The PROMIS 29 instrument is a 29-item questionnaire assessing each of the following domains: anxiety, depression, fatigue, physical function, pain interference, pain intensity, sleep disturbance, and ability to participate in social roles and activities. The minimum score is 4 per domain, except for pain intensity for which the minimum score is 0. The maximum score per domain is 20, except for pain intensity for which the maximum score is 10. For the physical function and the ability to participate in social roles and activities domains a high score is indicative of positive outcome. For the anxiety, depression, fatigue, sleep disturbance, pain interference, and pain intensity a low score is indicative of positive outcome.

Physical Activity Scale for the Elderly (PASE): self-reported level of physical activity in individuals aged 65 years or older during the previous 7 days. The score accounts for the type of activities performed and the time of performance of each activity. The minimum score is 0. The scale has theoretical maximum value of 864, if subjects spent 24 hours per day over 7 days engaged in vigorous activities. However, as this is not possible, a maximum value of 400 is typically considered as that would correspond to being engage in vigorous activities for 8 hours per day + more moderate activities for 4.5 hours per day.

Trail Making Test (TMT): Test to assess executive cognitive function. It has two parts: TMT A (number sequence only) considers visual search, and TMT B (alternating numbers and letters) evaluates executive control. The participant is asked to draw a line between 24 circles randomly arranged on a page that have to be linked in consecutive order. The TMT is scored by how long it takes to complete the test.

Exercise Self-efficacy Questionnaire (SEE): 9-item questionnaire that focuses on the self-efficacy expectations for exercise for older adults. The minimum score is 0. The maximum score is 90. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome.

Tai Chi experts will use video recordings to score proficiency using a developed instrument to score each of the six Tai Chi movements performed by participants. The minimum score is 6. The maximum score is 30 per Tai Chi exercise. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome. The change in Tai Chi proficiency score ranges from -24 to +24. A change equal to 0 represents no change in proficiency.

Timed-Up-and-Go: Test to determine fall risk. The participant is asked to stand up from a chair, walk 3 meters, turns 180°, walk 3 meters back, and sits back down with back resting against the chair.

Dual task Timed-Up-and-Go test: The participant is asked to stand up from a chair, walk 3 meters, turns 180°, walk 3 meters back, and sits back down with back resting against the chair while counting backwards by three.

Balance on each leg (two trials): The participant is asked to stand on one leg for as long as they can up to a maximum of 30 seconds.

Sit-to-Stand Test: The participant is asked to stand up from a chair while keeping their arms crossed across their chest and repeat the task as many times as possible during a period of 30 seconds. A higher count of completed repetitions is considered indicative of improved performance.

Balance on both legs with feet shoulder-width apart: The participant is asked to stand with feet shoulder-width apart for about 20 seconds. A sensor positioned at the waist is used to track the displacement of the center of mass during the 20 second data collection (for each task). The root mean square (RMS) value of the acceleration data collected using the sensor positioned at the waist is estimated in the antero-posterior (AP) and medio-lateral (ML) directions for the middle 15 seconds of each experiment. Data is gathered at baseline and post-intervention and derived for each condition for a total of 4 variables.

Balance on both legs with feet close to each other: The participant is asked to stand with feet as close as possible (without contact between the feet) for about 20 seconds. A sensor positioned at the waist is used to track the displacement of the center of mass during the 20 second data collection (for each task). The root mean square (RMS) value of the acceleration data collected using the sensor positioned at the waist is estimated in the antero-posterior (AP) and medio-lateral (ML) directions for the middle 15 seconds of each experiment. Data is gathered at baseline and post-intervention and derived for a total of 4 variables.

Study participants will undergo an instrumented gait evaluation. Stride length will be defined as the distance from ipsilateral foot contact to the next ipsilateral foot contact during gait. Normalized-to-height stride length will be derived by computing the stride length using the gait evaluation equipment and dividing such value by the height of the study participant.

Study participants will undergo an instrumented gait evaluation. Stride time will be defined as the time interval between two consecutive foot contacts of the same limb. Stride time variability will be computed as the standard deviation of the stride time estimates derived using the gait evaluation equipment.

Physical activity (e.g. number of steps per day) derived using a wrist-worn activity monitor.

MiniBEST Test: Test to assess dynamic balance. It is a 14-item test scored using a three-level ordinal scale. This test will be performed only in individuals who are willing to be tested in the laboratory (as opposed to via a remote visit). The minimum score is 0. The maximum score is 28. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome.

The percentage of participants who complete the study.

The scale is based on a 10-item questionnaire with five response options (from Strongly agree to Strongly disagree) in which participants rate the usability of the system. The scale ranges from 0 to 100, with higher scores indicating greater usability.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Qualitative interview about ease of use of the Tele-Tai Chi platform. The outcome will be the percentage of participants who provide positive feedback regarding the platform's ease of use.

All dataset analyzed.

Activities-specific Balance Confidence (ABC): 16-item self-report measure in which participants rate their balance confidence to perform motor activities. The minimum score is 0. The maximum score is 100. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome.

The dataset was incomplete and extreme outliers were eliminated. Participants with missing data were excluded from the analysis.

The PROMIS 29 instrument is a 29-item questionnaire assessing each of the following domains: anxiety, depression, fatigue, physical function, pain interference, pain intensity, sleep disturbance, and ability to participate in social roles and activities. The minimum score is 4 per domain, except for pain intensity for which the minimum score is 0. The maximum score per domain is 20, except for pain intensity for which the maximum score is 10. For the physical function and the ability to participate in social roles and activities domains a high score is indicative of positive outcome. For the anxiety, depression, fatigue, sleep disturbance, pain interference, and pain intensity a low score is indicative of positive outcome.

The dataset was incomplete.

Physical Activity Scale for the Elderly (PASE): self-reported level of physical activity in individuals aged 65 years or older during the previous 7 days. The score accounts for the type of activities performed and the time of performance of each activity. The minimum score is 0. The scale has theoretical maximum value of 864, if subjects spent 24 hours per day over 7 days engaged in vigorous activities. However, as this is not possible, a maximum value of 400 is typically considered as that would correspond to being engage in vigorous activities for 8 hours per day + more moderate activities for 4.5 hours per day.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Trail Making Test (TMT): Test to assess executive cognitive function. It has two parts: TMT A (number sequence only) considers visual search, and TMT B (alternating numbers and letters) evaluates executive control. The participant is asked to draw a line between 24 circles randomly arranged on a page that have to be linked in consecutive order. The TMT is scored by how long it takes to complete the test.

The dataset was incomplete. Participants with missing data were excluded from the analysis. For the analysis of Part A only 24 patients were considered.

Exercise Self-efficacy Questionnaire (SEE): 9-item questionnaire that focuses on the self-efficacy expectations for exercise for older adults. The minimum score is 0. The maximum score is 90. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Tai Chi experts will use video recordings to score proficiency using a developed instrument to score each of the six Tai Chi movements performed by participants. The minimum score is 6. The maximum score is 30 per Tai Chi exercise. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome. The change in Tai Chi proficiency score ranges from -24 to +24. A change equal to 0 represents no change in proficiency.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Timed-Up-and-Go: Test to determine fall risk. The participant is asked to stand up from a chair, walk 3 meters, turns 180°, walk 3 meters back, and sits back down with back resting against the chair.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Dual task Timed-Up-and-Go test: The participant is asked to stand up from a chair, walk 3 meters, turns 180°, walk 3 meters back, and sits back down with back resting against the chair while counting backwards by three.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Balance on each leg (two trials): The participant is asked to stand on one leg for as long as they can up to a maximum of 30 seconds.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Sit-to-Stand Test: The participant is asked to stand up from a chair while keeping their arms crossed across their chest and repeat the task as many times as possible during a period of 30 seconds. A higher count of completed repetitions is considered indicative of improved performance.

The dataset is incomplete. Participants with missing data were excluded from the analysis.

Balance on both legs with feet shoulder-width apart: The participant is asked to stand with feet shoulder-width apart for about 20 seconds. A sensor positioned at the waist is used to track the displacement of the center of mass during the 20 second data collection (for each task). The root mean square (RMS) value of the acceleration data collected using the sensor positioned at the waist is estimated in the antero-posterior (AP) and medio-lateral (ML) directions for the middle 15 seconds of each experiment. Data is gathered at baseline and post-intervention and derived for each condition for a total of 4 variables.

The dataset was incomplete. Participants with missing data were excluded from the analysis. The root mean square (RMS) value of the acceleration in the antero-posterior (AP) and medio-lateral (ML) directions at the waist sensor was calculated for middle 15 seconds of each experiment.

Balance on both legs with feet close to each other: The participant is asked to stand with feet as close as possible (without contact between the feet) for about 20 seconds. A sensor positioned at the waist is used to track the displacement of the center of mass during the 20 second data collection (for each task). The root mean square (RMS) value of the acceleration data collected using the sensor positioned at the waist is estimated in the antero-posterior (AP) and medio-lateral (ML) directions for the middle 15 seconds of each experiment. Data is gathered at baseline and post-intervention and derived for a total of 4 variables.

Data analysis was limited only to subjects who completed in the intervention and completed both baseline and post-intervention assessments. The root mean square (RMS) value of the acceleration in the antero-posterior (AP) and medio-lateral (ML) directions at the waist sensor was calculated for middle 15 seconds of each experiment.

Study participants will undergo an instrumented gait evaluation. Stride length will be defined as the distance from ipsilateral foot contact to the next ipsilateral foot contact during gait. Normalized-to-height stride length will be derived by computing the stride length using the gait evaluation equipment and dividing such value by the height of the study participant.

The dataset is incomplete. Participants with missing data were excluded from the analysis.

Study participants will undergo an instrumented gait evaluation. Stride time will be defined as the time interval between two consecutive foot contacts of the same limb. Stride time variability will be computed as the standard deviation of the stride time estimates derived using the gait evaluation equipment.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

Physical activity (e.g. number of steps per day) derived using a wrist-worn activity monitor.

The dataset was incomplete. Participants with missing data were excluded from the analysis.

MiniBEST Test: Test to assess dynamic balance. It is a 14-item test scored using a three-level ordinal scale. This test will be performed only in individuals who are willing to be tested in the laboratory (as opposed to via a remote visit). The minimum score is 0. The maximum score is 28. Positive changes (i.e., higher scores post-intervention) would represent a positive outcome.

Data analysis was limited only to subjects who completed both baseline and post-intervention assessments in the laboratory.

Reports of flu-like symptoms during study. All cases self-resolved with supportive care.

Reports of falls during non-study-related activities. No medical care required.

Reports of headache during non-study-related activities.

Reports of knee pain in participants with pre-existing knee pain.

Reports of muscle soreness primarily related to non-study-related activities.

Reports of skin irritation related to use of wearable devices.

Reports of dizziness during non-study-related activities.