Skip to main content
. 2021 May 21;12:666971. doi: 10.3389/fphar.2021.666971

TABLE 5.

Results of robust mixed linear regression model for predictors of anti-dsDNA antibody titer. Individual Patient ID was included as random effect in the model. Predictor values were measured at all timepoints.

Dependent: anti-dsDNA antibody titer
Predictors Estimate Std. error t value Pvalue
(Intercept) 335.7107 76.5060 4.3880 0.0007
Month 6 37.1296 30.4093 1.2210 0.2369
Month 12 −29.3079 30.7385 −0.9535 0.3528
C4 −8.0294 2.1743 −3.6928 0.0027
sBCMA −0.0285 0.0076 −3.7340 0.0015
sCD40L 0.0308 0.0086 3.5974 0.0018
#CD4 T cells 0.4038 0.1212 3.3317 0.0029
#CD8 T cells −0.4150 0.1558 −2.6639 0.0160

MODEL FIT: AIC = 384.81, BIC = 400.93 Pseudo-R 2 (fixed effects) = 0.76 Pseudo-R 2 (total) = 0.87. In bold p ≤ 0.050. sBCMA, soluble B cell maturation antigen; sCD40L, soluble CD40 Ligand; C4: complement factor 4; # absolute number of cells.