = 4.70) than with

open eyes (dC7: mean = 3.97 s-1, SD = 0.52 5; dCoP: mean = 7.05 s-1, SD =

0.95) and more variable in the AP (dC7: mean = 5.82 s-1, SD = 1.25; dCoP:

mean = 11.60 s-1, SD = 3.47) than the ML direction (dC7: mean = 4.22 s-1,

SD = 0.99; dCoP: mean = 8.15 s-1, SD = 2.55). Variability of sway was greatest

in the AP direction with eyes closed and smallest in the ML direction with eyes

open.

Table 1 Effect table for MANOVA on the two outcome measures representing body sway

Main effect/interaction Outcome variable

SD dC7 (univariate) SD dCoP (univariate)

Dance exercise F = 3.67#

Dance exercise 9 sway direction F = 6.17*

Dance exercise 9 vision condition F = 10.42* F = 10.32*

Dance exercise 9 vision condition 9 role F = 12.91* F = 10.25*

Sway direction F = 33.95* F = 18.15*

Vision condition F = 16.40* F = 22.75*

Sway direction 9 vision condition F = 8.11* F = 7.98*

* p B 0.05 (significant)# p B 0.10 (marginal)

Am J Dance Ther

123

An effect of practising danzon on body sway was found with less sway after the

exercise. Regarding the change of SD dCoP, a larger reduction was found without

vision (mean = -2.32 s-1, SD = 3.38) than with open eyes (mean = -0.29 s-1,

SD = 1.60) and in the AP (mean = -1.83 s-1, SD = 2.01) than the ML sway

direction (mean = -0.78 s-1, SD = 2.57). Interestingly, both outcome measures

demonstrated a significant interaction between exercise, vision condition, and role.

Figure 1 shows SD dCoP as a function of the danzon exercise for both directions

of sway and vision conditions as well as the participants’ roles. The change in AP

sway in terms of SD dCoP and SD dC7 with closed eyes after the dance lesson for

the two corresponding participants in each pair is shown in Fig. 2. In four out of five

pairs, dCoP sway showed a greater reduction for the ‘‘leader’’ role compared to the

‘‘follower’’ (Fig. 2a). These four ‘‘leading’’ individuals also showed reductions in

their dC7 sway in contrast to their ‘‘following’’ partners, who showed either no

change or increase (Fig. 2b).

Discussion

In a 30 min dance lesson of moderate intensity, inexperienced older adults learning

basic steps of danzon showed a significant reduction of body sway in the AP direction

during normal quiet standing without vision. This finding indicates that older

individuals show immediate improvement of body balance control following a single

dance lesson, despite having to divide their attention to various aspects of learning,

such as performing the danzon sequences in synchrony with their partner and the

music. Interestingly, the individual who acted in the ‘‘leader’’ role showed greater

reduction of sway even when statistical differences in physical body height were

controlled for. Since three of the four pairs showing this effect pattern were all female,

this finding cannot be attributed to biological gender. Rather, it seems that the

requirement to learn to take the ‘‘lead’’ role induced a greater benefit for controlling

body sway without visual feedback. However, it could also mean that the ‘‘leader’’

step sequence is intrinsically more beneficial even though the ‘‘follower’’ needs to be

as skilled in dancing the sequence as the ‘‘leader,’’ in order to perform in a

coordinated and synchronized fashion. With the exception of the leader helping the

follower to perform a turn, dancers were mirroring movements throughout the

choreography. For example at the beginning, middle, and end of the song, both

dancers had to perform the ‘‘danzon entrance’’ step. Even during this symmetric

dance maneuver, however, the ‘‘leader’’ might have implicitly adopted a more active

role than the ‘‘follower’’ by guiding the ‘‘follower’s’’ movements using fingertip

‘‘light touch’’ cues (Johannsen, Guzman-Garcıa, & Wing, 2009). One remote

possibility could be that the ‘‘follower’’ experienced some form of vestibular

aftereffect due to the turn in the training sequence, which could conceivably

counteract any effects of the training. Although a few minutes passed between the end

of the lesson and the following sway assessment, the ‘‘follower’’ might have swayed

slightly more than they would have if they had not performed the turn. In general, this

outcome suggests that during danzon training, pairs should be encouraged to switch

roles regularly in order to attain similar levels of balance benefits.

Am J Dance Ther

123

The results of the present pilot study, albeit significant, are based on a small

sample. Nevertheless, we are not aware of any previously published study of the

physical benefit of the danzon for healthy older adults. Hence, this study should be

of considerable potential interest in the search for effective interventions in

maintaining fitness and balance in older adults. An extension of the dance approach

in a larger study seems warranted. Whether the immediate benefits of learning to

dance for postural stability are caused by the complex dual-task nature of the

exercise, increased levels of arousal or improved sensory processing need to be

B

A

Fig. 1 Body sway (SD of CoP rate of change) in the AP (a) and the ML (b) direction, before and afterthe danzon lesson for both vision conditions and the two ‘‘roles.’’ Open markers indicate the ‘‘leader’’role, closed markers the ‘‘follower.’’ The error bars represent plus and minus one standard error of mean

Am J Dance Ther

123

studied further. Follow-up research should also control for the physical activity,

long-term effects, and active ‘‘dance roles’’ taken by each of the two individuals.

Also, our findings do not imply that danzon exercises are beneficial for standing

balance only in the elderly when practiced as an unfamiliar dance. On the contrary,

we expect that long-term practice, with the accompanying increased exposure and

greater familiarity, would augment any benefits of the dance exercise on body

balance. On the other hand, our results may imply that a dance native to a specific

elderly population, other than the danzon, yet minimally challenging, might become

the dance of choice for improving postural stability.

Danzón role

'Follower' 'Leader'

Cha

nge

in a

/p d

CoP

sw

ay (

s-1)

-6

-5

-4

-3

-2

-1

0

1

Pair IPair IIPair IIIPair IVPair V

Eyes closed

Male participant

Danzón role'Follower' 'Leader'

Cha

nge

in a

/p d

C7

sway

(s-1

)

-4

-3

-2

-1

0

1

2

3

Pair IPair IIPair IIIPair IVPair V

Eyes closed

Male participant

A

B

Fig. 2 Change in normalized AP sway following the danzon lesson in terms of SD dCoP (a) and SD dC7(b) during quiet standing with closed eyes for the individuals of each pair

Am J Dance Ther

123

In conclusion, we have shown that practicing a dance style of moderate intensity,

such as the danzon, with minimal skill, has potential benefits on the sensorimotor

control of body sway, following just a single dance session. Although we cannot be

certain that these immediate training effects last longer than the relatively short time

span of one session, one might speculate that this relatively easy dance routine may

result in long-term improvement in dynamic balance stability and a reduced risk for

fall-related injury if done regularly. Another implication for dancing interventions

in general is that roles should be interchanged regularly, as to maximize the benefits

for both participants in a pair.

Acknowledgement We acknowledge the financial support of the Biotechnology and Biological

Sciences Research Council (BBSRC; BBF0100871) awarded to Professor Alan Wing and the Mexican

National Council for Science and Technology (CONACYT) awarded to Azucena Guzman-Garcıa, PhD

Candidate. Further information about danzon can be provided by the authors on request.

References

Buchner, D. M., Cress, M. E., de Lateur, B. J., Esselman, P. C., Margherita, A. J., Price, R., et al. (1997).

A comparison of the effects of three types of endurance training on balance and other fall risk

factors in older adults. Aging Clinical and Experimental Research, 9(1–2), 112–119.

Federici, A., Bellagamba, S., & Rocchi, M. B. L. (2005). Does dance-based training improve balance in

adult and young old subjects? A pilot randomized controlled trial. Aging Clinical ExperimentalResearch, 17(5), 385–389.

Fitts, P. M., & Posner, M. I. (1967). Human performance. Belmont, CA: Brooks/Cole.

Flores y Escalante, J. (2006). Imagenes del Danzon, Iconografıa del Danzon en Mexico. [Danzon Images,Danzon Iconography in Mexico]. Mexico City: AMEF.

Howe, T. E., Rochester, L., Jackson, A., Banks, P. M. H., & Blair, V. A. (2007). Exercise for improving

balance in older people. Cochrane Database Systematic Review, 4. CD004963. doi:10.1002/146518

58.CD004963.pub2.

Johannsen, L., Guzman-Garcıa, A., & Wing, A. M. (2009). Interpersonal light touch assists balance in the

elderly. Journal of Motor Behaviour, 41(5), 397–399.

Judge, J. O. (2003). Balance training to maintain mobility and prevent disability. American Journal ofPreventive Medicine, 25(3), 150–156.

Keogh, J. W. L., Kilding, A., Pidgeon, P., Ashley, L., & Gillis, D. (2009). Physical benefits of dancing for

healthy older adults: A review. Journal of Aging and Physical Activity, 17(4), 479–500.

Lord, S., Matters, B., St. George, R., Thomas, M., Bindon, J., KamYin Chan, D., et al. (2006). The effects

of water exercise on physical functioning in older people. Australasian Journal of Ageing, 25(1),

36–41.

McKinley, P., Jacobson, A., Leroux, A., Bednarczyk, V., Rossignol, M., & Fung, J. (2008). Community-

based Argentine tango dance program improves functional balance and confidence in older people.

Journal of Aging and Physical Activity, 16(4), 435–453.

Shigematsu, R., Chang, M., Yabushita, N., Sakai, T., Nakagaichi, M., Nho, H., & Tanaka, K. (2002).

Dance-based aerobic exercise may improve indices of falling risk in older women. Age and Ageing,31(4), 261–266.

Sofianidis, G., Hatzitaki, V., Douka, S., & Grouios, G. (2009). Effect of a 10-week traditional dance

program on static and dynamic balance control in elderly adults. Journal of Aging and PhysicalActivity, 17(2), 167–180.

Verghese, J. (2006). Cognitive and mobility profile of older social dancers. Journal of the AmericanGeriatrics Society, 54(8), 1241–1244.

Winter, D. A. (2005). Biomechanics and motor control of human movement. Hoboken, NJ: John Wiley &

Sons.

Wolfson, L. (1996). Balance and strength training in older adults: Intervention gains and Tai-Chi

maintenance. Journal of American Geriatrics Society, 44, 498–506.

Am J Dance Ther

123

Zhang, J. G., Ishikawa-Takata, K., Yamazaki, H., Morita, T., & Ohta, T. (2006). The effects of Tai Chi

Chuan on physiological function and fear of falling in the less robust elderly: An intervention study

for preventing falls. Archives of Gerontology and Geriatrics, 42(2), 107–116.

Author Biographies

Azucena Guzman-Garcıa BA, MRes, PhD CandidateMs. Guzman-Garcıa is a PhD candidate in behaviour and affective states in dementia from the Institute

for Ageing and Health at Newcastle University, UK. She studied psychology at the Universidad de las

Americas, A.C. in Mexico City and she obtained her master’s degree in Neuropsychology and

Rehabilitation from The University of Birmingham, in the UK. Ms. Guzman-Garcıa, a trained Latin

Ballroom dancer, has studied psychomotor practice at the Institut Superieur de Reeducation

Psychomotrice in Paris, France.

Leif Johannsen, Dr. rer. nat, Dipl-PsychDr. Johannsen is a researcher at the Technische Universitat Munchen (Munich, Germany), and the

University of Birmingham, in the UK. His psychology degree was awarded by the University of Potsdam

in 1999. He completed his doctorate in behavioural neuroscience at the University of Tubingen

(Germany) in 2005. His research focuses on effects of light touch contact between two individuals, and on

the interpersonal coordination of movements, posture, and balance.

Alan M. Wing, PhDProfessor Wing studied Physics and Psychology as an undergraduate at Edinburgh University, in the UK.

After completing a PhD in Psychology on timing of movement at McMaster University in Ontario,

Canada, he continued this research in a postdoctoral fellowship at Bell Labs in New Jersey. Professor

Wing then joined the Medical Research Council staff at the Cambridge Applied Psychology Unit where

he studied movement timing as well as anticipatory control of posture in balance and grip. In 1997 he

became Professor of Human Movement in the School of Psychology at The University of Birmingham

(UK) where he set up the Sensory Motor Neuroscience group to focus on sensory factors in control of

posture and movement.

Am J Dance Ther

123