Real-world, Device-based Peak Cadence Metrics in Black and White Adults with Multiple Sclerosis

Abstract

Background:

Walking performance is worse in Black than White adults with multiple sclerosis (MS), and that difference might extend into real-world, device-based measurements.

Objective:

The current study examined peak 30-minute cadence(Peak-30_CAD) and peak 1-minute cadence(Peak-1_CAD) in Black and White adults with MS.

Methods:

Participants with MS who self-identified as Black(n = 43) or White(n=96) completed the Timed 25-Foot Walk(T25FW) and Six-Minute Walk(6MW), and wore an accelerometer on a belt around the waist during the waking hours of the day for a 7-day period.

Results:

Peak-30_CAD and Peak-1_CAD along with T25FW and 6MW were significantly lower in Black than White participants when controlling for imbalances in age, sex, education, and disease duration, with the differences approaching a moderate magnitude. The differences in Peak-30_CAD and Peak-1_CAD were smaller and non-significant in additional sensitivity analyses further controlling for number of days of valid accelerometer data. Peak-30_CAD and Peak-1_CAD had strong correlations with T25FW and 6MW in Black and White participants.

Conclusion:

Peak-30_CAD and Peak-1_CAD were associated with laboratory-based walking performance in both Black and White adults with MS and may provide complementary real-world mobility metrics.

Introduction

Multiple sclerosis(MS) has a prevalence of nearly 1 million adults in the United States, and occurs most frequently among White and Black adults(1). The clinical course of MS is worse in Black than White adults(2) as reflected by reduced laboratory-based walking performance on Timed 25-Foot Walk(T25FW) and 6-minute Walk(6MW)(3,4) that might extend into real-world, device-based measurements that reflect walking capacity.

There is increasing evidence that step-based metrics from accelerometry may provide real-world measures of walking capacity in MS. One seminal study reported strong associations between peak 30-minute(Peak-30_CAD) and peak 1-minute(Peak-1_CAD) cadence with T25FW and 6MW, even after controlling for daily steps(i.e., real-world mobility behavior), in MS(5). Those results have been replicated in other studies(6,7).

This brief report examined Peak-30_CAD and Peak-1_CAD in samples of Black and White adults with MS. We compared Peak-30_CAD and Peak-1_CAD along with T25FW and 6MW for hypothesized consistency of worse walking outcomes in Black than White adults with MS. We examined and compared the hypothesized strong, bivariate and partial correlations between Peak-30_CAD and Peak-1_CAD with T25FW and 6MW in Black and White adults with MS.

Method

Participants

The convenience sample included participants with MS who self-identified as Black(n=43) or White(n=96) and satisfied the inclusion criteria: (a)MS diagnosis, (b)age between 20–79 years, (c)no relapses within the last 30 days, (d)ability to walk with or without assistance, and (e)willingness to complete the study procedures. The sample was sufficient for detecting a moderate difference between groups (d=.5)(3,4) based on alpha of .05, beta of .20, and 1:2 Black:White sample ratio(1) using G*Power 3.1.

Procedure

The procedure was approved by an Institutional Review Board. Participants provided written informed consent, completed an in-person visit for the collection of demographic and clinical information and the administration of walking performance assessments, and then wore an accelerometer during free-living conditions.

Measures

Participants wore ActiGraph GT3X accelerometers(Pensacola,FL,USA) on an elastic belt around the waist over the non-dominant hip during waking hours over a 7-day period. The data were sampled at 100 Hz and downloaded as 1-minute epochs. We applied the Troiano wear-time algorithm for identifying non-wear time, and defined a valid day as 10+ hours of wear time. The 1-minute epochs were processed into step-based metrics using custom R scripts. Peak-30_CAD(steps/min) was computed by (a)rank-ordering steps/min values within each day, (b)averaging the highest 30 non-consecutive values within each day, and (c)averaging the values across valid days(5–7). Peak-1_CAD(steps/min) was computed by averaging the highest single steps/min value within each day across valid days(5–7). Daily steps(steps/d) was computed by averaging total daily step counts across valid days. All participants had 2 or more valid days of data and were included in the data analysis.

The T25FW assessed walking speed(ft/sec) in MS(8). The 6MW measured walking endurance(ft) in MS(9).

Data analysis

The data were analyzed using IBM SPSS Statistics. Descriptive values for continuous, ordinal, and categorical variables were presented as mean(SD), median(IQR), and number(percentage), respectively. Independent samples t-tests and Chi-square tests compared samples on demographic and clinical variables, and independent samples t-tests were further undertaken on number of valid accelerometer days and daily accelerometer wear time. Analysis of covariance examined differences between samples in peak cadence(Peak-30_CAD,Peak-1_CAD), daily steps, and walking performance(T25FW,6MW), controlling for any imbalances in demographic and clinical variables, and we performed additional sensitivity analyses for peak cadence and daily steps controlling for any imbalances in accelerometer compliance variables (valid days and wear time). The magnitude of differences was interpreted based on Cohen’s thresholds–small(|d|=0.2), moderate(|d|=0.5), and large(|d|=0.8)(10). Bivariate correlations between peak cadence(Peak-30_CAD,Peak-1_CAD), daily steps, and walking performance(T25FW,6MW) were examined using Pearson’s correlation coefficients with 95% confidence intervals, and we compared the bivariate correlations between groups using Fisher’s r-to-z transformation. As additional sensitivity analyses, partial correlations between peak cadence(Peak-30_CAD,Peak-1_CAD) and physical performance(T25FW,6MW), controlling for daily steps, number of valid accelerometer days, and mean daily accelerometer wear time in separate analyses, were examined using Pearson’s partial correlation coefficients with 95% confidence intervals. The magnitude of correlation coefficients was interpreted based on Cohen’s thresholds–weak(|r|=0.1), moderate(|r|=0.3), and strong(|r|=0.5)(10).

Results

Descriptive statistics for demographic and clinical variables are in Table 1. There were statistically significant differences in age, sex, years of education, and disease duration between Black and White participants with MS.

Table 1.

Descriptive statistics for demographic and clinical variables for samples of Black and White adults with multiple sclerosis.

Variables	Black (n = 43)	White (n = 96)	p-value
Age (years)	43.5 (11.7)	54.4 (12.5)	<.001
Sex (n/% Female)	38/88.4%	69/71.9%	<.05
Education (y)	16.0 (2.6)	16.8 (2.1)	<.05
MS type (n/% RRMS)	39/90.7%	84/87.5%	.80
Disease duration (years)	10.8 (7.5)	15.2 (10.5)	<.05
Assistive Device Use (n/% using device)	7/16.3%	15/15.6%	.61
PDDS (0–8)	0 (3)	1 (3)	.29

Note: Values represent mean (SD) and median (IQR). RRMS, relapsing-remitting multiple sclerosis. PDDS, Patient Determined Disease Steps scale.

There was a statistically significant(p<.001) difference in number of valid days of accelerometer data for the Black and White participants with means(SD; range) of 5.1(1.6; 2–7) and 6.2(1.1; 2–7) valid days, respectively. Wear time did not differ significantly(p=.69) between the Black and White participants with means(SD; range) of 811.4(83.5; 664.0–1129.1) and 817.6(85.2; 644.4–1045.0) minutes/day, respectively.

Descriptive statistics for peak cadence, daily steps, and walking performance are in Table 2. There were consistent, statistically significant differences in peak cadence, daily steps, and walking performance between Black and White participants, controlling for age, sex, years of education, and disease duration. The differences in those outcomes approached Cohen’s threshold for a moderate effect size(|d|=0.5). The sensitivity analyses on peak cadence metrics controlling for age, sex, years of education, and disease duration along with number of valid days of accelerometer data indicated smaller (Cohen’s d values of −0.36) and non-significant (p-values of .07) differences between the Black and White participants. The same result was observed in the sensitivity analysis for daily steps (p-value of .18; Cohen’s d of −0.27).

Table 2.

Descriptive statistics and effect sizes for peak cadence metrics, daily steps, and walking performance in samples of Black and White adults with multiple sclerosis.

Variables	Black (n = 43)	White (n = 96)	p-value	Cohen’s d
Peak-30 Cadence (steps/min)	45.5 (38.4, 52.6)	56.6 (52.1, 61.2)	<.05	−0.48
Peak-1 Cadence (steps/min)	75.0 (66.9, 83.2)	87.7 (82.5, 92.9)	<.05	−0.48
Daily Steps (steps/d)	4286 (3532, 5040)	5436 (4955, 5917)	<.05	−0.47
T25FW (ft/s)	4.9 (4.4, 5.5)	5.7 (5.3, 6.0)	<.05	−0.43
6MW (ft)	1370 (1233, 1507)	1568 (1481, 1655)	<.05	−0.45

Note: Values represent adjusted mean scores (95% confidence interval) controlling for age, sex, education, and disease duration. T25FW, timed 25-foot walk; 6MW, 6-minute walk.

Bivariate and partial correlations between peak cadence and daily steps and walking performance are in Table 3. Peak-30_CAD and Peak-1_CAD had strong correlations(|r|>0.5) with T25FW, 6MW, and daily steps in Black and White participants, and there were no statistically significant differences in the correlation coefficients between samples(all p>.24). Peak-30_CAD and Peak-1_CAD generally had strong correlations(|r|>0.50) with T25FW and 6MW when controlling for daily steps in Black and White participants. Peak-30_CAD and Peak-1_CAD had strong correlations(|r|>0.50) with T25FW and 6MW when controlling for number of valid days of accelerometer data and accelerometer wear time in Black and White participants.

Table 3.

Correlations between peak cadence metrics, daily steps, and walking performance in samples of Black and White adults with multiple sclerosis.

Variables	Black (n = 43)		White (n = 96)
	Peak-30CAD (steps/min)	Peak-1CAD (steps/min)	Peak-30CAD (steps/min)	Peak-1CAD (steps/min)
Bivariate Correlations
Daily Steps (steps/d)	.77 (.61,.87)	.67 (.46,.81)	.83 (.76,.88)	.72 (.61,.80)
T25FW (ft/s)	.56 (.31,.74)	.61 (.38,.77)	.67 (.54,.77)	.73 (.62,.81)
6MW (ft)	.63 (.41,.78)	.68 (.48,.81)	.67 (.54,.77)	.74 (.63,.82)
Partial Correlations (Controlling for Daily Steps)
T25FW (ft/s)	.50 (.23,.70)	.59 (.35,.76)	.48 (.31,.62)	.59 (.44,.71)
6MW (ft)	.51 (.25,.70)	.63 (.41,.78)	.47 (.30,.61)	.61 (.47,.72)
Partial Correlations (Controlling for Number of Valid Days of Accelerometer Data)
T25FW (ft/s)	.56 (.31,.74)	.61 (.38,.77)	.64 (.50,.75)	.71 (.59,.80)
6MW (ft)	.63 (.41,.78)	.68 (.48,.81)	.64 (.50,.75)	.72 (.61,.80)
Partial Correlations (Controlling for Accelerometer Wear Time)
T25FW (ft/s)	.54 (.29,.72)	.60 (.37,.76)	.63 (.48,.74)	.71 (.59,.80)
6MW (ft)	.62 (.39,.78)	.67 (.46,.81)	.63 (.48,.74)	.72 (.61,.80)

Note: Values represent bivariate and partial correlation coefficients (95% confidence interval), and all bivariate and partial correlation coefficients are statistically significant at p<.001. T25FW, timed 25-foot walk; 6MW, 6-minute walk.

Discussion

Walking performance based on the T25FW and 6MW is worse in Black than White adults with MS(3,4), and we observed differences that approached a moderate magnitude for worse Peak-30_CAD and Peak-1_CAD, along with worse T25FW and 6MW, in Black than White adults with MS, controlling for age, sex, years of education, and disease duration. We conducted an additional sensitivity analysis that further controlled for number of days of valid accelerometer data, and we observed smaller, non-significant differences in peak cadence metrics, indicating the compliance metrics are important when examining group differences in those metrics. The apparent differences in both laboratory and real-world measures may reflect a complex interplay of disease severity, health disparities, socioeconomic factors, environmental context, and differential exposure of structural inequities(11). Those small-to-moderate and consistent differences might further reflect residual confounding by unmeasured social, environment, clinical, and treatment-related factors.

We reported strong bivariate correlations between Peak-30_CAD and Peak-1_CAD with T25FW and 6MW that were not statistically different between Black and White adults with MS. The correlations generally remained strong in sensitivity analyses controlling for daily steps, the number of valid days of accelerometer data, and accelerometer wear time in Black and White participants. Those results were consistent with prior research(5–7) and support Peak-30_CAD and Peak-1_CAD as real-world measures of walking capacity, independent of mobility behavior based on daily steps, in the context of daily life among Black and White adults with MS.

There are limitations of this study. The limitations include the convenience sample and cross-sectional design along with modest Black sample size and mild disability overall, based on the median PDDS scores, that limits generalizability among those with higher disability levels. The sample predominantly had a relapsing-remitting clinical course of MS that limits generalizability among progressive disease courses. We did not include direct measures of structural racism and social determinants of health.

Conclusion

Our results may support the inclusion of the step-based metrics(Peak-30_CAD and Peak-1_CAD) in clinical and research practice involving racially diverse samples that include Black and White adults with MS. This might be salient for interventions that target ambulation in MS, as such interventions often involve pharmacological or rehabilitation-based interventions focusing on laboratory-based measures of walking performance(12), but the interventions have limited inclusion of real-world, device-based measurements(5) and diverse samples regarding race(13). We offer an approach that can extend understanding of interventions for improving ambulation in racially diverse samples of Black and White adults with MS, and this could be buttressed by patient-reported outcome measures(PROMs)(14). The combination of Peak-30_CAD and Peak-1_CAD using accelerometry over a 7-day period during free-living conditions with PROMs of walking might provide a fuller picture of intervention effects on walking capacity under real-world conditions.

Availability of Data and Materials –

The dataset used and analyzed during the current study is available from the corresponding author on reasonable request.