Cognitive and Mobility Profile of Older Social Dancers

Joe Verghese, MD

OBJECTIVES: To define the cognitive and physical at-tributes of regular social dancing so as to help establish itshealth benefits and help plan future dance interventionsto prevent adverse outcomes in older adults such as falls,slow gait, and dementia.

DESIGN: Cross-sectional survey with two-group compar-ison.

SETTING: Bronx County, New York.

PARTICIPANTS: Twenty-four cognitively normal oldersocial dancers (OSDs) were compared with 84 age-, sex-,and education-matched older nondancers (ONDs) partic-ipating in a community-based study.

MEASUREMENTS: Motor and cognitive performancewas assessed using validated clinical and quantitativemethods.

RESULTS: There were no differences in the frequency ofparticipation in other cognitive and physical leisure activ-ities, chronic illnesses, or falls between OSDs and ONDs.Cognitive test performance was not different betweenOSDs and ONDs. OSDs had better balance but notstrength than ONDs. OSDs had a longer meanstride � standard deviation than ONDs (117.8 � 10.5 cmvs 103.4 � 20.2 cm, P 5.008) on quantitative gait assess-ment, with a more stable pattern during walking with re-duced stance time (63.9% vs 65.9%, P 5.01), longer swingtime (36.1% vs 34.1%, P 5.01), and shorter double sup-port time (27.9% vs 30.9%, P 5.03).

CONCLUSION: The results of this study suggest thatlong-term social dancing may be associated with betterbalance and gait in older adults. J Am Geriatr Soc 54:1241–1244, 2006.

Key words: dance; elderly; balance; gait; cognition

Although social dancing is a popular form of recreationfor seniors, there is little information on its long-term

physical and mental benefits.1 Systematic gait and cognitiveassessments of social dancers are lacking.1 The limited re-

search suggests that regular dancing may improve balanceand strength, decrease risk of falls, and prevent cognitivedecline.1–5

Although regular dancers have good body rhythmwhile dancing, it does not necessarily follow that socialdancing will translate to a better gait pattern during normalwalking. Moreover, it is uncertain whether social dancinghas any effect on specific or global measures of gait or cog-nition.

Cognitive and motor performance of nondisabled cog-nitively normal older social dancers (OSDs) participating ina community-based study was compared with that of age-,sex-, and education-matched cognitively normal older non-dancers (ONDs) in this study. Based on prior research,1–5 itwas hypothesized that OSDs would have better gait, bal-ance, and cognition than ONDs. Defining the cognitive andphysical attributes of regular social dancing will help es-tablish its health benefits and help plan future dance inter-ventions to prevent adverse outcomes in older adults, suchas falls, slow gait, and dementia.

METHODS

Study Population

Subjects for the current study were identified from nondis-abled community-residing participants in a longitudinalgait and mobility study.6 The aim of the study was to iden-tify clinical markers of disability. The inclusion criteria wereage 70 and older and living in Bronx County. Exclusioncriteria were presence of disability (defined as inability toperform one or more of the following activities of dailyliving: bathing, dressing, grooming, feeding, toileting,walking around the home, and getting up from a chair)and inability to follow instructions for study protocols.Twenty-five OSDs were identified over a 1-year period(September 2004 to August 2005). One OSD with clinicaldementia was excluded. The 24 OSDs were matched by age(� 6 months), sex, and education (� 1 year) with 84ONDs evaluated over the same study period with the samestudy protocols. The local institutional review board ap-proved the study protocol. All subjects signed informedconsents.

Dance

At study visits, research assistants documented participa-tion in physical and mental leisure activities, including so-

Address correspondence to Joe Verghese, MD, Einstein Aging Study, AlbertEinstein College of Medicine, 1165 Morris Park Avenue, Bronx, NY 10461,USA, E-mail: jverghes@aecom.yu.edu

DOI: 10.1111/j.1532-5415.2006.00808.x

From the Department of Neurology, Albert Einstein College of Medicine,Bronx, NY.

JAGS 54:1241–1244, 2006r 2006, Copyright the AuthorsJournal compilation r 2006, The American Geriatrics Society 0002-8614/06/$15.00

cial dancing. The questionnaire was based on a validatedleisure activity scale.5 Subjects were asked about thenumber of days per week they danced and the number ofyears they had been dancing.

Cognitive Assessments

The tests included in the neuropsychological test batteryhave been validated for use in the normal aging population.Performance on tests of cognitive domains that exercisemay influence was examined.

General Mental Status

The Blessed Information Memory Concentration test,which was used to assess general mental status,7 has hightest-retest reliability and correlates well with Alzheimer’spathology.

Memory

The Free and Cued Selective Reminding Test is a widelyused and validated test of episodic memory.8

Executive Function

A number of cognitive processes are grouped under the ru-bric of executive function. Performance on the Digit Span,Digit Symbol, and Block Design Tests from the WechslerAdult Intelligence ScaleFrevised was examined.9 Perfor-mance on the Verbal Fluency Test10 and Trail MakingTests11 was also examined.

Depressive Symptoms

Depressive symptoms were assessed using the Geriatric De-pression Scale.12

Quantitative Gait

Research assistants conducted quantitative gait evalua-tions, independent of the clinician’s evaluation, using acomputerized gait mat (180 � 35.5 � 0.25 inches) withembedded pressure sensors (GAITRite, CIR Systems, Ha-vertown, PA).6 Subjects were asked to walk on the mat attheir normal walking speed for two trials in a quiet, well-lithallway. Start and stop points were marked using whitelines on the floor and included 3 feet each for initial accel-eration and terminal deceleration. Monitoring devices werenot attached to the participants during the test. The soft-ware computes quantitative parameters (see Appendix 1 fordefinitions) based on footfalls recorded. Each trial was onewalkway in length, and values analyzed were the mean oftwo trials computed by the software. Excellent reliabilityand validity for GAITRite assessments were reported inprevious research.6,13

Motor

Subjects were examined using the following validated testsof strength and balance.

Balance subtest of the Physical Performance Battery(PPB):14 participants were scored 1 if they could holda side-by side stand for 10 seconds but were unable tohold a semitandem stand for 10 seconds, 2 if they helda semitandem stand for 10 seconds but were unable to

hold a full tandem stand for more than 2 seconds, 3 ifthey held the full tandem stand for 3 to 9 seconds, and4 if they held the full tandem stand for 10 seconds.

Chair rise test from the PPB:14 time taken (seconds) tocomplete five chair rises. The chair rise is a test oflower extremity strength and balance.14,15

Unipedal stance (seconds): this test measures ability tostand on one foot, selected by subject, for a maximumof 30 seconds.

Grip strength (kg of force) was measured using adynamometer.

Statistical Analysis

Baseline characteristics of eligible OSDs and ONDs, as wellas their performance on cognitive and motor tests, werecompared using descriptive statistics. Categorical variableswere compared using the chi-square test, applying the Fish-er test as appropriate. Continuous variables were comparedusing independent sample t tests. All tests were two-tailed,with an alpha level of 5%.

RESULTS

Study Population

Of the 24 OSDs, eight were men and 16 women. The meanage of OSDs was 80.0 and of ONDs was 80.8. The OSDdanced for a mean � standard deviation of 4.3 � 3.0 daysper month (range 1–12 days). The mean duration of socialdancing was 36.5 � 26.5 years (range 3–75 years, median30 years). The types of dancing included ballroom dancing(n 5 10), line dancing (n 5 3), swing dancing (n 5 3), squaredancing (n 5 1), and unspecified (n 5 7). Of the 24 OSDs,three (12.5%) rated their dancing skills as expert, 10(41.7%) as intermediate, three (12.5%) as beginner; eight(30.3%) did not rate themselves. None of the OSDs wereprofessional or competitive dancers. Other than dancing,there were no significant differences in the frequency ofparticipation in other cognitive (watching television, read-ing newspapers, writing for pleasure, playing board or cardgames, doing puzzles, and playing musical instruments) andphysical (brisk walking, aerobic exercise, swimming, yoga,weight training, gardening, and light housework) leisureactivities between OSDs and ONDs. Table 1 confirms bal-anced age, sex, and education levels of OSDs and ONDs.There were no differences in prevalence of chronic illnessand falls between the two groups.

Cognition

Table 1 shows that there were no significant differencesin performance on tests of general and specific cognitivestatus, including executive function.

Motor

OSDs had better balance than ONDs, measured using thePPB and unipedal stance (Table 1), although lower extrem-ity (chair rise) and upper extremity (grip) strength measureswere not different. OSDs walked faster than ONDs. OSDshad longer steps and strides than ONDs, which resulted inshorter stance time, longer swing time, and shorter doublesupport time, reflecting a more stable walking pattern

1242 VERGHESE AUGUST 2006–VOL. 54, NO. 8 JAGS

(Table 2). There were no significant group differences ingait and cognitive parameters when the group was stratifiedby sex (data not shown).

DISCUSSION

This study shows that OSDs have better balance and gaitpatterns but not strength than ONDs. These findings inregular social dancers corroborate previous studies thathave reported improvements in balance in older adults fol-

lowing short-term dance interventions.3,4,16 Regular danc-ing may improve balance in several ways. Older adultsattempt to improve stability by shortening their step lengthand increasing their double-support time,17,18 although thisis not a stable gait pattern and is associated with greater riskof falls.19 Quantitative gait testing in this study revealedthat OSDs walked faster but took longer steps (and strides)than ONDs. They spent less time in the stance phase andmore time in the swing phase, which reflects a more stablegait pattern.17–19 Although OSDs had a lower prevalence offalls than ONDs, the differences were not significant in thissmall sample. Older adults have more motor tone whilestanding and may overreact to balance perturbations.1

Regular dancing may modify this unsafe response. Dancingmay also improve cardiovascular fitness. Findings in thisand other studies20,21 also raise the possibility that increas-ing step or stride length in older adults could be examined asan intermediate study endpoint in rehabilitative interven-tions to improve balance and reduce risk of falls.

Professional dancers are reported to have better bal-ance and may show different postural adaptation to balanceperturbations than nondancers.1,22,23 For instance, dancersrely more on proprioception and less on visual input thannondancers.23 Ballet dancers are trained to have a stablevisual reference.23 Even in professional dancers, patterns ofmuscle activation in response to postural changes differdepending on the type of dancing.24 Professional dancersreceive longer and more intensive training than OSDs, and

Table 1. Comparison of Demographic, Cognitive, and Balance Status of Older Social Dancers (OSDs) and Older Non-dancers (ONDs)

Variables OSDs ONDs P-value

Age, mean � SD 80.0 � 6.5 80.8 � 4.9 .54Female, % 70.1 69.7 .82High school or higher, % 88.4 87.5 .87Medical illnesses, %

Diabetes mellitus 20.8 10.3 .18Hypertension 41.6 33.3 .47Coronary artery disease 8.0 5.8 .10Osteoarthritis 37.5 35.6 .99Rheumatoid arthritis 0.0 1.1 .99Stroke 8.3 9.2 .63Falls 4 11 .29

Cognitive tests, mean � SDBlessed test score 2.0 � 1.9 2.0 � 2.0 .92Free and Cued Selective Reminding Test score 47.9 � 0.3 47.9 � 0.8 .79Block design score 23.1 � 8.5 23.1 � 8.9 .97Digit span total 14.5 � 2.9 14.9 � 3.5 .52Digit Symbol Substitution score 45.3 � 16.9 46.1 � 14.2 .75Verbal Fluency Test score 39.1 � 14.0 37.8 � 13.4 .67Trail Making Test A, seconds 54.3 � 19.0 54.9 � 20.0 .95Trail Making Test B, seconds 146.5 � 74.7 165.4 � 114.4 .95Geriatric Depression Scale score 1.9 � 1.7 2.1 � 1.8 .87

Balance, mean � SDBalance score (range 0–4) 3.7 � 0.6 3.3 � 0.9 .03Unipedal stance, seconds 12.2 � 10.1 7.2 � 8.2 .04Chair rise, seconds 12.8 � 4.0 14.4 � 6.6 .36Grip strength, kg 24.0 � 7.7 21.7 � 6.8 .17

SD 5 standard deviation.

Table 2. Comparison of Quantitative Gait Parameters Dur-ing Normal Walking of Older Social Dancers (OSDs) andOlder Nondancers (ONDs)

Variable

OSDs ONDs

P-valueMean � Standard Deviation

Velocity, cm/s 100.5 � 18.7 87.2 � 23.6 .02Cadence, steps/min 101.9 � 12.9 99.9 � 14.3 .94Step length, cm 58.6 � 5.9 52.3 � 10.2 .01Stride length, cm 117.8 � 10.5 103.4 � 20.2 .008Swing, % 36.1 � 1.9 34.1 � 2.9 .01Stance, % 63.9 � 1.9 65.9 � 2.9 .01Double support, % 27.9 � 3.3 30.9 � 5.0 .03Support base, cm 8.9 � 3.1 10.6 � 4.3 .16

Note: Gait parameters are for the right side.

OLDER SOCIAL DANCERS 1243JAGS AUGUST 2006–VOL. 54, NO. 8

it is not known whether regular social dancing will result inthe same postural adaptations seen in professional dancers.

It was previously reported that older adults whodanced frequently had a lower risk of developing demen-tia than those who rarely danced or never danced,5 butcognitive test performance was not significantly differentbetween OSDs and ONDs in this study. Both groups hadcomparable levels of participation in leisure activities otherthan dancing, which may also minimize cognitive differ-ences. Matching was done by age, sex, and education, all ofwhich have been shown to be strongly associated with riskof dementia. Matching by these variables may, hence, re-duce cognitive differences between OSDs and ONDs.

Limitations of this preliminary study include its cross-sectional design, which does not permit inferences on cau-sality, the small sample size, and variable amount of socialdancing in the OSD group. Despite comparable levels ofphysical and cognitive leisure activities, OSDs had bettergait and balance, suggesting activity-specific benefits. Com-parisons of OSDs with other control groups such as inactiveolder peers, master athletes, or chess players in future stud-ies will help further define specific benefits of this activity.Although there were no group differences in disease sever-ity, and matching was done on key variables, the possibilitythat the association between dancing and physical perfor-mance might in part reflect unmeasured disease severity orresidual confounding by unmeasured variables rather thanthe benefits of dancing cannot be discounted.

Social dancing is a popular and accessible form of rec-reation in senior centers.1 The long-term effects of regularsocial dancing on adverse outcomes such as mobility, falls,and cognitive decline in older adults should be addressed inthe context of future clinical trials and longitudinal studies.If other studies corroborate these findings, social dancingcould be examined as a feasible and enjoyable way to en-courage sedentary older adults to increase physical activityto improve their heath.1

ACKNOWLEDGMENTS

I would like to thank Gail Kuslansky, Adina Cappell, andMichael Potenza for assistance with developing the data-base.

Financial Disclosure: Dr. Verghese is supported by aPaul B. Beeson Career Development Award (NIA-K23AG024848) and National Institute on Aging Grant RO1AGO25119.

Author Contributions: Dr. Verghese was responsiblefor developing the study design, acquisition of data, anal-ysis and interpretation of data, and initial draft of the man-uscript.

Sponsor’s Role: The sponsor had no role in the design,methods, subject recruitment, data collections, analysis, orpreparation of the manuscript.

REFERENCES

1. Judge JO. Balance training to maintain mobility and prevent disability. Am

J Prev Med 2003;25:150–156.

2. Crotts D, Thompson B, Nahom M et al. Balance abilities of professional

dancers on select balance tests. J Orthop Sports Phys 1996;23:12–17.

3. Hopkins DR, Murrah B, Hoeger WW et al. Effect of low-impact aerobic dance

on the functional fitness of elderly women. Gerontologist 1990;30:189–192.

4. Shigematsu R, Chang M, Yabushita N et al. Dance-based aerobic exercise may

improve indices of falling risk in older women. Age Ageing 2002;31:261–266.

5. Verghese J, Lipton RB, Katz MJ et al. Leisure activities and the risk of dementia

in the elderly. N Engl J Med 2003;348:2508–2516.

6. Verghese J, Katz MJ, Derby CA et al. Reliability and validity of a telephone-

based mobility assessment questionnaire. Age Ageing 2004;33:628–632.

7. Blessed G, Tomlinson E, Roth M. The association between quantitative meas-

ures of dementia and of senile change in the cerebral grey matter of elderly

subjects. Br J Psychiatry 1968;114:797–811.

8. Buschke H. Cued recall in amnesia. J Clin Neuropsychol 1984;6:433–440.

9. Wechsler D. Manual for the Wechsler Adult intelligence Scale. Washington,

DC: Psychological Corporation, 1955.

10. Benton AL, Hamsher K, Sivan AB. Multilingual Aphasia Examination, 3rd Ed.

Iowa City, IA: AJA Associates, 1983.

11. Reitan RM. Trail making test results for normal and brain-damaged children.

Percept Mot Skills 1971;33:575–581.

12. Yesavage JA, Brink TL, Rose TL et al. Development and validation of a ger-

iatric depression rating scale: A preliminary report. J Psychiatr Res 1982;

17:27.

13. Bilney B, Morris M, Webster K. Concurrent related validity of the GAITRite

walkway system for quantification of the spatial and temporal parameters of

gait. Gait Posture 2003;17:68–74.

14. Guralnik JM, Simonsick EM, Ferrucci L et al. A short physical performance

battery assessing lower extremity function: Association with self-reported dis-

ability and prediction of mortality and nursing home admission. J Gerontol

1994;49:85–94.

15. Schenkman M, Hughes MA, Samsa G et al. The relative importance of

strength and balance in chair rise by functionally impaired older individuals.

J Am Geriatr Soc 1996;44:1441–1446.

16. Emery CF, Hsiao ET, Hill SM et al. Short-term effects of exercise and music on

cognitive performance among participants in a cardiac rehabilitation program.

Heart Lung 2003;32:368–373.

17. Judge JO, Ounpuu S, Davis RB. Effects of aging on the biomechanics and

physiology of gait. Clin Geriatr Med 1996;12:659–677.

18. Winter DA, Patla AE, Frank JS et al. Biomechanical walking pattern changes in

the fit and healthy elderly. Phys Ther 1990;70:340–347.

19. Maki BE. Gait changes in older adults: Predictors of falls or indicators of fear.

J Am Geriatr Soc 1997;45:313–320.

20. Protas EJ, Mitchell K, Williams A et al. Gait and step training to reduce falls in

Parkinson’s disease. Neurorehabilitation 2005;20:183–190.

21. van Loo MA, Moseley AM, Bosman JM et al. Test-re-test reliability of walking

speed, step length and step width measurement after traumatic brain injury:

A pilot study. Brain Inj 2004;18:1041–1048.

22. Stretanski MF, Weber GJ. Medical and rehabilitation issues in classical ballet.

Am J Phys Med Rehabil 2002;81:383–391.

23. Golomer E, Cremieux J, Dupui P et al. Visual contribution to self-induced

body sway frequencies and visual perception of male professional dancers.

Neurosci Lett 1999;267:189–192.

24. Trepman E, Gellman RE, Micheli LJ et al. Electromyographic analysis of

grand-plie in ballet and modern dancers. Med Sci Sports Exerc 1998;30:1708–

1720.

Appendix 1. Definition of Gait Parameters

Velocity (cm/s) is the distance covered on two trials divided byambulation time.Step length is measured from the heel point of the currentfootfall to the heel point of the previous footfall on the oppositefoot.Cadence is the number of steps taken in a minute.Stride length is the distance between the heel points of twoconsecutive footfalls of the same foot (left to left, right to right).Double support is the time elapsed between first contact of thecurrent footfall and the last contact of the previous footfall,added to the time elapsed between the last contact of thecurrent footfall and the first contact of the next footfall.Support base is the perpendicular distance from heel point ofone footfall to the line of progression of the opposite foot.

1244 VERGHESE AUGUST 2006–VOL. 54, NO. 8 JAGS