Panoramic Bracelet:

The Panoramic Digital Health Bracelet is a wearable inertial measurement unit "intended for use by researchers and healthcare professionals for high frequency or continuous collection of physical data in home and professional healthcare settings during research studies." It includes a 3 axis accelerometer and gyroscope, as well as magnetometer, pressure, and temperature sensors. The bracelet has no interface or screen for participants to interact with.

GENEActiv:

The GENEActiv device is a wearable inertial measurement unit "designed for public health research and clinical trials." The device is a circular unit with attachable wrist-bands, and has no interface or screen for participants to interact with. The GENEActiv device is FDA 510(k) exempt. The device records acceleration data and can collect data remotely for up to 1 month.

Stance duration was defined as the duration of time the foot is in contact with the ground in a step during gait. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Swing time was defined as the time during which the foot was not in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Stride length was defined as the distance covered when two steps were taken, one with each foot. It was measured from the heel of one foot to the heel of same foot when it came into contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Step duration was the period between the first contact of a foot and the first following contact of the contralateral foot. Participants had a single in-laboratory visit on Day 1 during which they performed gait monitoring procedures. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Cadence referred to the number of steps taken per minute during gait cycle. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Stride duration refers to the duration between successive heel contacts of the same foot. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Double support time referred to the phase during walking when both feet were in contact with the ground. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Single support time referred to the time when only one limb was in contact with the ground. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Stance duration was defined as the duration of time the foot is in contact with the ground in a step during gait. Participants had a single in-lab visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Swing duration was defined as the time during which the foot was not in contact with the ground. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Stride length was defined as the distance covered when two steps were taken, one with each foot. It was measured from the heel of one foot to the heel of same foot when it came into contact with the ground. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Step duration was the period between the first contact of a foot and the first following contact of the contralateral foot. Participants had a single in-lab visit during which they performed gait monitoring procedures. For in-lab simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Cadence referred to the number of steps taken per minute during gait cycle. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Gait speed was the time taken to walk a specified distance. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Stride duration refers to the duration between successive heel contacts of the same foot. Participants had a single in-lab visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

Total sleep time for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's total sleep opportunity (TSO) window and computing sleep metrics within that window.

The percentage of time asleep for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

Number of wake bouts for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

Sleep onset latency which is time to transition from full wakefulness to sleep, for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

Wake after sleep onset is time for which participants were awake after having initially fallen asleep, for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

Total scratch duration for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

Total scratch episodes for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

Available minutes represented any recorded accelerometer data per day per device in minutes. Number of available minutes for GENEActiv and Panoramic Bracelet was derived by the in-house built Sci-kit Digital Health (SKDH) package.

Number of minutes of wear include the total number of minutes the device is worn including sleep time and awake time. The wear time parameter (number of minutes of wear) for GENEActiv and Panoramic Bracelet was derived by the in-house built SKDH package.

The wear time parameter (number of wear minutes while awake) for GENEActiv and Panoramic Bracelet was derived by the in-house built SKDH package.

Wearability/comfort questionnaire included a total of 10 items for which the participants (12 to 17 years) or their caregivers (participants aged 3 to 11 years) were required to provide a response on a scale of 0 to 4, where 0 = strongly disagree, and 4 = strongly agree, higher scores indicated more comfort, except for questions 5 ("The device changes the way I move") and 6 ("The device changes the way I behave") , where 0 = strongly agree and 4 = strongly disagree, higher scores indicate more comfort. Number of participants according to response to each item of the wearability/comfort questionnaire is reported.

Number of days to complete enrollment corresponded to the total number of days it took to enroll all study participants within the age groups of 3-5 years, 6-11 years and 12-17 years.

Percentage of tasks completed from a total of 7 tasks (i.e., walking at natural, slow and fast speeds, tile and carpet walks, 2 sessions of in-lab simulated activities) is reported in this outcome measure.

Wearability/comfort questionnaire included a total of 10 items for which the participants (12 to 17 years) or their caregivers (participants aged 3 to 11 years) were required to provide a response on a scale of 0 to 4, where 0 = strongly disagree, and 4 = strongly agree, higher scores indicate more comfort except for questions 5 ("The device changes the way I move") and 6 ("The device changes the way I behave") , where 0 = strongly agree and 4 = strongly disagree, higher scores indicate more comfort. Number of participants according to response to each item of the wearability/comfort questionnaire is reported.

Wearability/comfort questionnaire included a total of 10 items for which the participants (12 to 17 years) or their caregivers (participants aged 3 to 11 years) were required to provide a response on a scale of 0 to 4, where 0 = strongly disagree, and 4 = strongly agree, higher scores indicate more comfort except for questions 5 ("The device changes the way I move") and 6 ("The device changes the way I behave") , where 0 = strongly agree and 4 = strongly disagree, higher scores indicate more comfort. Number of participants according to response to each item of the wearability/comfort questionnaire is reported.

Double support time referred to the phase during walking when both feet were in contact with the ground. Participants had a single in-lab visit during which they performed gait monitoring procedures. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Single limb support time referred to the time when only one limb was in contact with the ground. Participants had a single in-lab visit during which they performed gait monitoring procedures. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Stance duration was defined as the duration of time the foot is in contact with the ground in a step during gait. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Swing time was defined as the time during which the foot was not in contact with the ground. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Stride length was defined as the distance covered when two steps were taken, one with each foot. It was measured from the heel of one foot to the heel of same foot when it came into contact with the ground. Participants had a single in-lab visit during which they performed gait monitoring procedures. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Step duration was the period between the first contact of a foot and the first following contact of the contralateral foot. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Cadence referred to the number of steps taken per minute during gait cycle. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Gait speed was the time taken to walk a specified distance. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Stride duration refers to the duration between successive heel contacts of the same foot. Participants were required to work on tile and carpet at their natural speed while wearing the device under study (ActiGraph) and APDM system. Raw accelerometry data collected by ActiGraph devices was processed by GaitPy to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors. Analysis was performed using mixed model regression analyses with device, floor type, and their interaction as fixed effects, and participant as random effect.

Double support time referred to the phase during walking when both feet were in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Single support time referred to the time when only one limb was in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Stance duration was defined as the duration of time the foot is in contact with the ground in a step during gait. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Swing time was defined as the time during which the foot was not in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Stride length was defined as the distance covered when two steps were taken, one with each foot. It was measured from the heel of one foot to the heel of same foot when it came into contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Step duration was the period between the first contact of a foot and the first following contact of the contralateral foot. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Cadence referred to the number of steps taken per minute during gait cycle. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Gait speed was the time taken to walk a specified distance. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Stride duration refers to the duration between successive heel contacts of the same foot. Participants had to walk on GAITRite walkway (20 feet*3) wearing Panoramic bracelet device and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by devices were processed by manufacturer's algorithm for APDM and SKDH gait algorithm for Panoramic bracelet device to generate gait metrics.

Participants aged 6 to 11 years wore the ActiGraph CenterPoint Insight Watch on the non-dominant wrist (worn continuously) and lumbar region (worn after waking up in the morning and removed while sleeping at night or nap during daytime) for approximately 15 days (+ 3 days). Participants wore the APDM Opal sensors on each wrist, chest, lumbar, and above each shoe, only on Day 1.

Participants aged 12 to 17 years wore the ActiGraph CenterPoint Insight Watch on the non-dominant wrist (worn continuously) and lumbar region (worn after waking up in the morning and removed while sleeping at night or nap during daytime) for approximately 15 days (+ 3 days). Participants wore the APDM Opal sensors on each wrist, chest, lumbar, and above each shoe, only on Day 1.

Participants aged 3 to 5 years wore the Panoramic Digital Health Bracelet on the non-dominant wrist for approximately 15 days (+ 3 days). GENEActiv device was also worn on the same non-dominant wrist from Day 7 (+ 2 days) to Day 15 (+ 2 days). Participants wore the APDM Opal sensors on each wrist, chest, lumbar, and above each shoe, only on Day 1.

Participants aged 6 to 11 years wore the Panoramic Digital Health Bracelet on the non-dominant wrist for approximately 15 days (+ 3 days). GENEActiv device was also worn on the same non-dominant wrist from Day 7 (+ 2 days) to Day 15 (+ 2 days). Participants wore the APDM Opal sensors on each wrist, chest, lumbar, and above each shoe, only on Day 1.

Participants aged 12 to 17 years wore the Panoramic Digital Health Bracelet on the non-dominant wrist for approximately 15 days (+ 3 days). GENEActiv device was also worn on the same non-dominant wrist from Day 7 (+ 2 days) to Day 15 (+ 2 days). Participants wore the APDM Opal sensors on each wrist, chest, lumbar, and above each shoe, only on Day 1.

Total of all reporting groups

Double support time referred to the phase during walking when both feet were in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

Full Analysis Set (FAS) included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Single support time referred to the time when only one limb was in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometery data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Stance duration was defined as the duration of time the foot is in contact with the ground in a step during gait. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Swing time was defined as the time during which the foot was not in contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Stride length was defined as the distance covered when two steps were taken, one with each foot. It was measured from the heel of one foot to the heel of same foot when it came into contact with the ground. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Step duration was the period between the first contact of a foot and the first following contact of the contralateral foot. Participants had a single in-laboratory visit on Day 1 during which they performed gait monitoring procedures. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Cadence referred to the number of steps taken per minute during gait cycle. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Stride duration refers to the duration between successive heel contacts of the same foot. Participants had to walk on GAITRite walkway (20 feet*3) wearing ActiGraph and APDM devices at participant's natural speed and at varied speeds. Natural speed was participant's self-chosen pace for 3 rounds. Varied speeds included participant's self-chosen slow pace ("Walk slow") and fast pace ("Walk fast") for 3 rounds. Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from GAITRite walkway and APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Number Analyzed' signifies number of participants evaluable for the specified rows. The outcome measure was planned to be evaluated only for Cohort A.

Double support time referred to the phase during walking when both feet were in contact with the ground. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Single support time referred to the time when only one limb was in contact with the ground. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Stance duration was defined as the duration of time the foot is in contact with the ground in a step during gait. Participants had a single in-lab visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Swing duration was defined as the time during which the foot was not in contact with the ground. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Stride length was defined as the distance covered when two steps were taken, one with each foot. It was measured from the heel of one foot to the heel of same foot when it came into contact with the ground. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Step duration was the period between the first contact of a foot and the first following contact of the contralateral foot. Participants had a single in-lab visit during which they performed gait monitoring procedures. For in-lab simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Cadence referred to the number of steps taken per minute during gait cycle. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Gait speed was the time taken to walk a specified distance. Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Participants had a single in-laboratory visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Stride duration refers to the duration between successive heel contacts of the same foot. Participants had a single in-lab visit during which they performed gait monitoring procedures. For in-laboratory simulated activities, participants were required to alternate between activity blocks through a random path created by dice roll while wearing devices. Activities were done for approximately 20 minutes (approximately 7 minutes of activities [activity block 1], approximately 5 minutes of break and approximately 7 minutes of activities [activity block 2]). Activities: 2 sitting (e.g., look through a book, draw/write), 2 standing (e.g., turn a switch or hang a coat), 2 walking (e.g., move blocks/objects, toss a beanbag and pick it up). Raw accelerometry data collected by ActiGraph devices (comparator) was processed by the GaitPy algorithm to generate gait quality metrics (lumbar position). Gait metrics data was directly extracted from APDM sensors (reference).

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. The outcome measure was planned to be evaluated only for Cohort A.

Total sleep time for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's total sleep opportunity (TSO) window and computing sleep metrics within that window.

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Overall Number of Participants Analyzed ' signifies number of participants evaluable for this outcome measure and 'Number Analyzed' signifies number of participants analyzed for specified rows. The outcome measure was planned to be evaluated only for Cohort B.

The percentage of time asleep for GeneActiv and Panoramic Bracelet was derived by the SleepPy algorithm. SleepPy algorithm was used for detecting a participant's TSO window and computing sleep metrics within that window.

FAS included all participants who had a signed and dated informed consent document, had assented to the study and were enrolled in the study. Here, 'Overall Number of Participants Analyzed ' signifies number of participants evaluable for this outcome measure and 'Number Analyzed' signifies number of participants analyzed for specified rows. The outcome measure was planned to be evaluated only for Cohort B.