ClinicalTrials.gg
CompletedNCT06090604Updated Nov 18, 2026Results posted

Walking Balance Training Post-Stroke

An interventional study of High intensity gait training in a balance challenging Movement amplification environment in Stroke, sponsored by Northwestern University. Completed at 1 site in United States. Open to participants aged 18 Years to 85 Years. Per ClinicalTrials.gov, last updated 2026-11-18.

Sponsored by Northwestern University · Not applicable, Interventional, and Treatment

How we know

Updated Nov 18, 2026Results postedStudy documents posted+1 moreGo to Updates ↓

Help build ClinicalTrials.gg — short surveys, remote interviews or testing, in your language. Take part →

Phase
Not applicable
Study type
Interventional
Enrollment
33
Allocation
Not applicable
Ages
18 Years to 85 Years
Sex
All
01

Where this study stands

Counted across the records on this site · 2026-11-22

Among 24 studies in Stroke by Northwestern University — 6 open to participants now, 8 under way, none of them phase 3. Stroke · Northwestern University.

Enrollment
15th of 20
33 · median 52
Started
2023
with 1 other
No phase
17 of 24
5 open now
Registered
20th of 24
by date

Among the 24 studies in Stroke by Northwestern University, this one is the 15th largest by enrollment of the 20 interventional ones that give a count, is one of 2 that started in 2023, is one of 17 with no phase, and was the 20th registered.

Stroke

8,283 studies on the registry are indexed under Stroke; 2,065 are open to participants now.

Browse Stroke studies →

Open to participants now

6 of the 24 studies in Stroke by Northwestern University are open now, in 1 country. Recruiting first, the newest first:

1 more — sign in →

By country, each count listing its studies:

All 6 open studies →

Where they have run

Who these studies take

Of the 1,418 Stroke studies recruiting now, 698 (49%) set no upper age limit and 281 (20%) stop at 75 or younger; 77 take children under 18; 197 take healthy volunteers.

This study: 18 Years to 85 Years · all sexes · no healthy volunteers.

The full table on the hub →

This tab as a report to print or save as PDF → With an account.

Was this useful?
Get alerts for this landscapeDownload CSVNarrow it in search →With an account: a weekly mail of what changed among 24 studies in Stroke by Northwestern University, or all of them as a file.
02

Study summary

Stroke is one of the leading causes of disability in the United States, affecting \~795,000 people annually. Among ambulatory people with chronic stroke, impaired balance is common and substantially limits mobility (those with the poorest balance walk the least). This project will explore if a novel gait training intervention using a robotic device to amplify a person's self-generated movements can improve walking balance in people with chronic stroke. The development of effective interventions to increase walking balance among people with chronic stroke will positively impact quality of life and ability to participate in walking activities.

Read the detailed description

Background: People with chronic stroke (PwCS) have substantial walking balance dysfunctions that limit independence and participation in walking activities. There is a pressing need to develop effective methods to enhance walking balance in PwCS. Interventions that amplify self-generated movements may accelerate motor learning by enhance a person's perception of movement errors. This method could potentially be applied to help PwCS improve walking balance. To this end, we have developed a cable-driven robot to create a Movement Amplification Environment (MAE) during treadmill walking. The MAE challenges walking balance by applying lateral forces to the pelvis that are proportional in magnitude to real-time lateral center of mass (COM) velocity. Our purpose is to evaluate effects of MAE training on walking balance in PwCS.

Specific Aims: Aim 1a: To quantify locomotor adaptations PwCS make to maintain walking balance in a MAE. We will evaluate the impact of the two foremost gait training variables: Gait Speed and MAE Strength. Aim 1b: To evaluate short-term changes in the ability of PwCS to control their lateral COM excursion during walking (a quantifiable walking balance measure) immediately after MAE training. Aim 2: To evaluate long-term changes in walking balance and daily walking following a 5-week high-intensity gait training intervention performed in a MAE.

Approach: Aims 1a \& b: Twenty PwCS will each participate in two experiments evaluating gait biomechanics (COM dynamics and stepping patterns) during and immediately following MAE training. Aim 1a will evaluate locomotor adaptations when walking in a MAE and the effect of gait speed (slow, preferred, and fast) and MAE strength (low, medium, and high). Aim 1b will use visual feedback to challenge PwCS to control their lateral COM excursion during walking immediately before and after walking in different external conditions (No applied forces, or MAE). Outcomes will assess if short-term walking balance is enhanced immediately following MAE exposure. Aim 2: Ten PwCS will participate in 10-sessions of high intensity gait training performed in a MAE. We will assess pre- to post- changes in walking balance using clinical gait and balance measures, biomechanical assessments, and participation in daily walking (steps/day). Outcomes will assess if long-term walking balance is enhanced following MAE gait training.

Impact: Training walking balance of PwCS by amplifying their own self-generated movements is a radical departure from current practice and could substantially enhance walking balance. This study will provide a foundational understanding of how PwCS maintain walking balance in a MAE and if beneficial behaviors persist following MAE training. Successful outcomes will motivate a randomized controlled trial assessing efficacy of MAE training on walking balance in PwCS.

03

Conditions studied

  • Stroke

Browse trials for

Keywords

  • Stroke
  • Balance control
  • Walking dysfunction
04

Who can participate

Ages eligible
18 Years to 85 Years
Sexes eligible
All
Accepts healthy volunteers
No

Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.

Inclusion criteria

  1. 18 to 80 years of age;
  2. diagnosis of hemiparetic stroke;
  3. more than 6 months post-stroke;
  4. ability to ambulate over ground for 10 meters with or without a single cane, and/or ankle-foot orthosis.

Exclusion criteria

Exclusion Criteria:

  1. cognitive impairment (score of ≤22/30 on the scale);
  2. aphasia (score of ≥71/100 on the Mississippi Aphasia Screening Test);
  3. excessive spasticity in lower limbs (score of > 3 on the Modified Ashworth Scale);
  4. unable to tolerate 10-min. of standing;
  5. enrollment in concurrent physical therapy;
  6. use of braces/orthotics crossing the knee joint;
  7. known pregnancy;
  8. severe cardiovascular, musculoskeletal, or other neurological conditions affecting gait and balance.
  9. brainstem and/or cerebellar lesion
  10. perinatal stroke
Terms in these criteria, in plain language(1)

Underlined terms in the criteria open these too. Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.

MMSE and MoCA

Short tests of memory and thinking, each scored out of 30. Lower scores suggest more difficulty; criteria give a range, either to include people with memory problems or to leave them out.

05

Study design

Phase
Not applicable
Primary purpose
Treatment
Allocation
Not applicable
Intervention model
Single group
Masking
None (open label)
Enrollment
33 participants (actual)

Study arms

  • Experimental
    Control of lateral COM motion during and after walking practiced in the MAE

    For Aim 1a, participant's preferred and fast treadmill walking speeds will be determined followed by 2-min of baseline walking. Each participant will perform eight 2-min trials of treadmill walking 1) Null Environment: no forces, 2) MAE Low Gain: 25 Nsm-1, 3) MAE Medium Gain: 35 Nsm-1, 4) MAE High Gain: 45 Nsm-1, that will be repeated at both treadmill walking speeds. The trial order will be randomized. For Aim 1b, we will assess participant's maximum ability to control their lateral COM motion with no forces applied, using three 21-meter walking trials with visual projections on the treadmill to provide feedback used to challenge their lateral COM motion control. Five 2-min trials in a Null environment will be followed by COM control assessment. Participants will rest and repeat the above sequence in a MAE. The order of the external environments will be randomized across participants. Participants may participate in more than one aim (1a, 1b and 2).

    Other: High intensity gait training in a balance challenging Movement amplification environment

Interventions

  • OtherHigh intensity gait training in a balance challenging Movement amplification environment

    For Aim 2, each participant will engage in ten 45-min high-intensity gait training sessions performed in a Movement Amplification Environment (MAE). Training sessions will occur 2x/week and led by a licensed Physical Therapist. The goal of each training session is to achieve 40-min of stepping practice within a targeted range of heart rate (HR) of 70 to 80% of estimated age-predicted HR max and perceived excursion (RPE) in a MAE. Each session will begin and end with a 2.5-min warm-up and cool down, with 40-min of training using two 10-min Speed-training blocks and two 10-min Balance-training blocks. Balance training activities will include lateral maneuvers, head turns, obstacle negotiation, and backward walking. The MAE gain, training speeds and balance activities will be progressively increased within and across sessions to challenge the participant.

06

What researchers measure

Primary outcomes

  1. Gait Stability (Short-term Lyapunov Exponent)

    Gait stability was measured using the short-term Lyapunov exponent that quantifies the average logarithmic rate of divergence of a system after small kinematic deviations. This metric has demonstrated construct and predictive validity in prior simulation and walking studies where divergence is respective to the gait cycle. The lateral COM velocity (VCOM) was used for this calculation as a measure of local, dynamic stability. Lower values of the short-term Lyapunov exponent indicates higher stability and higher values indicates lower stability and greater challenge to walking balance. For Aim 1a, gait stability was calculated at session 1 (baseline). For Aim 1b, gait stability was calculated at session 2 and 3. For Aim 2, gait stability was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  2. Lateral Center of Mass (COM) Excursion

    Lateral center of mass (COM) excursion occurring during 5 consecutive gait cycles was calculated as our metric of ability to control lateral COM motion during walking. Kinematic marker data was processed using Visual3D (C-Motion, Germantown, MD) and a custom MATLAB (Mathworks, Natick, MA) program. Marker data was gap-filled and low-pass filtered (Butterworth, 6 Hz cut-off frequency). Time of initial foot contact (IC) and toe-off (TO) events were identified for each step based on fore-aft positions of the calcaneus and 2nd metatarsal markers. A reduction in the mean lateral COM excursion suggests improved walking balance or adaptive response towards changing gains. For Aim 1a, COM excursion was calculated at session 1. For Aim 1b, gait stability was calculated at session 2 and 3. For Aim 2, COM excursion was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  3. Lateral Center of Mass (COM) Velocity

    Lateral center of mass (COM) velocity was calculated by taking the derivative of filtered COM position (same filter as marker data). The average of absolute value of lateral COM velocity peaks in both directions were calculated. For Aim 1a, COM velocity was calculated at session 1. For Aim 1b, velocity was calculated at session 2 and 3. For Aim 2, COM velocity was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  4. Ten Minute Walk Test (10MWT) at Preferred Speed

    Ten minute walk test (10MWT) is a simple measurement of an individual's average walking speed measured at self-selected speed (instruction: "walk at your normal comfortable pace"). For Aim 2, 10MWT was collected at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  5. Ten Minute Walk Test (10MWT) at Fast Speed

    Ten minute walk test (10MWT) is a simple measurement of an individual's average walking speed measured at fast speed (instruction: "walk as fast as you can safely walk"). For Aim 2, 10MWT was collected at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  6. Dynamic Gait Index (DGI)

    DGI is a clinical assessment tool used to evaluate an individual's balance, walking ability, and fall risk. It tests how well a person can adapt their gait to external demands, such as navigating obstacles and turning their head, rather than just walking on a flat surface. Each task is graded on a 4-point ordinal scale (0 to 3), where 3 indicates normal function with no impairments, and 0 indicates severe impairment (e.g., losing balance or requiring physical assistance). The total score of DGI is 24, with a minimum score of 0 and maximum score of 24. Higher scores indicate better balance, while lower scores indicate poor balance. For Aim 2, DGI was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

Secondary outcomes

  1. Berg Balance Scale (BBS)

    BBS is a 14-item measure that assesses static balance with excellent validity and reliability in individuals post-stroke. The total score for this test is 56. Minimum score is 0 and maximum score is 56. Higher scores indicate better balance and lower fall-risk, while lower scores indicate poor balance and increased fall-risk. For Aim 2, BBS was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  2. Timed up and go Test (TUG)

    TUG was used to assess functional mobility, walking balance, and fall risk. It is a quick, valid, reliable and widely used clinical performance based measure in individuals post-stroke. While there is no time limit to the TUG test, more time taken to finish the task indicates poor balance and greater fall-risk. For Aim 2, TUG was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  3. Activities-specific Balance Confidence (ABC) Scale

    ABC scale is a self-reported measure of an individual's confidence while performing specific postural and ambulatory activities. The ABC is a reliable and valid measure of balance confidence in individuals post-stroke. The total score for ABC scale is 100. Minimum score is 0 and maximum score is 100. Higher scores indicate better balance confidence, while lower scores indicate poor balance confidence. For Aim 2, ABC scale was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  4. Six Minute Walk Distance (6MWD)

    Six minute walk test is a sub-maximal exercise test used to assess aerobic capacity and endurance. The distance covered during 6 minutes of walking indicate participants exercise tolerance. The vitals measures before and after the 6MWT indicate the cardiovascular and respiratory response to exercise. For Aim 2, 6MWD was collected at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  5. Daily Stepping in the Home and Community

    The amount of daily stepping in the home and community taken per day averaged across the 7-day collection periods was collected before and after 10 high intensity gait training session in MAE in Aim 2. Daily stepping was measured and recorded using a StepWatch4 (Modus, Edmonds, WA) activity monitor. The microprocessor-based accelerometer was worn around the ankle during all waking hours (except bathing) for the duration of the stepping assessment period. For Aim 2, daily stepping was calculated at session 4 (pre-training) and session 15 (post-training).

    Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  6. Step Width

    Step width was calculated as the medio-lateral distance between the left and right 5th Metatarsal markers at initial foot contact. For Aim 1a, step width was calculated at session 1. For Aim 1b, step width was calculated at session 2 and 3.

    Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

  7. Step Length

    Step length was calculated as the anteroposterior distance between the calcaneal markers of the leading and trailing limbs at heel strike for each stride. For Aim 1a, step length was calculated at session 1. For Aim 1b, step length was calculated at session 2 and 3.

    Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.

07

Results

Posted Nov 18, 2026

Participant flow

33 participants had consented for this study for participation. Out of which, 3 participants were excluded and 1 participant dropped out.

Participant flow — Overall Study
MilestoneAim 1Aim 2
Started2211
Completed1910
Not completed31
Withdrew: Lost to follow-up01
Withdrew: Excluded from the study30

Outcome measures

PrimaryGait Stability (Short-term Lyapunov Exponent)

Gait stability was measured using the short-term Lyapunov exponent that quantifies the average logarithmic rate of divergence of a system after small kinematic deviations. This metric has demonstrated construct and predictive validity in prior simulation and walking studies where divergence is respective to the gait cycle. The lateral COM velocity (VCOM) was used for this calculation as a measure of local, dynamic stability. Lower values of the short-term Lyapunov exponent indicates higher stability and higher values indicates lower stability and greater challenge to walking balance. For Aim 1a, gait stability was calculated at session 1 (baseline). For Aim 1b, gait stability was calculated at session 2 and 3. For Aim 2, gait stability was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · second-¹ (s-¹)
Gait Stability (Short-term Lyapunov Exponent)
second-¹ (s-¹)Aim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 12.09 ± 0.342.33 ± 0.372.34 ± 0.302.33 ± 0.412.11 ± 0.272.31 ± 0.442.35 ± 0.332.31 ± 0.37—————
Session 2————————2.44 ± 0.352.45 ± 0.39———
Session 3——————————2.17 ± 0.272.16 ± 0.28—
Session 4 (Pre-training)————————————2.24 ± 0.24
Session 15 (Post-training)————————————2.12 ± 0.23
Statistical analysis
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.001 (Effect of gain (null, low, medium and high)) · Partial eta squared (ηp²): 0.3967
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Effect of speed (preferred and fast)) · Partial eta squared (ηp²): 0.0026
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Interaction effect for gain (null, low, medium and high) by speed (preferred and fast)) · Partial eta squared (ηp²): 0.0111
  • Aim 1b_Early_MAE vs Aim 1b_Early_Null · t-test, 2 sided · p = <0.001 · Mean difference (final values): 0.27 · 95% CI .163 to .376
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.64 (Sequence (order of field exposure)) · Partial eta squared (ηp²): 0.013
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.7 (Field (MA or Null)) · Partial eta squared (ηp²): 0.009
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · Partial eta squared (ηp²) · p = 0.58 (Interaction effect for Sequence and Field (MA, null)) · Partial eta squared (ηp²): 0.018
  • Aim 2 · Wilcoxon Signed-Rank Test · p = 0.049 · Mean difference (final values): 0.12
PrimaryLateral Center of Mass (COM) Excursion

Lateral center of mass (COM) excursion occurring during 5 consecutive gait cycles was calculated as our metric of ability to control lateral COM motion during walking. Kinematic marker data was processed using Visual3D (C-Motion, Germantown, MD) and a custom MATLAB (Mathworks, Natick, MA) program. Marker data was gap-filled and low-pass filtered (Butterworth, 6 Hz cut-off frequency). Time of initial foot contact (IC) and toe-off (TO) events were identified for each step based on fore-aft positions of the calcaneus and 2nd metatarsal markers. A reduction in the mean lateral COM excursion suggests improved walking balance or adaptive response towards changing gains. For Aim 1a, COM excursion was calculated at session 1. For Aim 1b, gait stability was calculated at session 2 and 3. For Aim 2, COM excursion was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · cms
Lateral Center of Mass (COM) Excursion
cmsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 10.112 ± 0.0290.102 ± 0.0270.107 ± 0.0270.111 ± 0.0280.100 ± 0.0250.094 ± 0.0260.097 ± 0.0260.097 ± 0.026—————
Session 2————————0.095 ± 0.0250.097 ± 0.027———
Session 3——————————0.102 ± 0.0240.106 ± 0.026—
Session 4 (Pre-training)————————————0.106 ± 0.028
Session 15 (Post-training)————————————0.098 ± 0.030
Statistical analysis
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.05 (Effect of gain (null, low, medium and high)) · Partial eta squared (ηp²): 0.3085
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.001 (Effect of speed (preferred and fast)) · Partial eta squared (ηp²): 0.7033
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = =0.05 (Interaction effect for gain (null, low, medium and high) by speed (preferred and fast)) · Partial eta squared (ηp²): 0.1464
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE · t-test, 2 sided · p = 0.004 · Mean difference (final values): -0.008 · 95% CI -.012 to -.002
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.52 (Sequence (order of field exposure)) · Partial eta squared (ηp²): 0.024
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.35 · Partial eta squared (ηp²): 0.052
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.10 (Interaction effect for Sequence and Field (MA, null)) · Partial eta squared (ηp²): 0.155
  • Aim 2 · t-test, 2 sided · p = 0.087 · Mean difference (final values): 0.007 · 95% CI -.001 to 0.016
PrimaryLateral Center of Mass (COM) Velocity

Lateral center of mass (COM) velocity was calculated by taking the derivative of filtered COM position (same filter as marker data). The average of absolute value of lateral COM velocity peaks in both directions were calculated. For Aim 1a, COM velocity was calculated at session 1. For Aim 1b, velocity was calculated at session 2 and 3. For Aim 2, COM velocity was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · cms/second
Lateral Center of Mass (COM) Velocity
cms/secondAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 10.0031 ± 0.00080.0030 ± 0.00090.0030 ± 0.00080.0032 ± 0.00090.0044 ± 0.00400.0033 ± 0.00080.0034 ± 0.00080.0034 ± 0.0008—————
Session 2————————0.0032 ± 0.00080.0031 ± 0.0007———
Session 3——————————0.0032 ± 0.00080.0032 ± 0.0009—
Session 4 (Pre-training)————————————0.0030 ± 0.0006
Session 15 (Post-training)————————————0.0032 ± 0.0004
Statistical analysis
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Effect of gain (null, low, medium and high)) · Partial eta squared (ηp²): 0.1238
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.05 · Partial eta squared (ηp²): 0.6219
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Interaction effect for gain (null, low, medium and high) by speed (preferred and fast)) · Partial eta squared (ηp²): 0.0815
  • Aim 1b_Early_MAE vs Aim 1b_Early_Null · t-test, 2 sided · p = 0.97 · Mean difference (final values): -0.000003 · 95% CI -.00017 to .00017
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.31 (Sequence (order of field exposure)) · Partial eta squared (ηp²): 0.061
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.003 (Field (MA or Null)) · Partial eta squared (ηp²): 0.414
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.13 (Interaction effect for Sequence and Field (MA, null)) · Partial eta squared (ηp²): 0.132
  • Aim 2 · t-test, 2 sided · p = 0.285 · Mean difference (final values): 0.000 · 95% CI -.001 to .000
SecondaryBerg Balance Scale (BBS)

BBS is a 14-item measure that assesses static balance with excellent validity and reliability in individuals post-stroke. The total score for this test is 56. Minimum score is 0 and maximum score is 56. Higher scores indicate better balance and lower fall-risk, while lower scores indicate poor balance and increased fall-risk. For Aim 2, BBS was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · units
Berg Balance Scale (BBS)
unitsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————47.50 ± 5.46
Session 15 (Post-training)————————————50.40 ± 3.47
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.013 · Mean difference (final values): -2.90 · 95% CI -5.018 to -.782
SecondaryTimed up and go Test (TUG)

TUG was used to assess functional mobility, walking balance, and fall risk. It is a quick, valid, reliable and widely used clinical performance based measure in individuals post-stroke. While there is no time limit to the TUG test, more time taken to finish the task indicates poor balance and greater fall-risk. For Aim 2, TUG was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · seconds
Timed up and go Test (TUG)
secondsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————13.77 ± 2.40
Session 15 (Post-training)————————————13.29 ± 3.48
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.519 · Mean difference (final values): 0.484 · 95% CI -1.148 to 2.116
SecondaryActivities-specific Balance Confidence (ABC) Scale

ABC scale is a self-reported measure of an individual's confidence while performing specific postural and ambulatory activities. The ABC is a reliable and valid measure of balance confidence in individuals post-stroke. The total score for ABC scale is 100. Minimum score is 0 and maximum score is 100. Higher scores indicate better balance confidence, while lower scores indicate poor balance confidence. For Aim 2, ABC scale was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · units
Activities-specific Balance Confidence (ABC) Scale
unitsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————73.65 ± 12.65
Session 15 (Post-training)————————————76.65 ± 12.67
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.144 · Mean difference (final values): -3.01 · 95% CI -7.255 to 1.239
SecondarySix Minute Walk Distance (6MWD)

Six minute walk test is a sub-maximal exercise test used to assess aerobic capacity and endurance. The distance covered during 6 minutes of walking indicate participants exercise tolerance. The vitals measures before and after the 6MWT indicate the cardiovascular and respiratory response to exercise. For Aim 2, 6MWD was collected at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · meters
Six Minute Walk Distance (6MWD)
metersAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————266.21 ± 79.75
Session 15 (Post-training)————————————283.26 ± 66.13
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.071 · Mean difference (final values): -17.05 · 95% CI -35.922 to 1.822
SecondaryDaily Stepping in the Home and Community

The amount of daily stepping in the home and community taken per day averaged across the 7-day collection periods was collected before and after 10 high intensity gait training session in MAE in Aim 2. Daily stepping was measured and recorded using a StepWatch4 (Modus, Edmonds, WA) activity monitor. The microprocessor-based accelerometer was worn around the ankle during all waking hours (except bathing) for the duration of the stepping assessment period. For Aim 2, daily stepping was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · Number of steps per day
Daily Stepping in the Home and Community
Number of steps per dayAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————3410.24 ± 1804.62
Session 15 (Post-training)————————————2934.75 ± 1617.75
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.133 · Mean difference (final values): 475.49 · 95% CI -174.973 to 1125.958
SecondaryStep Width

Step width was calculated as the medio-lateral distance between the left and right 5th Metatarsal markers at initial foot contact. For Aim 1a, step width was calculated at session 1. For Aim 1b, step width was calculated at session 2 and 3.

Time frame:
Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · cms
Step Width
cmsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 10.309 ± 0.0510.310 ± 0.0500.314 ± 0.0500.318 ± 0.0510.306 ± 0.0560.310 ± 0.0550.310 ± 0.0510.311 ± 0.053—————
Session 2————————0.313 ± 0.0580.313 ± 0.059———
Session 3——————————0.301 ± 0.0510.303 ± 0.053—
Statistical analysis
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.05 (Effect of gain (null, low, medium and high)) · Partial eta squared (ηp²): 4.44
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.05 (Effect of speed (preferred and fast)) · Partial eta squared (ηp²): 4.55
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Interaction effect for gain (null, low, medium and high) by speed (preferred and fast)) · Partial eta squared (ηp²): 1.70
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE · t-test, 2 sided · p = 0.06 · Mean difference (final values): 0.012 · 95% CI -.0003 to .023
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.06 (Sequence (order of field exposure)) · Partial eta squared (ηp²): 4.08
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.27 (Field (MA or Null)) · Partial eta squared (ηp²): 1.3
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.70 (Interaction effect for Sequence and Field (MA, null)) · Partial eta squared (ηp²): 0.15
SecondaryStep Length

Step length was calculated as the anteroposterior distance between the calcaneal markers of the leading and trailing limbs at heel strike for each stride. For Aim 1a, step length was calculated at session 1. For Aim 1b, step length was calculated at session 2 and 3.

Time frame:
Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · cms
Step Length
cmsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 10.281 ± 0.1140.277 ± 0.1130.276 ± 0.1120.280 ± 0.1090.339 ± 0.1210.341 ± 0.1210.343 ± 0.1220.337 ± 0.118—————
Session 2————————0.306 ± 0.1240.311 ± 0.126———
Session 3——————————0.316 ± 0.1260.318 ± 0.126—
Statistical analysis
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Effect of gain (null, low, medium and high)) · Partial eta squared (ηp²): 0.08
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = <0.05 (Effect of speed (preferred and fast)) · Partial eta squared (ηp²): 86.13
  • Aim 1a_Preferred speed_Null vs Aim 1a_Preferred speed_Low MAE vs Aim 1a_Preferred speed_Medium MAE vs Aim 1a_Preferred speed_High MAE vs Aim 1a_Fast speed_Null vs Aim 1a_Fast speed_Low MAE vs Aim 1a_Fast speed_Medium MAE vs Aim 1a_Fast speed_High MAE · ANOVA · p = >0.05 (Interaction effect for gain (null, low, medium and high) by speed (preferred and fast)) · Partial eta squared (ηp²): 1.39
  • Aim 1b_Early_MAE vs Aim 1b_Early_Null · t-test, 2 sided · p = 0.03 · Mean difference (final values): -0.011 · 95% CI -.019 to -.001
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · ANOVA · p = 0.04 (Sequence (order of field exposure)) · Partial eta squared (ηp²): 4.77
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · t-test, 2 sided · p = 0.21 (Field (MA or Null)) · Mean difference (final values): 1.66
  • Aim 1b_Early_MAE vs Aim 1b_Late_MAE vs Aim 1b_Early_Null vs Aim 1b_Late_Null · t-test, 2 sided · p = 0.27 (Interaction effect for Sequence and Field (MA, null)) · Mean difference (final values): 1.31
PrimaryTen Minute Walk Test (10MWT) at Preferred Speed

Ten minute walk test (10MWT) is a simple measurement of an individual's average walking speed measured at self-selected speed (instruction: "walk at your normal comfortable pace"). For Aim 2, 10MWT was collected at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · meters/second
Ten Minute Walk Test (10MWT) at Preferred Speed
meters/secondAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————0.851 ± 0.242
Session 15 (Post-training)————————————0.854 ± 0.199
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.952 · Mean difference (final values): -0.003 · 95% CI -.112 to .106
PrimaryTen Minute Walk Test (10MWT) at Fast Speed

Ten minute walk test (10MWT) is a simple measurement of an individual's average walking speed measured at fast speed (instruction: "walk as fast as you can safely walk"). For Aim 2, 10MWT was collected at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · meters/second
Ten Minute Walk Test (10MWT) at Fast Speed
meters/secondAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————1.175 ± 0.386
Session 15 (Post-training)————————————1.143 ± 0.337
Statistical analysis
  • Aim 2 · t-test, 2 sided · p = 0.333 · Mean difference (final values): 0.032 · 95% CI -.039 to .103
PrimaryDynamic Gait Index (DGI)

DGI is a clinical assessment tool used to evaluate an individual's balance, walking ability, and fall risk. It tests how well a person can adapt their gait to external demands, such as navigating obstacles and turning their head, rather than just walking on a flat surface. Each task is graded on a 4-point ordinal scale (0 to 3), where 3 indicates normal function with no impairments, and 0 indicates severe impairment (e.g., losing balance or requiring physical assistance). The total score of DGI is 24, with a minimum score of 0 and maximum score of 24. Higher scores indicate better balance, while lower scores indicate poor balance. For Aim 2, DGI was calculated at session 4 (pre-training) and session 15 (post-training).

Time frame:
Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
Reported as:
Mean · units
Dynamic Gait Index (DGI)
unitsAim 1a_Preferred speed_NullAim 1a_Preferred speed_Low MAEAim 1a_Preferred speed_Medium MAEAim 1a_Preferred speed_High MAEAim 1a_Fast speed_NullAim 1a_Fast speed_Low MAEAim 1a_Fast speed_Medium MAEAim 1a_Fast speed_High MAEAim 1b_Early_MAEAim 1b_Late_MAEAim 1b_Early_NullAim 1b_Late_NullAim 2
Session 4 (Pre-training)————————————14.30 ± 2.15
Session 15 (Post-training)————————————15.00 ± 1.41
Statistical analysis
  • Aim 2 · Wilcoxon-signed rank · p = 0.125 · Mean difference (final values): 0.70

Adverse events

Collected over Adverse event data were monitored for each participant for 3 weeks in Aim 1 and for 7 weeks in Aim 2.. Non-serious events are listed at a 0% frequency threshold.

Adverse event summary by group
GroupDeathsSeriousOther
Aim 10/19 (0%)0/19 (0%)0/19 (0%)
Aim 20/10 (0%)0/10 (0%)0/10 (0%)

Baseline characteristics

Age, Continuous
Age, Continuous(Years)Aim 1Aim 2Total
Mean62.84 ± 8.4763.30 ± 6.6763.00 ± 7.81
Sex: Female, Male
Sex: Female, Male(Participants)Aim 1Aim 2Total
Female538
Male14721
Race (NIH/OMB)
Race (NIH/OMB)(Participants)Aim 1Aim 2Total
American Indian or Alaska Native000
Asian235
Native Hawaiian or Other Pacific Islander000
Black or African American10212
White7512
More than one race000
Unknown or Not Reported000
Ethnicity (NIH/OMB)
Ethnicity (NIH/OMB)(Participants)Aim 1Aim 2Total
Hispanic or Latino101
Not Hispanic or Latino171027
Unknown or Not Reported101
10 Meter Walk Test (10MWT) at preferred walking speed
10 Meter Walk Test (10MWT) at preferred walking speed(Meters per second)Aim 1Aim 2Total
Mean0.82 ± 0.190.85 ± 0.250.83 ± 0.21
Timed Up and Go (TUG)
Timed Up and Go (TUG)(Seconds)Aim 1Aim 2Total
Mean13.87 ± 4.5913.77 ± 2.413.84 ± 3.98
Dynamic Gait Index (DGI)
Dynamic Gait Index (DGI)(units on a scale)Aim 1Aim 2Total
Mean15.89 ± 2.8514.3 ± 2.1515.34 ± 2.73
Berg Balance Scale (BBS)
Berg Balance Scale (BBS)(units on a scale)Aim 1Aim 2Total
Mean48.16 ± 4.2147.50 ± 5.4647.93 ± 4.69

3 further baseline measures are reported on the registry.

08

Study locations

1 site
  • Northwestern University
    Chicago, Illinois 60611, United States
09

References and documents

Study documents

PDFs hosted by the registry, ClinicalTrials.gov.

Individual participant data

Plan to share: Yes — De-identified data from this study will be uploaded on the Northwestern University Research and Data repository (Arch) so that it is fully available to other parties interested in probing the data further and performing secondary analyses.

Supporting information: Study protocol

Time frame: After completion of the study

Access criteria: Data will be publicly available.

10

Timeline

1 registry update since Sep 25, 2026
ElapsedEstimatedEstimated, not yet reachedSimilar studies: half by the dot; the band is the middle halfHalf of similar studies reached it by the dot; the band is the middle half, from a quarter to three quarters

Ran from Aug 2023 to primary completion in May 2025, 20 months.

Results were first posted in Nov 2026, 1 year 6 months after primary completion.

Of 3,121 non-industry interventional studies with no phase under Stroke, half reached primary completion within 23 months of starting (the median), counting those still running; a quarter within 10 and three quarters within 46. This study's 20 months is below that.

Counting every sponsor, half of 3,265 studies reached it within 23 months; a quarter within 10 and three quarters within 46.

How these figures are worked out, and which studies are left out of them.

Registered
Sep 18, 2023
First posted
Oct 19, 2023
StartActual
Aug 18, 2023
Primary completionActual
May 2, 2025
CompletionActual
May 2, 2026
Results first posted
Nov 18, 2026
Last update
Nov 18, 2026
What these dates mean
Start.
When the first participant enrolled; while estimated, when the study expects to open to recruitment.
Primary completion.
When the last participant was examined or treated to collect data for the study's main question, its primary outcome. A study that stopped early gives the date its data collection stopped.
Completion.
When the last participant's last visit took place, for every outcome, secondary ones and side effects included: the end of the study. Often later than primary completion, sometimes the same day.
Actual and Est.
Actual means the registry says the date has happened. Est. (estimated) is a date the sponsor plans or expects. It can still move, and the sponsor is expected to change it to actual once it has passed.
Results
Results posted
posted Nov 18, 2026
Primary outcomes
4 edited, 1 time frame changed, 1 added, 4 removed, 1 description edited
What changed in the outcomes
  • EditedMinimum average lateral centerLateral Center of massMass (COM) excursionExcursionTime frame: For Aim 1, COM excursion will be calculated at session 1 and after a week on session 2. For Aim 2, COM excursion will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • EditedMinimum lateral Margin Of StabilityLateral Center of Mass (MOSCOM) VelocityTime frame: For Aim 1, margin of stability will be calculated at session 1 and after a week on session 2. For Aim 2, margin of stability will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • Edited10 MeterTen Minute Walk Test (10MWT) at Preferred SpeedTime frame: For Aim 1, 10MWT will be performed at session 1 to determine baseline functional status. For Aim 2, 10MWT will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • Edited6Ten Minute Walk Test (6WMT10MWT) at Fast SpeedTime frame: For Aim 2, 6WMT will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • AddedGait Stability (Short-term Lyapunov Exponent)Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • RemovedTimed Up and Go (TUG)Time frame: For Aim 1, TUG will be performed at session 1 to determine baseline functional status. For Aim 2, TUG will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long.
  • RemovedBerg Balance Scale (BBS)Time frame: For Aim 1, BBS will be performed at session 1 to determine baseline functional status. For Aim 2, BBS will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long.
  • RemovedActivities-specific Balance Confidence (ABC) ScaleTime frame: For Aim 1, ABC will be performed at session 1 to determine baseline functional status. For Aim 2, ABC will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long.
  • RemovedStepping Activity in the Home and CommunityTime frame: For Aim 2, the stepping activity will be assessed for a period of one week after session 1 and session 12 (post 5 weeks of training). Participants will be asked to wear the monitor during all waking hours.
  • Time frameDynamic Gait Index (DGI)For Aim 1, DGI will be performed at session 1 to determine baseline functional status. For Aim 2, DGI will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • DescriptionDynamic Gait Index (DGI): how it is measured was reworded
Nov 18, 2026
Secondary outcomes
2 edited, 5 added, 5 removed
What changed in the outcomes
  • EditedStep lengthWidthTime frame: For Aim 1, step length will be performed at session 1 and after a week on session 2. For Aim 2, step length will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • EditedStep widthLengthTime frame: For Aim 1, step width will be performed at session 1 and after a week on session 2. For Aim 2, step width will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • AddedBerg Balance Scale (BBS)Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • AddedTimed up and go Test (TUG)Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • AddedActivities-specific Balance Confidence (ABC) ScaleTime frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • AddedSix Minute Walk Distance (6MWD)Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • AddedDaily Stepping in the Home and CommunityTime frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
  • RemovedHeart rateTime frame: For Aim 2, heart rate will be collected before and after each of the 10 one hour long training sessions over 5 weeks. Vitals will be collected at session 1 and after 5 weeks of training at session 12. Both these sessions will be two hours long.
  • RemovedBlood pressureTime frame: For Aim 2, blood pressure will be collected before and after each of the 10 one hour long training sessions over 5 weeks. Vitals will be collected at session 1 and after 5 weeks of training at session 12. Both these sessions will be two hours long.
  • RemovedExertionTime frame: For Aim 2, participants level of exertion will be collected throughout the 10 training sessions over 5 weeks. Each training session will be one hour long.
  • RemovedNumber of steps taken during training sessionTime frame: For Aim 2, the total number of steps during training will be collected after each of the 10 training session over 5 weeks. Each training session will be one hour long.
  • RemovedChedoke-McMaster Stroke Assessment (CMSA) scaleTime frame: For both Aims 1 and 2, CMSA scale will be collected at session 1 to determine participants stroke recovery. This session will be two hours long.
Nov 18, 2026
Show all 1 update
  1. Nov 18, 2026
    Results posted
    Primary outcomes 4 edited, 1 time frame changed, 1 added, 4 removed, 1 description edited
    What changed in the outcomes
    • EditedMinimum average lateral centerLateral Center of massMass (COM) excursionExcursionTime frame: For Aim 1, COM excursion will be calculated at session 1 and after a week on session 2. For Aim 2, COM excursion will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • EditedMinimum lateral Margin Of StabilityLateral Center of Mass (MOSCOM) VelocityTime frame: For Aim 1, margin of stability will be calculated at session 1 and after a week on session 2. For Aim 2, margin of stability will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • Edited10 MeterTen Minute Walk Test (10MWT) at Preferred SpeedTime frame: For Aim 1, 10MWT will be performed at session 1 to determine baseline functional status. For Aim 2, 10MWT will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • Edited6Ten Minute Walk Test (6WMT10MWT) at Fast SpeedTime frame: For Aim 2, 6WMT will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • AddedGait Stability (Short-term Lyapunov Exponent)Time frame: Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • RemovedTimed Up and Go (TUG)Time frame: For Aim 1, TUG will be performed at session 1 to determine baseline functional status. For Aim 2, TUG will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long.
    • RemovedBerg Balance Scale (BBS)Time frame: For Aim 1, BBS will be performed at session 1 to determine baseline functional status. For Aim 2, BBS will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long.
    • RemovedActivities-specific Balance Confidence (ABC) ScaleTime frame: For Aim 1, ABC will be performed at session 1 to determine baseline functional status. For Aim 2, ABC will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long.
    • RemovedStepping Activity in the Home and CommunityTime frame: For Aim 2, the stepping activity will be assessed for a period of one week after session 1 and session 12 (post 5 weeks of training). Participants will be asked to wear the monitor during all waking hours.
    • Time frameDynamic Gait Index (DGI)For Aim 1, DGI will be performed at session 1 to determine baseline functional status. For Aim 2, DGI will be performed at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • DescriptionDynamic Gait Index (DGI): how it is measured was reworded
    Secondary outcomes 2 edited, 5 added, 5 removed
    What changed in the outcomes
    • EditedStep lengthWidthTime frame: For Aim 1, step length will be performed at session 1 and after a week on session 2. For Aim 2, step length will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • EditedStep widthLengthTime frame: For Aim 1, step width will be performed at session 1 and after a week on session 2. For Aim 2, step width will be calculated at session 1 and after 5 weeks of training at session 12. Each of these sessions will be two hours long. → Session 1 was baseline; Sessions 2 and 3 occurred one week apart within 2 weeks to 10 months of baseline. Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • AddedBerg Balance Scale (BBS)Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • AddedTimed up and go Test (TUG)Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • AddedActivities-specific Balance Confidence (ABC) ScaleTime frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • AddedSix Minute Walk Distance (6MWD)Time frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • AddedDaily Stepping in the Home and CommunityTime frame: Pre-training assessment (Session 4) occurred one year after baseline, and post-training assessment (Session 15) followed five weeks later.
    • RemovedHeart rateTime frame: For Aim 2, heart rate will be collected before and after each of the 10 one hour long training sessions over 5 weeks. Vitals will be collected at session 1 and after 5 weeks of training at session 12. Both these sessions will be two hours long.
    • RemovedBlood pressureTime frame: For Aim 2, blood pressure will be collected before and after each of the 10 one hour long training sessions over 5 weeks. Vitals will be collected at session 1 and after 5 weeks of training at session 12. Both these sessions will be two hours long.
    • RemovedExertionTime frame: For Aim 2, participants level of exertion will be collected throughout the 10 training sessions over 5 weeks. Each training session will be one hour long.
    • RemovedNumber of steps taken during training sessionTime frame: For Aim 2, the total number of steps during training will be collected after each of the 10 training session over 5 weeks. Each training session will be one hour long.
    • RemovedChedoke-McMaster Stroke Assessment (CMSA) scaleTime frame: For both Aims 1 and 2, CMSA scale will be collected at session 1 to determine participants stroke recovery. This session will be two hours long.
    + 1 other change: verification date

From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗

11

Registry details

Key details

Study ID
NCT06090604
Responsible party
Keith Gordon (Professor, Northwestern University) — Principal investigator
First posted
Oct 19, 2023
Start date
Aug 18, 2023
Primary completion
May 2, 2025
Completion
May 2, 2026
Results posted
Nov 18, 2026
Last update
Nov 18, 2026

Study contacts

Keith Gordon, PhD
principal investigator · Assistant Professor, Northwestern University

Oversight

Data monitoring committee
No
FDA-regulated drug
No
FDA-regulated device
No
View the source record on ClinicalTrials.gov ↗

Not currently enrolling

This study is completed, as verified in Oct 2026. You cannot join it, but the record below documents what was studied.

Follow this study

Get an email when the registry record changes — status, dates, results — or when someone posts here.

Sign in to follow

Discussion

Questions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.

Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.

Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.

Start the discussion