Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 10 , Pages 1309-1313, October 2007

The Effect of a Dual-Task on Obstacle Crossing in Healthy Elderly and Young Adults

Presented in part to the Society for Neuroscience, November 2003.

  • Hyeong-Dong Kim, PhD

      Affiliations

    • Department of Physical Therapy, Catholic University of Daegu, South Korea
    • ,
  • Denis Brunt, EdD

      Affiliations

    • Department of Physical Therapy, East Carolina University, Greenville, NC.
    • Corresponding Author InformationCorrespondence to Denis Brunt, EdD, Dept of Physical Therapy, School of Allied Health Sciences, East Carolina University, Greenville, NC 27858-4353

Article Outline

Abstract 

Kim H-D, Brunt D. The effect of a dual-task on obstacle crossing in healthy elderly and young adults.

Objective

To investigate the effect of a dual-task on step initiation over an obstacle.

Design

Repeated-measures design between groups.

Setting

University laboratory.

Participants

Ten healthy, community-dwelling elderly adults and 10 healthy young adults.

Interventions

Not applicable.

Main Outcome Measures

The effect of a reaction time task on ground reaction forces, toe clearance, and temporal events in stepping over an obstacle.

Results

Ground reaction forces of the swing limb before toe-off did not differ between the groups and was not affected by task difficulty. Stepping with a random stimulus after toe-off was the most difficult task, whereas stepping with a predictable stimulus before toe-off was the easiest task. Reaction time and stepping time were greater and toe clearance was less for the elderly subjects. Both groups had a decrease in toe clearance and an increase in reaction time and stepping time as the task became more difficult.

Conclusions

Our findings indicate that even healthy older adults may be at risk for falls in situations where they are engaged in concurrent tasks. The data support the inclusion of dual-task activities in fall prevention programs.

Key Words: Accidental falls, Aging, Gait, Reaction time, Rehabilitation

 

WHEN 2 TASKS ARE performed simultaneously, “attention capacity”1, 2, 3 must be apportioned effectively. The difficulty of these tasks, or how tasks are prioritized, will influence how attention is partitioned. If, however, the combined difficulty of the tasks requires excessive attention, then interference between tasks could occur. That is, the quality of performance of both tasks could decrease, or 1 task be performed in preference to the other. This “dual-task” paradigm has become an important focus of research in elderly populations. The effect on balance of a secondary task has been of particular interest because of the risk of falls and their subsequent medical complications. Older adults have increased postural sway or loss of balance and take compensatory steps when presented with a secondary task during quiet stance.4, 5, 6, 7, 8

Melzer and Oddsson9 extended the dual-task paradigm from quiet stance to the initiation of a single step. Subjects stepped as quickly as possible in response to a cutaneous stimulus. In general, time to initiate the step and temporal phases of step initiation increased for older subjects when performed simultaneously with a cognitive task. In this study, we asked healthy young and elderly subjects to not just initiate stepping from quiet stance, but to step over an obstacle while simultaneously performing a reaction-time task.

Such an investigation appears to be important for 2 reasons. First, tripping over an obstacle is among the most commonly reported causes of falls in the elderly.10, 11, 12, 13, 14 Second, there are only small differences in stepping over an obstacle from quiet stance between young and healthy elderly15 and between healthy and frail elderly.16 Therefore, challenging the attention capacity of the elderly while stepping may reveal differences that place them at risk for tripping and possibly falling. In addition, increasing the difficulty of a task during walking also increases attention demands.17 We therefore introduced 2 levels of task difficulty into the protocol. First, presentation of the reaction time stimulus was either predictable or randomized. Second, the stimulus cue was presented at different levels of stability, that is, either in single- or double-limb stance.

In this study, the predicted task-dependent consequences of the dual-task paradigm were a decrease in toe clearance over an obstacle, an increase in temporal parameters of stepping, and an increase in reaction time.

Back to Article Outline

Methods 

Participants 

Ten healthy, community-dwelling elderly adults (mean age, 74.6±5.0y) and 10 healthy, younger adults (mean age, 27.7±4.1y) participated in the study. Criteria for inclusion of the elderly subjects was a score greater than 50 on both the Berg Balance Scale18, 19 and Frenchay instrumental activities of daily living,20 and greater than 20 on a physical function test.21 Elderly participants reported no falls in the previous 12 months and scored greater than 24 on the Mini-Mental State Examination.22 The reliability and validity of these tests have been reported as satisfactory.23, 24, 25, 26 The study was approved by the university institutional review board and all participants signed an informed consent form before their participation.

Experimental Procedures 

Subjects stood with each foot on a force platforma and, after receiving a visual cue, stepped over a 10-cm high obstacle at a self-paced speed with the right (swing) limb. The visual cue was a light-emitting diode placed in the center of the obstacle. Subjects were instructed to continue walking after stepping over the obstacle. Lines were marked on the forceplate to ensure that foot placement remained constant for all subjects. A mat switch detected heel-strike of the swing limb. Amplified force platformb signals were sampled on-line at a rate of 1000Hz for 5 seconds. Reflective markers were placed on the head of the fifth metatarsal of the right stepping limb and the front edge of the obstacle. The vertical distance between the markers provided an estimate of toe clearance. Motion datac were captured at 100Hz for 5 seconds. Figure 1 illustrates the experimental setup.

Subjects first completed reaction time trials while standing and trials of baseline stepping. This was followed by stepping trials in which subjects were asked to press a handheld microswitch in response to a threshold cutaneous stimulus (reaction time task). The stimulus was a single 1-ms square wave pulse delivered to 1 arm from a stimulator and stimulus isolation unit.d The cutaneous stimulus was presented either before and after toe-off with the swing limb. In addition, the trials were blocked so that the stimulus was presented either at every trial or randomly during the trials. After the practice trials, the subjects completed approximately 100 successful experimental trials. Rest periods were provided to avoid fatigue.

Statistical Analysis 

All trials of all subjects were averaged for each condition; we used analysis of variance techniques to determine main and interaction effects. We used single degree of freedom mean contrasts to determine significant comparisons (P<.05). The independent variables were the 2 groups (young and healthy elderly adults) and 5 levels of stepping conditions (baseline and random or predictable stimulus trials before and after toe-off). Dependent measures included timing, slope, and amplitude measures of the anteroposterior ground reaction force (Fx) under the stance and swing limbs (normalized to body weight [BW]), time to swing limb toe-off, and swing time. Timing data were referenced from movement onset, defined as the first detectable onset of force platform activity (Fx), as determined visually with an interactive cursor of 1-ms resolution. Simple reaction time was analyzed from the onset of the stimulus to the subject’s response. Toe clearance was assessed as the perpendicular distance from the marker placed on the metatarsal head to the marker on the obstacle. Figure 2 shows data from an individual trial by a young subject that illustrate the dependent variables.

  • View full-size image.
  • Fig 2. 

    Anteroposterior (Fx) ground reaction force dependent variables of a trial for a young subject, where the cutaneous stimulus was presented at every trial and before toe-off. The vertical arrows represent time of stimulus and reaction time (RT). Vertical lines represent movement onset, swing toe-off, swing heel-strike, and stance toe-off. Dependent measures of the stance and swing limb ground reaction forces are indicated. In this trial, reaction time was 360ms and peak shear force (Fx) of the stance limb was 57% of BW.

Back to Article Outline

Results 

The timing of the cutaneous stimulation after movement onset for both the predictable and randomized trials before swing limb toe-off was 285±63ms and 292±44ms, respectively. Corresponding times of stimulation after toe-off were 449±75ms and 493±111ms, respectively. The timing of the stimulus presentation between predictable and randomized trials did not differ significantly. Previous research has shown that Fx of the swing limb correlates significantly with the velocity of step initiation.27, 28 There were no main effects for peak Fx or the slope and time to peak Fx. The velocity of the initiation of stepping, therefore, was not influenced by the secondary task.

Group Effects 

Toe clearance was greater for the younger adults than for the older adults (14.1±0.65cm and 12.6±0.86cm, respectively) (F1,18=4.97, P<.05), and reaction time of the elderly (418±104ms) was significantly slower than the young adults (297±68ms) (F1,18=8.61, P<.01). Stance peak Fx was significantly greater for younger adults (46%±133% of BW) than for the older adults (26%±12% of BW) (F1,18=12.67, P<.01). Stance peak Fx coincides with swing toe-off and will increase with a faster lift off.15 Time to swing toe-off was significantly earlier for the young than for the older adults (383±25ms, 435±40ms, respectively) (F1,18=5.44, P<.05). Swing time was also significantly shorter for the young than for the older adults (533±31ms, 621±38ms, respectively) (F1,18=4.78, P<.05).

Condition Effects 

There was a condition effect for toe clearance (F4,64=8.17, P<.01). It was greater for baseline stepping (14.6±0.78cm) versus stepping with the reaction time tasks (combined mean, 13.0±0.96cm). In addition, toe clearance for predictable trials before toe-off (13.5±1.2cm) was significantly greater than for the random trials after toe-off (12.6±1.31cm). The data were similar for reaction time (F4,72=36.93, P<.001), where baseline reaction time (219±42ms) was significantly less than the dual-task conditions (combined mean, 393±83ms), and reaction time was less for predictable trials before toe-off (338.5±86ms) versus random trials after toe-off (442.5±116ms).

Stance peak Fx was significantly greater when the stimulus was provided at every trial (38%±16% of BW before toe-off, 39%±18% of BW after toe-off), compared with respective random trials (34%±18% of BW before toe-off, 34%±19% of BW after toe-off) (F4,72=2.72, P<.05). As would be expected, time to swing toe-off was significantly earlier when the stimulus was provided for every trial (361±113ms before toe-off, 393±74ms after toe-off), compared with respective random trials (441±76ms before toe-off, 425±95ms after toe-off) (F4,18=3.06, P<.05). Further, there were significant condition effects for swing time (F4,18=5.78, P<.001). Similar to the swing toe-off, swing time was less when the stimulus was provided at every trial (533±72ms before toe-off, 556±36ms after toe-off), compared with respective random trials (583±62ms, 601±57ms).

Degree of Task Difficulty 

These data imply a degree of difficulty with respect to the presentation of the independent variables. That is, a certain combination of the levels of stepping condition had greater influence on the task-dependent consequences. For example, Fig 3, Fig 4 show the mean data for reaction time and toe clearance where a predictable cue before toe-off is the least difficult task and a random cue after toe-off is the most difficult task. That is, toe clearance decreased and reaction time increased when the secondary reaction time task was randomly presented during single-limb stance. It is reasonable to assume that the degree of difficulty of a random cue before toe-off is similar to a predictable cue after toe-off.

  • View full-size image.
  • Fig 3. 

    Reaction time for baseline and dual-task conditions. The dual-task stimulus was presented at every trial and random trial both before and after toe-off. Abbreviations: EA, every trial after toe-off; EB, every trial before toe-off; RA, random trials after toe-off; RB, random trials before toe-off.

Back to Article Outline

Discussion 

In this study, we investigated the effects of a secondary reaction time task when stepping over an obstacle from a position of a quiet stance in healthy young and older adults. We hypothesized that the performance of the stepping task (toe clearance) and/or the reaction time task would decline with presentation of the secondary task, especially in older adults. The presentation of the reaction time task was predictable or random and during single- or double-limb stance. Reaction time and swing time increased and toe clearance decreased as the task became more difficult. Random presentation of the reaction time stimulus after swing limb toe-off appeared to be the most difficult stepping condition. Regardless of when the reaction time stimulus was presented, the components of the ground reaction force to peak Fx of the stepping limb did not differ between stimulus conditions and between young and older adults. This is important because it eliminates the potential influence of the velocity of step initiation on reaction time, toe clearance, and other task parameters.

Although this study demonstrated significant group differences in toe clearance and reaction time, we did not detect any group-by-condition interaction. That is, group-related differences in toe clearance, temporal events after swing toe-off, and reaction time remained similar as task parameters were manipulated to make the task more demanding. Nevertheless, toe clearance was consistently less and reaction time and swing time greater for the elderly compared with young adults. These data are consistent with previous findings in dual-task experiments in which older adults had increased reaction time and decreased performance in gait or postural control paradigms.29, 30, 31, 32

The notion of task-appropriate scaling of toe clearance during obstacle crossing is crucial to ensuring a safe and efficient obstacle crossing. Falls are more likely to occur with inadequate toe clearance rather than through contacting the obstacle with the heel.33 It has been reported that the toe clearance over an obstacle ranges from approximately 9 to 13cm for either young or healthy elderly subjects.34 Our study showed a mean toe clearance ranging from 11.8 to 15.1cm. Not only is swing limb toe clearance essential for safely crossing an obstacle, there also must be a stable base of support provided by the stance limb. The older subjects demonstrated significantly longer swing time (and therefore, longer single-limb stance time) than the younger subjects. Stance peak Fx was significantly less in the older adults than in the young adults, a finding that suggests older adults modulated the duration and magnitude of swing limb lift off differently than did the young adults.15, 16 Swing and stance time were longer for the older adults and resulted in a slower stepping speed. Using a longer swing and stance time may reflect the fact that older adults tend to use a more cautious strategy in this test condition.

These data suggest that both the timing and the uncertainty of the presentation of the secondary reaction time task influenced toe clearance. The notion of task difficulty may therefore be significant in placing the elderly at risk for falls. Our findings suggest that an increase in risk for falls occurs when an attention-demanding task is unexpected when a subject is in a posture of decreased stability. In fact, a low level of task difficulty was a plausible explanation for there being no change in foot clearance in elderly subjects stepping over an obstacle while walking.35 There was only a small decrement in the secondary task, which was continuous throughout the walk. Analysis of a single subject, however, where the change in the secondary task was significant revealed definite changes in gait-related variables. Other studies have reported a performance decrease in the secondary but not the primary task. For example, Brown et al36 asked subjects to step over a foam obstacle while walking. A secondary reaction time task was presented before, or while, stepping over the obstacle. They reported an increase in reaction time for both old and young subjects before crossing the obstacle and for older subjects while crossing the obstacle. No subjects hit the obstacle, but the authors did not provide any performance data on actual obstacle clearance. Reaction time was also increased for older adults when heel strike was constrained to a specific target while walking.17 In both the Brown36 and Sparrow17 studies, manipulation of the reaction time signal altered the degree of difficulty of the task; however, neither study reported the potential impact of task difficulty on the consequences of the primary task.

Back to Article Outline

Conclusions 

The results of this study provide an increased understanding of the interaction of age and dual tasking in stepping over obstacles. Many distractions encountered in daily life are unexpected and can reduce attention to stepping mechanics. Our findings indicate that, depending on the difficulty of the task, even healthy older adults may be at risk for falls. Several clinical assessment tools have been devised that include a dual-task component. These tests have been effective in identifying people with a history of falls and/or are at a greater risk of falling.37, 38, 39, 40 This study, therefore, seems to support the inclusion of dual-task activities in fall prevention programs. Finally, our results raise 2 interesting questions for future study: (1) whether the group differences we found in this study would be exaggerated under more demanding or dynamic conditions than those we investigated, and (2) whether frail and balance-impaired older adults would have more difficulty performing the same tasks than would healthy older adults.

Suppliers

Back to Article Outline

References 

  1. Moray N. Where is attention limited? (A survey and a model). Acta Psychol (Amst). 1967;27:84–92
  2. Keele SW. Attention and human performance. Pacific Palisades: Goodyear; 1997;
  3. Wickens C. Engineering psychology and human performance. 2nd ed.. New York: Harper Collins; 1992;
  4. Brown LA, Shumway-Cook A, Woollacott MH. Attentional demands and postural recovery: the effects of aging. J Gerontol A Biol Sci Med Sci. 1999;55:M165–M171
  5. Maylor EA, Allison S, Wing AM. Effects of spatial and nonspatial cognitive activity on postural stability. Br J Psychol. 2001;92:319–338
  6. Maylor EA, Wing AM. Age differences in postural stability are increased by additional cognitive demands. J Gerontol B Psychol Sci Soc Sci. 1996;51:143–154
  7. Shumway-Cook A, Woollacott M, Kerns KA, Baldwin M. The effects of two types of cognitive tasks on postural stability in older adults with and without a history of falls. J Gerontol A Biol Sci Med Sci. 1997;52:M232–M240
  8. Teasdale N, Bard C, LaRue J, Fleury M. On the cognitive penetrability of posture control. Exp Aging Res. 1993;19:1–13
  9. Melzer I, Oddsson LI. The effect of a cognitive task on voluntary step execution in healthy elderly and young individuals. J Am Geriatr Soc. 2004;52:1255–1262
  10. Blake AJ, Morgan K, Bendall MJ, et al. Falls by elderly people at home: prevalence and associated factors. Age Ageing. 1988;17:365–372
  11. Campbell AJ, Borrie MJ, Spears GF, Jackson SL, Brown JS, Fitzgerald JL. Circumstances and consequences of falls experienced by a community population 70 years and over during a prospective study. [published erratum in: Age Ageing 1990;19:345-6] Age Ageing. 1990;19:136–141
  12. Gabell A, Nayak US. The effect of age on variability in gait. J Gerontol. 1984;39:662–666
  13. Overstall PW, Exton-Smith AN, Imms FJ, Johnson AL. Falls in the elderly related to postural imbalance. BMJ. 1997;1(6056):261–264
  14. Tinetti ME, Speechley M. Prevention of falls among the elderly. N Engl J Med. 1989;302:1055–1059
  15. Brunt D, Liu SM, Trimble M, Bauer J, Short M. Principles underlying the organization of movement from quiet stance. Gait Posture. 1999;10:121–128
  16. Brunt D, Santos V, Kim HD, Light K, Levy C. Initiation of movement from quiet stance: comparison of gait and stepping in elderly subjects of different levels of functional ability. Gait Posture. 2005;21:297–302
  17. Sparrow WA, Bradshaw EJ, Lamoureux E, Tirosh O. Ageing effects on the attention demands of walking. Hum Mov Sci. 2002;21:961–972
  18. Berg KO, Wood-Dauphinee SL, Williams JI, et al. Measuring balance in the elderly: preliminary development of an instrument. Physiother Can. 1989;41:304–311
  19. Berg KO, Maki BE, Williams JI, Holliday PJ, Wood-Dauphinee SL. Clinical and laboratory measures of postural balance in an elderly population. Arch Phys Med Rehabil. 1992;73:1073–1080
  20. Schuling J, de Haan R, Limburg M, Groenier KH. The Frenchay Activities Index: assessment of functional status in stroke patients, Stroke 1993;24:1173-7.
  21. Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36) (I. Conceptual framework and item selection). Med Care. 1992;30:473–483
  22. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state” (A practical method for grading the cognitive state of patients for the clinician). J Psychiatr Res. 1975;12:189–198
  23. Tombaugh TN, McIntyre NJ. The Mini-Mental State Examination (A comprehensive review). J Am Geriatr Soc. 1992;40:922–935
  24. Turnbull JC, Kersten P, Habib M, McLellan L, Mullee MA, George S. Validation of the Frenchay Activities Index in a general population 16 years and older. Arch Phys Med Rehabil. 2000;81:1034–1038
  25. Halsaa KE, Brovold T, Graver V, Sandvik L, Bergland A. Assessments of interrater reliability and internal consistency of the Norwegian version of the Berg Balance Scale. [published erratum in: Arch Phys Med Rehabil 2007;88:544] Arch Phys Med Rehabil. 2007;88:94–98
  26. Yip JY, Wilber KH, Myrtle RC, Grazman DN. Comparison of older adult subject and proxy responses on the SF-36 health-related quality of life instrument. Aging Ment Health. 2001;5:136–142
  27. Breniere Y, Do MC. When and how does steady state gait movement induced from upright posture begin?. J Biomech. 1986;19:1035–1040
  28. Brunt D, Short M, Trimble M, Liu SM. Control strategies for initiation of human gait are influenced by accuracy constraints. Neurosci Lett. 2000;285:1–3
  29. Chen HC, Schultz AB, Ashton-Miller JA, Giordani B, Alexander NB, Guire KE. Stepping over obstacles: dividing attention impairs performance of old more than young adults. J Gerontol A Biol Sci Med Sci. 1996;51:M116–M122
  30. Lajoie Y, Teasdale N, Bard C, Fleury M. Upright standing and gait: are there changes in attentional requirements related to normal aging?. Exp Aging Res. 1996;22:185–198
  31. Marsh AP, Geel SE. The effects of age on the attentional demands of postural control. Gait Posture. 2000;12:105–113
  32. Teasdale N, Stelmach GE, Breuning A. Postural sway characteristics of the elderly under normal and altered visual and support surface conditions. J Gerontol. 1991;46:B238–B244
  33. Chou LS, Draganich LF. Stepping over an obstacle increases the motions and moments of the joints of the trailing limb in young adults, Biomechanics 1997;30:331-7.
  34. Patla AE, Rietdyk S. Visual control of limb trajectory over the obstacles during locomotion: effect of obstacle height and width, Gait Posture 1993;1:45-60.
  35. Schrodt LA, Mercer VS, Giuliani CA, Hartman M. Characteristics of stepping over an obstacle in community dwelling older adults under dual-task conditions. Gait Posture. 2004;19:279–287
  36. Brown LA, McKenzie NC, Doan JB. Age-dependent differences in the attentional demands of obstacle negotiation. J Gerontol A Biol Sci Med Sci. 2005;60:924–927
  37. Lundin-Olsson L, Nyberg L, Gustafson Y. “Stops walking when talking” as a predictor of falls in elderly people. Lancet. 1997;349:617
  38. Lundin-Olsson L, Nyberg L, Gustafson Y. Attention, frailty and falls: the effect of a manual task on basic mobility. J Am Geriatr Soc. 1998;46:758–761
  39. Shumway-Cook A, Brauer S, Woollacott M. Predicting the probability for falls in community dwelling older adults using the Timed Up & Go test. Phys Ther. 2000;80:896–903
  40. Verghese J, Buschke H, Viola L. Validity of divided attention tasks in predicting falls in older individuals: a preliminary study. J Am Geriatr Soc. 2002;50:1572–1576
  • a Advanced Mechanical Technology, 176 Waltham St, Watertown, MA 02472.
  • b BIOPAC Systems, 42 Aero Camino, Goleta, CA 93117.
  • c Qualisys Inc, 148 Eastern Blvd, Ste 110, Glastonbury, CT 06033.
  • d Grass Instrument Co, 101 Old Colony Ave, Quincy, MA 02169.

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.

 Reprints are not available from the author.

PII: S0003-9993(07)01251-8

doi:10.1016/j.apmr.2007.07.001

Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 10 , Pages 1309-1313, October 2007

Article Tools

* Image Usage & Resolution