Contributions
Abstract: P698
Type: Poster Sessions
Abstract Category: Clinical aspects of MS - Clinical assessment tools
Introduction: Objectively measuring physical activity in the natural environment via remote monitoring can be useful for documenting activity status for clinical care and as an exploratory outcome measure for Multiple Sclerosis (MS) clinical trials. As technology advances, newer devices require validation for measurement of step count in people with MS.
Objective: Determine the validity of a consumer-grade (wristworn) accelerometer (Fitbit-Flex2), against the earlier generation Fitbit-Flex and the ActiGraph-GT3X (waist), an accelerometer commonly used in clinical research, to measure step count in people with MS.
Methods: 27 people with relapsing or progressive MS enrolled in a longitudinal observational study recording activity using a Fitbit Flex were invited to participate in a 7-day home validation wearing the Fitbit Flex2 and ActiGraph. The main outcome was steps per epoch (minute and daily). Disability was assessed using the EDSS (expanded disability status scale); via an online validated-questionnaire at the time of monitoring, and in person during study check-ins.
Results: The median EDSS at baseline was 4 (range 0.0-6.5). 13 (48%) of the cohort were women, and 14 (52%) had relapsing-remitting MS and 48% progressive MS. High Interclass correlations between Flex and Flex2 (0.97) and between the ActiGraph and Flex2 (0.92) over 1-week of continuous activity monitoring. Bland-Altman plots showed no systematic bias between Flex2 and Flex (mean difference: 53 more steps recorded by Flex2). Bland-Altman plots showed greater number of step per day recorded by Flex2 than ActiGraph (mean difference: 1018 more steps). 93% of these data were within the 95% limits of agreement. The majority (88%) of the subjects showed no difference in wear-time for the Flex and Flex2. ActiGraph was worn for fewer hours a day than both Fitbit devices and our analysis was adjusted for usage.
Accounting for MS subtype did not change the results for Flex vs Flex 2. However, in progressive MS, ActiGraph vs Flex 2 there was a significant relationship between mean difference and step count (slope: 0.237, p=0.0008). Linear regression with Bland-Altman resulted in plots with 96% falling within the limits of agreement.
Conclusion: Average daily step counts can be determined in MS patients using either the Flex or Flex2 device. These consumer-targeted devices can be an alternative to research-grade accelerometers, particularly for longer-term remote activity monitoring in people with MS
Disclosure: Valerie Block and Chao Zhao have no relevant disclosures.
Jeffrey Gelfand has received personal compensation for consulting from Biogen and Merck. Personal compensation for medical legal consulting. Research support from Genentech unrelated to this abstract.
Bruce Cree has received personal compensation for consulting from Abbvie, Akili, Biogen, EMD Serono, Ge Neuro, Novartis, Sanofi Genzyme.
Abstract: P698
Type: Poster Sessions
Abstract Category: Clinical aspects of MS - Clinical assessment tools
Introduction: Objectively measuring physical activity in the natural environment via remote monitoring can be useful for documenting activity status for clinical care and as an exploratory outcome measure for Multiple Sclerosis (MS) clinical trials. As technology advances, newer devices require validation for measurement of step count in people with MS.
Objective: Determine the validity of a consumer-grade (wristworn) accelerometer (Fitbit-Flex2), against the earlier generation Fitbit-Flex and the ActiGraph-GT3X (waist), an accelerometer commonly used in clinical research, to measure step count in people with MS.
Methods: 27 people with relapsing or progressive MS enrolled in a longitudinal observational study recording activity using a Fitbit Flex were invited to participate in a 7-day home validation wearing the Fitbit Flex2 and ActiGraph. The main outcome was steps per epoch (minute and daily). Disability was assessed using the EDSS (expanded disability status scale); via an online validated-questionnaire at the time of monitoring, and in person during study check-ins.
Results: The median EDSS at baseline was 4 (range 0.0-6.5). 13 (48%) of the cohort were women, and 14 (52%) had relapsing-remitting MS and 48% progressive MS. High Interclass correlations between Flex and Flex2 (0.97) and between the ActiGraph and Flex2 (0.92) over 1-week of continuous activity monitoring. Bland-Altman plots showed no systematic bias between Flex2 and Flex (mean difference: 53 more steps recorded by Flex2). Bland-Altman plots showed greater number of step per day recorded by Flex2 than ActiGraph (mean difference: 1018 more steps). 93% of these data were within the 95% limits of agreement. The majority (88%) of the subjects showed no difference in wear-time for the Flex and Flex2. ActiGraph was worn for fewer hours a day than both Fitbit devices and our analysis was adjusted for usage.
Accounting for MS subtype did not change the results for Flex vs Flex 2. However, in progressive MS, ActiGraph vs Flex 2 there was a significant relationship between mean difference and step count (slope: 0.237, p=0.0008). Linear regression with Bland-Altman resulted in plots with 96% falling within the limits of agreement.
Conclusion: Average daily step counts can be determined in MS patients using either the Flex or Flex2 device. These consumer-targeted devices can be an alternative to research-grade accelerometers, particularly for longer-term remote activity monitoring in people with MS
Disclosure: Valerie Block and Chao Zhao have no relevant disclosures.
Jeffrey Gelfand has received personal compensation for consulting from Biogen and Merck. Personal compensation for medical legal consulting. Research support from Genentech unrelated to this abstract.
Bruce Cree has received personal compensation for consulting from Abbvie, Akili, Biogen, EMD Serono, Ge Neuro, Novartis, Sanofi Genzyme.