Skip to main content

Table 4 Univariate and EDSS adjusted associations between key-parameters and performance at the Neuro Assessment Lab, including patient characteristics

From: Going beyond PA: Assessing sensorimotor capacity with wearables in multiple sclerosis—a cross-sectional study

Parameter

Dimension

Univariate regression

EDSS adjusted

PEAKSTD

Upper-limb

GRIP r = 0.44, p < 0.01

PEG r = 0.25, p = 0.028

TAP r = 0.25, p = 0.029

GRIP β = 0.33, p < 0.01

Gait

LIMP r = − 0.24, p = 0.040

NORM10 r = − 0.27, p = 0.019

MAX10 r = − 0.29, p = 0.010

Information processing capacity

Clinical estimate & age

EDSS r = − 0.31, p < 0.01

CADENCE

Upper-limb

GRIP r = 0.38, p < 0.01

PEG r = 0.43, p < 0.01

TAP r = 0.44, p < 0.01

DOT r = − 0.40, p < 0.01

TAP β = 0.25, p = 0.042

DOT β = − 0.40, p < 0.01

Gait

LIMP r = − 0.25, p = 0.032

COMPL r = 0.35, p < 0.01

FREQ r = 0.55, p < 0.01

NORM10 r = − 0.47, p < 0.01

MAX10 r = − 0.50, p < 0.01

TSUG r = − 0.51, p < 0.01

Information processing capacity

RT r = − 0.24, p = 0.034

Clinical estimate and age

EDSS r = − 0.57, p < 0.01

WSS r = − 0.43, p < 0.01

AGE r = − 0.40, p < 0.01

FM r = − 0.34, p < 0.01

FATREV

Upper-limb

GRIP r = 0.32, p < 0.01

PEG r = 0.50, p < 0.01

TAP r = 0.57, p < 0.01

DOT r = − 0.55, p < 0.01

GRIP β = 0.57, p < 0.01

PEG β = 0.24, p = 0.034

TAP β = 0.57, p < 0.01

DOT β = − 0.55, p < 0.01

Gait

ASYM r = − 0.24, p = 0.038

NOISE r = 0.37, p < 0.01

COMPL r = 0.56, p < 0.01

FREQ r = 0.63, p < .01

NORM10 r = − 0.62, p < 0.01

MAX10 r = − 0.67, p < 0.01

TSUG r = − 0.69, p < .01

Information processing capacity

RT r = − 0.37, p < 0.01

GNG r = − 0.27, p = 0.018

RT β = − 0.37, p < 0.01

GNG β = − 0.27, p = 0.018

Clinical estimate and age

EDSS r = − 0.58, p < 0.01

WSS r = − 0.63, p < 0.01

AGE r = − 0.36, p < 0.01

FM r = − 0.35, p < 0.01

LAT r = − 0.24, p = 0.034

  1. “EDSS adjusted” lists parameters that can be predicted by the respective TUM:W parameter with a model including the EDSS
  2. PEAKSTD: standard deviation of angular velocity peaks, higher values indicate higher intensities of upper-limb use, CADENCE: average steps/min in case of gait, higher values indicate less walking fragmentation, PEG: 1/trial duration, higher values indicate better performance; FATREV: bouts > 30s/> 5s, higher values indicate lower fragmentation