Skip to main content
. 2014 Aug 15;6:20–25. doi: 10.1016/j.nicl.2014.06.009

Table 2.

Fixed effect estimates for the general linear mixed model of MTR in ∆MTR lesions for subjects with RRMS. The model is significant with 9022 samples from 120 subject-timepoints in 18 subjects, χ2 = 5925 on 5 degrees of freedom, p < 0.0001. The ‘:’ in an effect denotes an interaction. dDF is the estimated denominator degrees of freedom. Predictions are included to help interpretation but statistics apply to the effect estimates.

RRMS subjects, 120 timepoints, 1037 lesions, 9022 samples
χ2 = 5925; 5 DF; p < 0.0001; R2 marginal
 = 0.45; R2 conditional = 0.67
Effect Prediction
(nMU)		 Estimate
(nMU)		 Std
Error		 dDF f p
Intercept
(de novo lesions, baseline)		 0.45 0.45 0.053
Acute −0.85 −1.29 0.016 8072 4087 <0.0001
Post-recovery 0.13 −0.31 0.014 7985 146 <0.0001
Repeat lesions −0.29 −0.74 0.025 5171 865 <0.0001
Repeat lesions: acute −1.31 0.28 0.037 8119 20.6 <0.0001
Repeat lesions: follow-up −0.49 0.12 0.030 8789 3.34 0.068
SPMS subjects, 221 timepoints, 1283 lesions, 5548 samples
χ2 = 4794; 5 DF; p < 0.0001; R2 marginal
 = 0.47; R2 conditional = 0.78
Effect Prediction
(nMU)		 Estimate
(nMU)		 Std
Error		 dDF f p
Intercept
(de novo lesions, baseline)		 0.51 0.51 0.051
Acute −0.99 −1.51 0.025 4295 4485 >0.0001
Post-recovery −0.30 −0.81 0.030 4410 700 >0.0001
Repeat lesions −0.74 −1.25 0.029 5144 1612 >0.0001
Repeat lesions: acute −1.82 0.42 0.033 4323 20.4 >0.0001
Repeat lesions: follow-up −1.25 0.29 0.039 4655 5.70 0.017