Article Figures & Data
Figures
- Figure 1.
Predicted probability for joint damage progression for 2 different prediction scores. A. Prediction score based on the regression coefficients in the extended model. B. Prediction score based on the number of prediction factors present in a patient in the simplified model.
- Figure 2.
Discriminative ability of the extended and simplified prediction model for joint damage progression between 0 and 36 months. Areas under the ROC curve were 0.77 (95% CI 0.72–0.81) and 0.75 (95% CI 0.70–0.80), respectively. In the extended model, anti-CCP, ESR, and erosions are categorized, and in the simplified model, these are dichotomized. ROC: receiver-operation characteristic; anti-CCP: anticyclic citrullinated peptide antibodies; ESR: erythrocyte sedimentation rate.
Tables
- Table 1.
Baseline characteristics and medication use of study population (not imputed data), separated by occurrence of joint damage progression between 0 and 3 years. Values are mean (SD), % (n), or median (IQR).
Characteristics Joint Damage Progression No Joint Damage Progression p n Mean/Med/% n Mean/Med/% Age, yrs 175 54 (14) 250 57 (14) 0.029 Female 175 59% (104) 250 66% (165) 0.167 Anti-CCP–positive 158 82% (129) 213 53% (112) < 0.001 RF–positive 175 87% (153) 246 66% (163) < 0.001 SE present 164 75% (123) 188 68% (128) 0.153 Smoking, ever 131 72% (94) 176 67% (118) 0.377 DAS28 170 5.5 (1.3) 241 4.9 (1.4) < 0.001 ESR, mm/h 168 40 (21–59) 238 21 (9–38) < 0.001 CRP, mg/l 118 22 (7–52) 206 3 (0–23) < 0.001 SJC28 173 11 (7–16) 244 10 (6–14) 0.017 SJC44 142 15 (10–22) 153 13 (9–19) 0.041 TJC28 173 7 (4–13) 244 6 (2–12) 0.070 TJC53 142 13 (6–19) 153 12 (5–19) 0.269 VAS pain, 0–100 157 47 (28–60) 228 47 (31–64) 0.241 Erosions at BL 175 62% (108) 250 32% (80) < 0.001 Ratingen score 175 2 (0–4) 250 0 (0–1) < 0.001 DMARD 169 250 0.253 Combination 21% (37) 20% (51) Monotherapy 75% (132) 72% (181) Oral prednisone 175 61% (107) 250 61% (152) 0.943 -
Joint damage progression was defined as ≥ 5 Ratingen points. Treatment with synthetic DMARD and oral prednisone was evaluated between 0 and 36 months. Statistic tests used were unpaired Student t test, chi-square test, and Mann-Whitney U test, as appropriate. IQR: interquartile range; anti-CCP: anticyclic citrullinated peptide antibodies; RF: rheumatoid factor; SE: shared epitope; DAS28: Disease Activity Score at 28 joints; ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; SJC28: swollen joint count at 28 joints; TJC28: tender joint count at 28 joints; VAS: visual analog scale; BL: baseline; DMARD: disease-modifying antirheumatic drug.
-
- Table 2.
Extended multivariable logistic regression model for the prediction of joint damage progression between 0 and 3 years followup. Erosions are according to the Ratingen score. The model was corrected for age and sex.
Variables Values β OR 95% CI p Anti-CCP/RF 0 — — — — 1 0.19 1.20 0.52–2.79 0. 666 2 1.34 3.80 1.92–7.50 < 0.001 Erosions 0 — — — — 1–5 0.95 2.57 1.57–4.23 < 0.001 5–10 1.35 3.87 1.68–8.95 0.002 > 10 2.61 13.55 3.34–54.99 < 0.001 ESR, mm/h < 25 — — — — 25–50 0.99 2.69 1.58–4.59 < 0.001 > 50 1.61 4.98 2.67–9.29 < 0.001 Age < 45 — — — — 45–64 −0.18 0.84 0.47–1.48 0.539 > 64 −0.89 0.41 0.21–9.81 0.010 Sex Female −0.45 0.64 0.40–1.03 0.065 Constant −1.43 0.24 0.08–0.75 0.014 -
Anti-CCP: anticyclic citrullinated peptide antibodies; RF: rheumatoid factor; ESR: erythrocyte sedimentation rate.
-
- Table 3.
Simplified multivariable regression model for prediction of joint damage progression. The 3 baseline prognostic factors were anti-CCP positivity > 25 U/ml, ESR > 25 mm/h, and presence of ≥ 1 erosion. The model was corrected for age and sex.
Variables Values n β OR 95% CI p No. risk factors 0 56 — — — — 1 128 1.36 3.91 1.26–12.09 0.018 2 146 2.93 10.95 3.70–32.40 < 0.001 3 95 3.61 37.01 11.74–116.6 < 0.001 Age < 45 — — — — 45–64 −0.23 0.79 0.45–1.39 0.415 > 64 −0.77 0.46 0.25–0.86 0.015 Sex Female −0.40 0.70 0.42–1.06 0.085 Constant −1.53 0.22 0.06–0.85 0.028 -
Anti-CCP: anticyclic citrullinated peptide antibodies; ESR: erythrocyte sedimentation rate.
-
In this issue
