Figure 2.

CombiWISE measures multiple sclerosis (MS) progression more accurately than Expanded Disability Status Scale (EDSS). (A) Patient-specific progression rates measured by EDSS and CombiWISE regression slopes strongly correlate. (B) Frequency distribution of measured CombiWISE slopes for patients with no reported change in EDSS [highlighted blue points in (A)]. The histogram demonstrates the insensitivity of EDSS; patients can progress as much as 4 CombiWISE units/year (or improve by up to 2 CombiWISE units/year) with no change in EDSS. (C) Boxplots depict the proportion of variance explained for CombiWISE versus EDSS regressions. The median is indicated with a solid black line, and the mean is indicated with a dashed blue line. The bottom and top of the box denotes the first and third quartiles, respectively. The whiskers indicate the 10th and 90th percentiles. A two-tailed paired t-test indicated strong evidence of a mean difference of 0.1357 (95% CI, 0.089–0.182) in R² values between the CombiWISE and EDSS regressions. (D) Histogram of difference in R² values between CombiWISE and EDSS regressions suggests that the values are relatively normally distributed. (E) Representative patient examples illustrate higher accuracy of linear regressions (solid blue lines with dashed 95% CI; upper panels) derived from longitudinal CombiWISE measurements (no fill circles; upper panels) in comparison to linear regressions (solid black lines; lower panels) derived from longitudinal EDSS measurements (black filled circles; lower panels) from the same three MS patients.