Continuous daily assessment of multiple sclerosis disability using remote step count monitoring

Abstract

Disability measures in multiple sclerosis (MS) rely heavily on ambulatory function, and current metrics fail to capture potentially important variability in walking behavior. We sought to determine whether remote step count monitoring using a consumer-friendly accelerometer (Fitbit Flex) can enhance MS disability assessment. 99 adults with relapsing or progressive MS able to walk ≥2-min were prospectively recruited. At 4 weeks, study retention was 97% and median Fitbit use was 97% of days. Substudy validation resulted in high interclass correlations between Fitbit, ActiGraph and manual step count tally during a 2-minute walk test, and between Fitbit and ActiGraph (ICC = 0.76) during 7-day home monitoring. Over 4 weeks of continuous monitoring, daily steps were lower in progressive versus relapsing MS (mean difference 2546 steps, p < 0.01). Lower average daily step count was associated with greater disability on the Expanded Disability Status Scale (EDSS) (p < 0.001). Within each EDSS category, substantial variability in step count was apparent (i.e., EDSS = 6.0 range 1097-7152). Step count demonstrated moderate-strong correlations with other walking measures. Lower average daily step count is associated with greater MS disability and captures important variability in real-world walking activity otherwise masked by standard disability scales, including the EDSS. These results support remote step count monitoring as an exploratory outcome in MS trials.

Keywords: Accelerometer; Multiple sclerosis; Outcome measurement; Progressive; Relapsing; Remote physical activity monitoring.

Fig 1

Bland–Altman plots to assess agreement between measurement techniques. a–c Comparisons of steps counted during 2-min walk test as recorded by manual counting, Fitbit Flex, and ActiGraph. d Bland–Altman plots for Fitbit Flex and ActiGraph steps per day as recorded over 7 days of activity monitoring in the home environment. The solid black line represents the mean difference between the measures and the black dashed horizontal lines represent mean difference ±2 standard deviations. Each point corresponds to a separate individual (a–c) and separate day (d). The solid lighter gray line indicates where ‘0’ difference between the measures would lie. The gray dashed lines are the 95% limits of agreement

Fig 2

Evolution of daily step count and of the reliability of monitoring over 28 days. a Average Steps (via Fitbit Flex) per Day, over initial 28 days. The number of steps taken each day for each individual participant is represented in light grey dots. The black line depicts the average number of steps per day, across all participants, with the variability denoted by the error bars. No reactivity was observed in our cohort. b For each n number of days, we plot the average of all the correlation measures between an average daily step count computed from a monitoring period of n successive days, and the average daily step count computed from the full 4 weeks monitoring period (ground truth). The correlations are computed with the Pearson r², across all patients

Fig 3

Average daily step count (measured using the Fitbit Flex) is associated with EDSS but also identifies variability of ambulatory activity within EDSS categories. The number of people in each EDSS block is depicted below the box plots. Greater disability (higher EDSS score) is associated with lower average steps per day (ambulatory activity), and lower disability (lower EDSS) is associated with greater ambulatory activity. Variability exists within EDSS blocks, including in those designated EDSS = 6 (need for unilateral support to walk minimal distances)

Fig 4

Greater average daily step count (measured using the Fitbit Flex) is correlated with better performance-based measures of ambulatory function and lower patient-reported impact of MS on ambulatory function. a: Average steps per day are inversely correlated to the time taken to complete the T25FW: faster walking speeds are associated with a greater average daily step count. (q = 0.65, p < 0.001). b: Likewise, average steps per day are positively associated with higher mobility and balance (faster TUG time scores). (q = -0.64, p < 0.001). c The 12-item MS walking scale (MSWS-12) is inversely correlated to the ambulatory activity (average number of steps per day) recorded by the Fitbit Flex. Higher MSWS-12 scores indicate greater impact of MS on a person's walking ability, and correspond to a lower average daily step count. (q = -0.63, p <0.001). The orange shading corresponds to the slope estimate, which is constrained to be linear (red line). a, b Inverse scores were taken to normalize the data and allow for linear regressions. nDay the number of valid days of Fitbit Flex step count data, T25FWInv inverse (1/T25FW) score of the T25FW, TUGInv inverse (1/TUG) score of the TUG

Fig 5

Bivariate associations between average steps per day and all demographic, clinic-based and patient-reported variables. The heat-map (to the right) graphically depicts the direction of correlation, with red tones trending towards positive (+1) and blue tones indicating negative correlations (-1). Correlations are computed using the Spearman's q. The Dendrogram (to the left) is the branching diagram, depicting the strength of the relationships between all of the included outcomes and shows all outcomes hierarchically clustered in relation to the strength of correlation (Spearman's q) regardless of the direction of association. The distance used for clustering here is equal to 1–C^2, with C the correlation. diseaseDuration the duration of the disease, subtype_cat MS subtype/phenotype category (Relapsing or Progressive), EDSSgroup EDSS grouped into blocks, nDay the number of valid days of Fitbit step count data, dist2M distance walked during the 2MW test, STEPS average number of steps/day over 4 weeks.