Athletic Training and Sports Health Care

Original Research 

Spatiotemporal Parameters of Treadmill Walking While Dual-Tasking in Those With Chronic Ankle Instability Versus Uninjured Controls

Ashley L. Duncan, MS; Abbey C. Thomas, PhD, ATC; Tricia Hubbard-Turner, PhD, ATC; Christopher J. Burcal, PhD, ATC; Michael J. Turner, PhD; Erik A. Wikstrom, PhD, ATC

Abstract

Purpose:

To determine if dual-tasking while walking affects patients with chronic ankle instability (CAI) differently than uninjured controls.

Methods:

Nineteen individuals with CAI and 19 controls participated. Participants performed three 60-second single-task (walking only) and three dual-task (walking and backwards counting) trials at a self-selected and a preselected speed. Spatiotemporal gait outcomes and their coefficients of variation were captured using an Optogait floor-based photocell system (Microgate, Bolzano, Italy).

Results:

Dual-tasking had no impact on spatiotemporal gait outcome relative to the single-task (walking only) condition and group differences were not identified.

Conclusions:

The results suggest that spatiotemporal gait outcomes in patients with CAI, relative to controls, are not disproportionately influenced by a dual-task paradigm consisting of backwards counting while treadmill walking.

[Athletic Training & Sports Health Care. 201X;X(X):XX–XX.]

Abstract

Purpose:

To determine if dual-tasking while walking affects patients with chronic ankle instability (CAI) differently than uninjured controls.

Methods:

Nineteen individuals with CAI and 19 controls participated. Participants performed three 60-second single-task (walking only) and three dual-task (walking and backwards counting) trials at a self-selected and a preselected speed. Spatiotemporal gait outcomes and their coefficients of variation were captured using an Optogait floor-based photocell system (Microgate, Bolzano, Italy).

Results:

Dual-tasking had no impact on spatiotemporal gait outcome relative to the single-task (walking only) condition and group differences were not identified.

Conclusions:

The results suggest that spatiotemporal gait outcomes in patients with CAI, relative to controls, are not disproportionately influenced by a dual-task paradigm consisting of backwards counting while treadmill walking.

[Athletic Training & Sports Health Care. 201X;X(X):XX–XX.]

Lateral ankle sprains are the most common type of acute musculoskeletal injury sustained during sports.1 Due to a quick return to full activity with minimal activity limitations, ankle sprains are considered a minor injury.2 This may explain why so few individuals complete rehabilitation despite a recommendation from a health care provider.3 Unfortunately, at least 40% of individuals who sprain their ankle go on to experience recurrent sprains and develop chronic ankle instability (CAI).4 The condition of CAI results in sensorimotor impairments commonly quantified as poor postural control5,6 and kinematic alterations throughout the gait cycle.7,8 These adaptations are thought to contribute to the decreased function observed in patients with CAI and initiate a cascade of events that leads to post-traumatic ankle osteoarthritis.2

Despite our knowledge of CAI-related impairments, little is known about how to prevent and/or stop the cascade of events from occurring and/or progressing. One possible reason for this is the limited focus on the interactions among sensorimotor control and cognition. For example, many ankle sprains occur when individuals are performing cognitive tasks (eg, determining how a defender might react) while simultaneously completing different types of dynamic motor tasks. But this type of scenario is not often trained during preventative and/or rehabilitative programs. An individual's ability to perform multiple tasks concurrently can be assessed using dual-tasking paradigms.9,10 Therefore, dual-tasking paradigms may provide additional information about if and how the concurrent completing of a cognitive and motor task influence the sensorimotor adaptations associated with and/or causing CAI. The limited CAI research that has used dual-tasking paradigms has focused on static postural control11–13 but yielded mixed results. Although informative because poor postural control is associated with CAI and is a risk factor for future lateral ankle sprains,14,15 static stance is not representative of activities that commonly cause lateral ankle sprains16 and may not be challenging enough to consistently elicit the central and/or peripheral mechanisms hypothesized to be responsible for the sensorimotor adaptations associated with CAI.17 Therefore, a more dynamic dual-tasking paradigm that is a better representation of real-world scenarios associated with the mechanisms of lateral ankle sprains is needed.

Recently, a dual-tasking paradigm during different gait speeds resulted in stride time variability alterations in both controls and patients with CAI, but the pattern of change differed between the groups.18 Thus, this dual-tasking paradigm (ie, walking at different speeds while simultaneously performing a cognitive task) may provide a better understanding of how cognition influences the gait and sensorimotor impairments observed in patients with CAI. Therefore, the purpose of this study was to identify if dual-tasking during treadmill walking affects patients with CAI differently compared to healthy uninjured controls at a self-selected and faster constrained speed. We hypothesized that patients with CAI would demonstrate significant alterations in spatiotemporal gait characteristics while dual-tasking relative to walking alone, based on the existing literature.18

Methods

Participants

Nineteen individuals with CAI (age: 20.1 ± 2.04 years, height: 167.2 ± 8.5 cm, mass: 68.1 ± 10.6 kg) and 19 healthy young adults (age: 20.9 ± 1.5 years, height: 167.3 ± 9.2 cm, mass: 67.3 ± 11.9 kg) volunteered to participate. Sample size estimates, based on CAI to control differences in gait kinematics and single- to dual-tasking alterations in healthy adults, illustrated effect sizes that ranged from 0.61 to 1.89. Given that these effect size estimates were not based on the variables of interest in the current study, we chose a conservative and small effect size estimate of 0.305. Using this effect size, an alpha level of 0.05, and 1-ß of 0.95, a total sample size of 38 participants was determined to be needed to find statistical significance.

All participants were recreationally active (ie, three aerobic exercise sessions for a total of 90 minutes a week) as determined by participant self-report and the National Aeronautics and Space Administration Physical Activity Status Scale, an indicator of aerobic fitness.19 This scale allows each participant to rate his or her level of physical activity over a set time period. Inclusion criteria for patients with CAI met the recommendations of the International Ankle Consortium.20 In brief, participants with CAI must have had the following: (1) at least one episode of giving way within the past year, (2) at least one recurrent ankle sprain prior to study participation, (3) “yes” responses to 5 or more questions on the Ankle Instability Instrument (AII), (4) a score of less than 90% on the Foot and Ankle Ability Measure (FAAM) Activities of Daily Living subscale, and (5) a score of less than 80% on the FAAM-Sports (FAAM-S) subscale. Participants with CAI must have also been free from previous ankle fractures and not had an acute lower extremity injury within the past 3 months. Potential participants with bilateral CAI were also excluded. Exclusion criteria for all participants included: a history of previous surgeries to the musculoskeletal structures of the lower limbs, a history of a fracture in either lower extremity requiring realignment, and acute injury to musculoskeletal structures of the lower extremity that may impact joint integrity and function. Injury characteristics of both groups can be seen in Table 1. All participants read and signed the approved informed consent form prior to participation.

Demographics and Injury History Characteristics

Table 1:

Demographics and Injury History Characteristics

Procedures

During a single test session, participants were instructed to wear comfortable shoes and clothes. Because instructions can influence dual-tasking results,21 a verbal script was generated and used throughout the test session to prevent discrepancies. Participants first underwent a baseline cognitive task familiarization while seated comfortably and wearing noise-cancelling headphones (Skullcandy Inc., Park City, UT) for familiarization. For cognitive task familiarization, participants were provided a pseudorandom three-digit number (eg, 683) and asked to count aloud backward in steps of 7 as quickly and accurately as possible. Previous research has shown that counting backwards by 7s alters spatiotemporal gait characteristics in young healthy adults.22–24 Each participant completed 10 practice trials, each lasting 60 seconds. A new three-digit number was used for each trial. Extensive practice trials were incorporated to help prevent a potential learning curve, which could artificially alter the impact of dual-tasking on gait. The accuracy of cognitive responses was not recorded because the aim of this investigation was to determine the impact of dual-tasking on motor behavior.

Following cognitive task familiarization, participants completed a walking task to determine their self-selected walking speed on the treadmill. To do this, the speed of the belt was increased in increments of 0.32 kph every 10 seconds until the participant reached a comfortable pace.25 Comfortable pace was operationally defined as leisurely walking. To confirm this selection, belt speed was increased two additional increments and then lowered by 0.16 kph every 10 seconds until the participants reached a comfortable walking speed. The average of the reported comfortable speeds was taken and used in subsequent test trials.25 Next, participants performed six walking trials (three single-task [walking only], three dual-tasking [walking and subtraction]) at their self-selected speed. All walking trials were 60 seconds in length and delivered in a random order, determined by the flipping of a coin. During each dual-tasking trial, a unique computer-generated three-digit number was used. Each trial was separated by a 20-second transition period during which participants continued walking on the treadmill at their self-selected speed and completed a visual analog scale that assessed their perception of task difficulty. During all test trials, a clipboard covered the treadmill console to prevent distractions and participants were reminded to hold their gaze straight ahead. Following a 10-minute rest, participants completed an identical testing protocol but at a preselected speed of 3.86 kph.25

All gait outcomes were captured using the Optogait floor-based photocell system (Microgate, Bolzano, Italy). This device is composed of adjacent transmitting and receiving bars, each containing 99 infrared LEDs. Foot-strike between the bars blocks the transmission of the infrared rays and allows for spatiotemporal gait parameters to be recorded. Data were sampled at 1,000 Hz and processed using dedicated software (Optojump Next, version 1.3.20.0; Microgate).26 Gait outcomes included the means and coefficient of variation (CoV) for stance time (seconds), swing time (seconds), swing percentage (%), double limb support (%), step time (seconds), and step length (cm). Briefly, stance and swing time represent the time in single limb support and swing, respectively. Swing and double limb support correspond to the percentage of the gait cycle spent in swing and double limb support, respectively. Step time represents the time from the heel strike of one foot to the heel strike of the opposite foot and step length represents the distance from the heel strike of one foot to the heel strike of the opposite foot. These outcomes represent the average values for both limbs and were selected based on their apparent sensitivity to dual-tasking in previous research using uninjured controls and patients with CAI.18,24

Data Analysis

Data were collected from 38 participants. However, due to corrupted data files, 3 participants were removed from analysis, leaving a total of 17 CAI and 18 uninjured control participants in the final analysis. These data were submitted to four group (control, CAI) × task (single-task, dual-task) repeated-measures multivariate analyses of variance, which assessed the means and CoV of the self-selected and preselected gait speeds. Between-group Hedges' g effect sizes and 95% confidence intervals were calculated at each walking speed and for all outcomes. Between-group single- to dual-task effect sizes were also calculated for all variables. Hedges' g effect sizes were interpreted as follows: less than 0.3 as small, 0.31 to 0.7 as moderate, and greater than 0.71 as large. Analyses were completed using IBM SPSS software (version 23; IBM Corporation, Armonk, NY) at an a priori alpha level of 0.05.

Results

Spatiotemporal Gait Parameters

Data and effect sizes for the Optogait outcomes at a preselected and self-selected walking speed are listed in Table 2. At a preselected walking speed, our analyses did not reveal any multivariate main effects or interactions (P > .05). Effect sizes were small for task (range: −0.26 to 0.21) and group (range: −0.21 to 0.30) and all 95% confidence intervals included zero. When walking at a self-selected speed, a main effect of group was identified (F(6, 28) = 2.545, P = .043), but follow-up univariate analysis did not reveal differences in any of the six variables (P > .05), suggesting subtle differences in each of the outcomes due to CAI. For the self-selected speed, effect sizes for task were small (range: −0.36 to 0.36) and all 95% confidence intervals crossed zero. Group effect sizes were larger (range: −0.51 to 0.53), but all 95% confidence intervals crossed zero.

Outcome Measures for Effects of Task and Group at a Preselected Walking Speed

Table 2:

Outcome Measures for Effects of Task and Group at a Preselected Walking Speed

CoV of Spatiotemporal Gait Parameters

Table 3 lists the means, standard deviations, effect sizes, and corresponding 95% confidence intervals of the CoV data for the preselected and self-selected walking speed. No significant multivariate main effects or interactions were observed for group or task in the CoV of Optogait outcomes at either walking speed (P > .05). Similarly, all effect sizes for task were weak, ranging from −0.19 to 0.11. Small-to-moderate between-group effect sizes were identified in the CoV measures during the self-selected walking speed (range: −0.40 to −0.02), suggesting a small negative effect of dual-tasking on these measures in the CAI group, but all 95% confidence intervals crossed zero. This was not identified in the preselected walking speed trials because effect sizes were generally smaller (range: −0.30 to 0.00).

Coefficient of Variation on Optogait Outcome Measures for Effects of Task and Group

Table 3:

Coefficient of Variation on Optogait Outcome Measures for Effects of Task and Group

Discussion

The gait outcomes demonstrated no difference between healthy individuals and patients with CAI in both single- and dual-tasking treadmill walking. Contrary to our hypothesis, we concluded there were no differences between patients with CAI and healthy young adults for any gait variables. These findings were consistent across baseline and dual-tasking conditions, but differed from a recent study using a similar methodology.18

The current CAI sample did present with FAAM and FAAM-S scores similar to those of a CAI sample that completed a treadmill walking study and demonstrated kinematic differences from uninjured controls.27 Overall, the current results were similar to the existing literature in that a cognitive task did not alter gait characteristics in young and otherwise healthy adults.25 This may suggest that healthy young adults have sufficient working memory to simultaneously complete a cognitive and motor task while walking. However, it is also possible that the cognitive load was not challenging enough, or that the average between limb spatiotemporal gait measures were not sensitive enough to elicit changes in a young otherwise healthy sample. It is also possible that CAI does not affect working memory enough to disrupt spatiotemporal gait parameters while counting backwards by 7s and walking simultaneously. Similarly, averaging the healthy and CAI involved limb may have inhibited our ability to identify between-group differences. Previous gait studies consistently demonstrate that, relative to uninjured controls, patients with CAI have kinematic differences including decreases in dorsiflexion, a more lateral center of pressure while walking, and more inversion pre-heel strike to post-heel strike.8,27 Because these gait alterations are observed in the involved relative to a matched limb and are kinematic and not spatiotemporal in nature, the cumulative results would support our hypothesis that the average between-limb spatiotemporal outcomes we tested may not be sensitive enough to detect the influences of dual-tasking paradigms in patients with CAI.

Differences in our findings and those of others may also be due to gait pattern variance in treadmill versus over-ground walking.28,29 Over-ground walking can emphasize “freezing gait,” stride variations, and changes in pace, whereas on a treadmill, the brain will prioritize attention to prevent injury and falling off.30 This prioritization may represent another possible explanation for why the chosen cognitive task did not elicit changes in our young healthy sample. Previous research has shown that during over-ground walking, young healthy adults decrease gait speed and swing time to account for the demands of counting backwards by 7s.22–24 It is possible that patients with CAI may respond differently from controls during over-ground walking and that tread-mill walking, perhaps due to motor task prioritization, masked group differences. It is also important to note that visual input is a main contributor in postural control by providing continuous input on body orientation and movement31 and that those with CAI appear to rely more heavily on visual information than uninjured controls.32 Therefore, our serial 7s, a verbal task that stresses the phonological loop of working memory,21 may not have stressed the appropriate processing areas or was not a significant enough stress to influence the average between-limb outcome measures in our younger sample. Future studies will need to stress other areas of working memory, particularly those related to vision, and when possible to use dual-tasking paradigms that are relevant to daily and physical activities to broaden the interpretation of central influences on both postural control and gait in patients with CAI.

Although we believe the above rationales are well grounded based on the existing literature, they do not explain the contrary findings between our work and that of Springer and Gottlieb.18 The authors noted that increasing the difficulty of a walking task, including the addition of the backwards counting by 7s task, caused a reduction in stride time variability in both young healthy adults and patients with CAI. However, the response pattern differed between the groups because a more challenging walking condition elicited stride time variability reductions only in those with CAI. Similar sample sizes were used, identical equipment was used to capture the gait data, and both samples report similar AII scores as part of the inclusion criteria (6.81 ± 1.3818 vs 7.4 ± 2.3). Based on instrumentation constraints, we speculate that Springer and Gottlieb18 also reported average between-limb outcomes, but the authors did not explicitly state this. Several other key differences may also explain the contrary findings. First, the previous investigation did not quantify FAAM and FAAM-S scores or key injury history metrics recommended by the International Ankle Consortium (eg, total number of ankle sprains, was a health care provider seen, etc).20 This is important because CAI is a heterogeneous condition16,17,20 and without this contextual information, it is difficult to directly compare the samples and determine if contrary results are due to methodological differences or simply applying the methodology to different subsamples of CAI. The previous report did quantify the time since the last sprain (21.25 ± 16.57 weeks), which is roughly half of the current sample's mean (12.3 ± 17.36 months). A longer time since injury may have allowed the current sample to adapt and compensate for their latest injury more than the previous sample.18 Although speculative, this is another example of how the investigations may have quantified different subsamples of the CAI population.

Another major difference between the investigations was the walking speeds studied. The current sample (CAI and control) self-selected a slower walking speed (approximately 0.92 m/s) compared to the sample of Springer and Gottlieb18 (approximately 1.35 m/s). The constrained (ie, fast) walking speed used in the current investigation was also slower at approximately 1.07 m/s, whereas the faster walking speed used previously was approximately 1.67 m/s. The ability of Springer and Gottlieb18 to influence gait characteristics during a dual-tasking paradigm at a higher gait speed, particularly in the control group, is consistent with the results of Beauchet et al.24 The authors noted that a backwards counting task significantly reduced a self-selected over-ground walking speed from 1.3 to 1.23 m/s in young healthy adults. Cumulatively, these results may suggest that higher gait velocities are needed to elucidate the effects of a backwards counting dual-tasking paradigm in young adults while walking on a treadmill and that only at such speeds could potential group differences be elucidated. Future research is needed to test this hypothesis by quantifying the effects of dual-tasking paradigms in CAI and control participants under a broader range of gait velocities.

Like all research, this investigation was not without limitations. First, we were limited to assessing the gait characteristics that were available via the Optogait system. For example, the Optogait system cannot distinguish left versus right foot falls during treadmill walking or capture step width. This hinders the exploration of limb-to-limb asymmetries and the development of a more cautious walking pattern, both of which may be influenced by dual-tasking paradigms. Second, we did not assess cognitive task accuracy, which should be done in future research. Assessing cognitive accuracy in conjunction with gait parameters could provide additional insight about prioritization strategies between the groups. We also only assessed two walking speeds and the impact of one type of cognitive task. As discussed, faster gait speeds or other dynamic tasks (eg, running, cutting, or jump landing) may be better suited to elucidate dual-tasking effects and group differences in young adults. Similarly, different types of cognitive tasks and the difficulty of those tasks may have unique influences on over-ground and/or treadmill walking gait characteristics within uninjured controls and/or participants with CAI. Finally, we did not quantify how dual-tasking could influence previously identified gait outcomes that differ between those with CAI and uninjured controls such as foot kinematics and center of pressure position. These may be more pronounced while dual-tasking than spatiotemporal measures.

Implications for Clinical Practice

The results suggest that this dual-tasking combination is not sensitive to CAI-associated impairments and thus should be used as a diagnostic tool. However, the use of dual-task paradigms as an intervention technique may be beneficial for those with CAI. To maximize potential benefits, we recommend personalized but challenging motor and cognitive tasks.

Conclusions

The main objective of this study was to determine how an additional cognitive load (backwards counting by 7s) affected patients with CAI during treadmill walking at different speeds relative to uninjured controls. We observed that the addition of a backwards counting cognitive task did not adversely influence the gait characteristics of patients with CAI or healthy controls. Based on the existing literature, these results suggest that either patients with CAI were able to successfully adapt to the additional constraints placed on them and the uninjured controls or that a more difficult task (eg, higher walking speed, running, cutting, or jump landing) is needed to elicit the effects of additional cognitive loading.

References

  1. Hootman JM, Dick R, Agel J. Epidemiology of collegiate injuries for 15 sports: summary and recommendations for injury prevention initiatives. J Athl Train. 2007;42:311–319.
  2. Gribble PA, Bleakley CM, Caulfield BM, et al. Evidence review for the 2016 international ankle consortium consensus statement on the prevalence, impact, and long-term consequences of lateral ankle sprains. Br J Sport Med. 2016;50:1496–1505. doi:10.1136/bjsports-2016-096189 [CrossRef]
  3. Feger MA, Glaviano NR, Donovan L, Hart JM, Park JS, Hertel J. Current trends in the management of lateral ankle sprain in the United States. Clin J Sport Med. 2017;27:145–152. doi:10.1097/JSM.0000000000000321 [CrossRef]
  4. Doherty C, Bleakley C, Hertel J, Caulfield B, Ryan L, Delahunt E. Recovery from a first-time lateral ankle sprain and the predictors of chronic ankle instability: a prospective cohort analysis. Am J Sports Med. 2016;44:995–1003. doi:10.1177/0363546516628870 [CrossRef]
  5. Groters S, Groen BE, van Cingel R, Duysens J. Double-leg stance and dynamic balance in individuals with functional ankle instability. Gait Posture. 2013;38:968–973. doi:10.1016/j.gaitpost.2013.05.005 [CrossRef]
  6. McKeon PO, Hertel J. Spatiotemporal postural control deficits are present in those with chronic ankle instability. BMC Musculoskelet Disord. 2008;9:76. doi:10.1186/1471-2474-9-76 [CrossRef]
  7. Doherty C, Bleakley C, Hertel J, Caulfield B, Ryan J, Delahunt E. Locomotive biomechanics in persons with chronic ankle instability and lateral ankle sprain copers. J Sci Med Sport. 2016;19:524–530. doi:10.1016/j.jsams.2015.07.010 [CrossRef]
  8. Herb CC, Chinn L, Dicharry J, McKeon PO, Hart JM, Hertel J. Shank-rearfoot joint coupling with chronic ankle instability. J Appl Biomech. 2014;30:366–372. doi:10.1123/jab.2013-0085 [CrossRef]
  9. Frazier EV, Mitra S. Methodological and interpretive issues in postur-cognition dual-tasking in upright stance. Gait Posture. 2008;27:271–279. doi:10.1016/j.gaitpost.2007.04.002 [CrossRef]
  10. Pashler H. Dual-task interference in simple tasks: data and theory. Psychol Bull. 1994;116:220–244. doi:10.1037/0033-2909.116.2.220 [CrossRef]
  11. Rahnama L, Salavati M, Akhbari B, Mazaheri M. Attentional demands and postural control in athletes with and without functional ankle instability. J Orthop Sports Phys Ther. 2010;40:180–187. doi:10.2519/jospt.2010.3188 [CrossRef]
  12. Burcal CJ, Wikstrom EA. Cognitive loading-induced postural control alterations are similar in those with chronic ankle instability and uninjured controls. Gait Posture. 2016;2016:95–98. doi:10.1016/j.gaitpost.2016.05.004 [CrossRef]
  13. Shiravi Z, Talebian Moghadam T, Hadian MR, Olyaei G. Effect of cognitive task on postural control of the patients with chronic ankle instability during single and double leg standing. J Bodyw Mov Ther. 2017;21:58–62. doi:10.1016/j.jbmt.2016.05.001 [CrossRef]
  14. Wang HK, Chen CH, Shiang TY, Jan MH, Lin KH. Risk-factor analysis of high school basketball-player ankle injuries: a prospective controlled cohort study evaluating postural sway, ankle strength, and flexibility. Arch Phys Med Rehabil. 2006;87:821–825. doi:10.1016/j.apmr.2006.02.024 [CrossRef]
  15. Willems TM, Witvrouw E, Delbaere K, Philippaerts R, De Bourdeaudhuij I, De Clercq D. Intrinsic risk factors for inversion ankle sprains in females: a prospective study. Scand J Med Sci Sports. 2005;15:336–345. doi:10.1111/j.1600-0838.2004.00428.x [CrossRef]
  16. Hiller CE, Nightingale EJ, Lin C-W, Coughlan GF, Caulfield B, Delahunt E. Characteristics of people with recurrent ankle sprains: a systematic review with meta-analaysis. Br J Sport Med. 2011;45:660–672. doi:10.1136/bjsm.2010.077404 [CrossRef]
  17. Wikstrom EA, Hubbard-Turner T, McKeon PO. Understanding and treating lateral ankle sprains and their consequences: a constraints-based approach. Sports Med. 2013;43:385–393. doi:10.1007/s40279-013-0043-z [CrossRef]
  18. Springer S, Gottlieb U. Effects of dual-task and walking speed on gait variability in people with chronic ankle instability: a cross sectional study. BMC Musculoskelet Disord. 2017;18:316. doi:10.1186/s12891-017-1675-1 [CrossRef]
  19. Heil DP, Freedson PS, Ahlquist LE, Price J, Rippe JM. Nonexercice regression models to estimate peak oxygen consumption. Med Sci Sports Exer. 1995;27:599–606. doi:10.1249/00005768-199504000-00020 [CrossRef]
  20. Gribble PA, Delahunt E, Bleakley CM, et al. Selection criteria for patients with chronic ankle instability in controlled research: a position statement of the international ankle consortium. J Orthop Sports Phys Ther. 2013;43:585–591. doi:10.2519/jospt.2013.0303 [CrossRef]
  21. Burcal CJ, Drabik EC, Wikstrom EA. The effect of instructions on postural-suprapostural interactions in three working memory tasks. Gait Posture. 2014;40:310–314. doi:10.1016/j.gaitpost.2014.04.203 [CrossRef]
  22. Ghanavati T, Salavati M, Karim N, et al. Intra-limb coordination while walking is affected by cognitive load and walking speed. J Biomech. 2014;47:2300–2005. doi:10.1016/j.jbiomech.2014.04.038 [CrossRef]
  23. Springer S, Giladi N, Peretz C, Yogev G, Simon ES, Hausdorff JM. Dual-tasking effects on gait variability: the role of aging, falls, and executive function. Mov Disord. 2006;21:950–957. doi:10.1002/mds.20848 [CrossRef]
  24. Beauchet O, Dubost V, Hermann FR, Kressig RW. Stride-to-stride variability while backward counting among healthy young adults. J Neuroeng Rehabil. 2005;2:26. doi:10.1186/1743-0003-2-26 [CrossRef]
  25. Qu X. Age-related cognitive task effects on gait characteristics: do different working memory components make a difference?J Neuroeng Rehabil. 2014;11:149. doi:10.1186/1743-0003-11-149 [CrossRef]
  26. Lee MM, Song CH, Lee KJ, Jung SW, Shin DC, Shin SH. Concurrent validity and test-retest reliability of the OPTOGait photoelectric cell system for the assessment of spatio-temporal parameters of the gait of young adults. J Phys Ther Sci. 2014;26:81–85. doi:10.1589/jpts.26.81 [CrossRef]
  27. Chinn L, Dicharry J, Hertel J. Ankle kinematics of individuals with chronic ankle instability while walking and jogging on a treadmill in shoes. Phys Ther Sport. 2013;14:232–239. doi:10.1016/j.ptsp.2012.10.001 [CrossRef]
  28. Wearing SC, Reed LF, Urry SR. Agreement between temporal and spatial gait parameters from an instrumented walking and treadmill system at matched walking speed. Gait Posture. 2013;38:380–384. doi:10.1016/j.gaitpost.2012.12.017 [CrossRef]
  29. Rosenblatt NJ, Hurt CP, Grabiner MD. Sensitivity of dynamic stability to changes in step width during treadmill walking in young adults. J Appl Biom. 2012;28:616–621.
  30. Wollacott M, Shumway-Cook A. Attention and the control of posture and gait: a review of an emerging area of research. Gait Posture. 2002;16:1–14. doi:10.1016/S0966-6362(01)00156-4 [CrossRef]
  31. Lord SR, Menz HB. Visual contributions to postural stability in older adults. Gerontology. 2000;46:306–310. doi:10.1159/000022182 [CrossRef]
  32. Song K, Burcal CJ, Hertel J, Wikstrom EA. Increased visual utilization in chronic ankle instability: a meta-analysis. Med Sci Sports Exer. 2016;48:2046–2056. doi:10.1249/MSS.0000000000000992 [CrossRef]

Demographics and Injury History Characteristics

ParameterUninjured ControlChronic Ankle InstabilityP
Number of ‘Yes’ responses on the Ankle Instability Instrument0.0 ± 0.07.4 ± 2.3< .001
Foot & Ankle Ability Measure: Activities of Daily Living (%)100.0 ± 0.083.0 ± 9.6< .001
Foot & Ankle Ability Measure: Sport (%)100.0 ± 0.071.4 ± 13.4< .001
Number of lateral ankle sprains0.0 ± 0.04.9 ± 6.8< .001
Months since last significant sprain12.3 ± 17.8.016
Number of ‘giving way’ episodes within the past 6 months0.0 ± 0.06.9 ± 7.2< .001
Percentage of participants who saw a health care provider for the initial sprain65%< .001
Number of days non-weight bearing after initial sprain9.3 ± 6.6< .001
Percentage of participants who completed supervised therapy33%.078
Average duration of therapy (weeks)6.0 ± 4.8.028
Participants who saw a health care provider for the most recent sprain22%.220
Number of days non-weight bearing after most recent sprain3.8 ± 3.7.002
Self-reported physical activity (1 to 10)5.9 ± 1.35.6 ± 1.5.967
Self-selected walking speed (kph)3.3 ± 0.83.4 ± 0.5.609

Outcome Measures for Effects of Task and Group at a Preselected Walking Speed

Outcome MeasureSingle Task (Mean ± SD)Dual Task (Mean ± SD)Task Effect Size (95% CI)aGroup Effect Size (95% CI)b
Preselected speed
  Stance time (sec)
    Control0.821 (0.059)0.820 (0.061)−0.01 (−0.66, 0.65)−0.21 (−0.88, 0.45)
    CAI0.848 (0.118)0.846 (0.117)−0.19 (−0.87, 0.48)
  Swing time (sec)
    Control0.376 (0.021)0.373 (0.020)−0.26 (−0.92, 0.39)0.30 (−0.36, 0.97)
    CAI0.377 (0.032)0.376 (0.032)−0.05 (−0.73, 0.62)
  Swing percentage %c
    Control31.430 (1.388)31.300 (1.474)−0.23 (−0.88, 0.43)0.30 (−0.37, 0.97)
    CAI30.940 (3.003)30.966 (2.939)0.13 (−0.54, 0.80)
  Double limb support %c
    Control37.135 (2.755)37.37 (2.929)0.21 (−0.45, 0.86)0.07 (−0.59, 0.73)
    CAI37.754 (5.846)38.070 (5.885)0.21 (−0.46, 0.89)
  Step time (sec)
    Control0.598 (0.036)0.60 (0.036)−0.17 (−0.83, 0.48)0.01 (−0.65, 0.68)
    CAI0.612 (0.068)0.611 (0.068)−0.10 (−0.78, 0.57)
  Step length (cm)
    Control56.370 (4.856)56.31 (5.026)−0.11 (−0.77, 0.54)0.11 (−0.56, 0.77)
    CAI57.059 (9.121)57.058 (9.168)0.00 (−0.67, 0.67)
Self-selected speed
  Stance time (sec)
    Control0.760 (0.039)0.759 (0.039)−0.01 (−0.67, 0.64)0.13 (−0.54, 0.79)
    CAI0.768 (0.056)0.771 (0.035)0.10 (−0.57, 0.77)
  Swing time (sec)
    Control0.365 (0.018)0.367 (0.019)0.34 (−0.32, 1.00)−0.51 (−1.18, 0.17)
    CAI0.370 (0.034)0.368 (0.030)−0.17 (−0.84, 0.51)
  Swing percentage %c
    Control32.465 (1.263)32.605 (1.358)0.36 (−0.30, 1.02)−0.43 (−1.10, 0.24)
    CAI32.702 (2.411)32.327 (1.720)−0.22 (−0.89, 0.45)
  Double limb support %c
    Control35.037 (2.505)34.744 (2.705)−0.36 (−1.02, 0.30)0.53 (−0.15, 1.20)
    CAI35.020 (3.677)35.358 (3.4490.21 (−0.47, 0.88)
  Step time (sec)
    Control0.563 (0.024)0.563 (0.024)0.17 (−0.48, 0.83)0.13 (−0.53, 0.80)
    CAI0.566 (0.051)0.559 (0.027)0.16 (−0.51, 0.84)
  Step length (cm)
    Control60.037 (2.858)60.018 (2.941)−0.02 (−0.67, 0.63)0.23 (−0.44, 0.89)
    CAI61.294 (3.842)61.568 (3.989)0.16 (−0.52, 0.83)

Coefficient of Variation on Optogait Outcome Measures for Effects of Task and Group

Outcome MeasureSingle Task (Mean ± SD)Dual Task (Mean ± SD)Task Effect Size (95% CI)aGroup Effect Size (95% CI)b
Preselected speed
  Stance time
    Control0.028 (0.015)0.029 (0.018)0.04 (−0.61, 0.70)0.00 (−0.66, 0.66)
    CAI0.028 (0.007)0.026 (0.006)−0.19 (−0.86, 0.48)
  Swing time
    Control0.060 (0.036)0.063 (0.042)0.06 (−0.59, 0.72)−0.09 (−0.75, 0.57)
    CAI0.056 (0.015)0.057 (0.017)0.09 (−0.58, 0.76)
  Swing %
    Control0.058 (0.038)0.061 (0.043)0.06 (−0.59, 0.71)−0.14 (−0.80, 0.53)
    CAI0.053 (0.016)0.054 (0.017)0.06 (−0.62, 0.73)
  Double limb support %
    Control0.051 (0.028)0.052 (0.034)0.02 (−0.63, 0.67)0.03 (−0.63, 0.69)
    CAI0.046 (0.012)0.047 (0.012)0.08 (−0.59, 0.76)
  Step time
    Control0.043 (0.003)0.045 (0.043)0.04 (−0.61, 0.69)−0.13 (−0.79, 0.53)
    CAI0.039 (0.012)0.039 (0.012)−0.01 (−0.68, 0.67)
  Step length
    Control0.043 (0.009)0.042 (0.009)−0.05 (−0.70, 0.60)−0.30 (−0.96, 0.37)
    CAI0.050 (0.017)0.047 (0.015)−0.13 (−0.81, 0.54)
Self-selected speed
  Stance time
    Control0.028 (0.018)0.029 (0.021)0.08 (−0.58, 0.73)−0.36 (−1.03, 0.30)
    CAI0.022 (0.003)0.021 (0.003)−0.14 (−0.81, 0.53)
  Swing time
    Control0.058 (0.042)0.061 (0.048)0.07 (−0.58, 0.72)−0.37 (−1.04, 0.30)
    CAI0.045 (0.010)0.044 (0.009)−0.08 (−0.75, 0.59)
  Swing %
    Control0.057 (0.044)0.062 (0.051)0.08 (−0.57, 0.73)−0.40 (−1.07, 0.27)
    CAI0.042 (0.011)0.042 (0.010)−0.06 (−0.73, 0.62)
  Double limb support %
    Control0.054 (0.039)0.059 (0.050)0.11 (−0.55, 0.76)−0.36 (−1.02, 0.31)
    CAI0.024 (0.009)0.042 (0.009)−0.03 (−0.71, 0.64)
  Step time
    Control0.046 (0.040)0.049 (0.046)0.06 (−0.59, 0.71)−0.29 (−0.95, 0.38)
    CAI0.033 (0.008)0.032 (0.006)−0.11 (−0.79, 0.56)
  Step length
    Control0.040 (0.009)0.039 (0.011)−0.04 (−0.69, 0.62)−0.02 (−0.68, 0.64)
    CAI0.045 (0.014)0.045 (0.014)−0.03 (−0.70, 0.64)
Authors

From the University of North Carolina at Charlotte, Charlotte, North Carolina (ALD, ACT, TH-T, MJT); the University of Nebraska Omaha, Omaha, Nebraska (CJB); and the University of North Carolina at Chapel Hill, Chapel Hill, North Carolina (EAW).

The authors have no financial or proprietary interest in the materials presented herein.

Correspondence: Erik A. Wikstrom, PhD, ATC, University of North Carolina at Chapel Hill, 209 Fetzer Hall, CB #8700, Chapel Hill, NC 27599. E-mail: wikstrom@unc.edu

Received: February 28, 2018
Accepted: October 22, 2018
Posted Online: February 26, 2019

10.3928/19425864-20190131-02

Sign up to receive

Journal E-contents