Table 2.
Associations between mSHS, joint space narrowing score, erosion score and frequency of CD21−/low double-negative B cells as well as age
| Dependent variable | Predictors | Parameter estimate (95% CI) | p-value |
|---|---|---|---|
| mSHSa | Intercept | − 28.55 (− .44.34; − 12.76) | 0.0009 |
| CD21- DN cells | 0.30 (− 0.03; 0.62) | 0.07 | |
| Age | 0.50 (0.28; 0.72) | < 0.0001 | |
| JSNb | Intercept | − 21.88 (− 34.45; − 9.31) | 0.0013 |
| CD21- DN cells | 0.29 (0.03; 0.54) | 0.03 | |
| Age | 0.31 (0.13; 0.48) | 0.001 | |
| ESc | Intercept | − 6.35 (− 11.76; − 0.94) | 0.02 |
| CD21- DN cells | − 0.02 (− 0.13; 0.09) | 0.71 | |
| Age | 0.20 (0.13; 0.28) | < 0.001 | |
| Smoking | 5.13 (0.22; 10.04) | 0.04 |
JSN joint space narrowing score, ES erosion score, mSHS modified Sharp van der Heijde score
amSHS model: adjusted R2 = 51%, i.e. 51% of the variation in mSHS is explained by the B cell population and age
bJNS model: adjusted R2 = 20%, i.e. CD21-DN cells explain 20% of the variation in JNS
cES model: adjusted R2 = 46%, i.e. age and smoking explain 46% of the variation in ES