Maxime Verhoeven

178 Chapter 9 a significant predictor for ACR50 response at 6 months, see Table 3. In all analyses the association of the short-term change in OST-score with treatment response was not modified by RA stage as tested in the models (p=0.44/p=0.22 and p=0.20/p=0.30 for ACR50 and EULAR good response at 3/6 months, respectively). Table 3 Odds ratio’s for predictive ability of variables on treatment response. Parameter OR (95%CI) multivariable OR (95%CI) multivariable EULAR good response at month 3 Baseline OST-score 0.93 (0.83 – 1.04) ∆OST-score at month 1 1.04 (0.90 – 1.19) Baseline DAS28 0.76 (0.41 – 1.41) ∆DAS28 at month 1 4.47 (1.73 – 11.58)* RA stage 0.20 (0.07 – 0.62)* 0.41 (0.12 – 1.37) EULAR good response at month 6 Baseline OST-score 0.95 (0.93 – 1.10) ∆OST-score at month 1 1.07 (0.89 – 1.27) Baseline DAS28 0.71 (0.34 – 1.49) ∆DAS28 at month 1 3.05 (1.05 – 8.90)* RA stage 0.03 (0.01 – 0.13)* 0.04 (0.01 – 0.19) ACR50 response at month 3 Baseline OST-score 0.93 (0.82 – 1.05) ∆OST-score at month 1 1.10 (0.94 – 1.31) Baseline DAS28 0.96 (0.55 – 1.78) ∆DAS28 at month 1 1.77 (0.81 – 3.88) RA stage 0.22 (0.06 – 0.75)* 0.35 (0.10 – 1.22) ACR50 response at month 6 Baseline OST-score 0.94 (0.84 – 1.05) ∆OST-score at month 1 1.00 (0.88 – 1.14) Baseline DAS28 0.83 (0.47 – 1.49) ∆DAS28 at month 1 3.92 (1.57 – 9.28)* RA stage 0.27 (0.09 – 0.82)* 0.58 (0.18 – 1.87) *= statistically significant (OR 95%CI not including 1). OST= optical spectral transmission; DAS28= disease activity score assessing 28 joints; OR= Odds ratio; ∆= change; RA stage= early (starting conventional synthetic DMARD therapy) is reference. DISCUSSION In this first longitudinal study of the HandScan, the concurrent correlations of HandScan (expressed as OST-score) and DAS28 were in general low-to-moderate, consistent with data of a previous cross-sectional study. 8 Although we established a longitudinal