Analysis to determine treatment differences between sites (comparing the 2 largest sites with the best outcomes and other sites). Linear and logistic mixed models to test whether each treatment comparison was associated with DAS28 changes after adjusting for covariates (female sex, baseline HAQ, baseline DAS28), divided by sites with best disease outcomes (2 largest sites B and H). Regression models for treatment and outcome were significant with site having an effect.
| Treatment | DAS28 Change 0–6 Months¥ B (p value) | DAS28 Change from 0–12 Months¥ B (p value) | DAS28 Remission at Last Visit‡ OR (p value) |
|---|---|---|---|
| Mean no. DMARD taken | |||
| Site B | 0.0480 (0.742) | −0.2271 (0.145) | 1.3747 (0.144) |
| Site H | 0.6483 (0.003)* | 0.2500 (0.238) | 0.6262 (0.301) |
| Medication changed ever | |||
| Site B | 0.9543 (0.001)* | 0.6524 (0.004)* | 0.4324 (0.014)* |
| Site H | 0.4499 (0.107) | 0.7533 (0.003)* | 0.4574 (0.150) |
| Medication increased ever | |||
| Site B | 1.0695 (0.001)* | 0.6696 (0.005)* | 0.3909 (0.007)* |
| Site H | 0.7809 (0.013)* | 0.9349 (0.001)* | 0.2520 (0.027)* |
| MTX monotherapy ever^ vs MTX combined with at least 1 other DMARD or biologic ever | |||
| Site B | 0.0693 (0.818) | −0.3975 (0.229) | 1.7805 (0.206) |
| Site H | 0.7934 (0.001)* | 0.2957 (0.196) | 0.6463 (0.387) |
| MTX SC ever | |||
| Site B | 0.5212 (0.044)* | 0.2789 (0.315) | 1.3407 (0.448) |
| Site H | 0.0594 (0.813) | 0.0715 (0.759) | 0.7875 (0.676) |
| Single DMARD ever^ vs combination of DMARD and/or biologics ever | |||
| Site B | 0.2362 (0.365) | −0.1956 (0.481) | 1.6738 (0.188) |
| Site H | 0.6515 (0.002)* | 0.3139 (0.118) | 0.6935 (0.389) |
| Single DMARD ever^ vs combination of DMARD ever | |||
| Site B | −0.0662 (0.792) | −0.3676 (0.193) | 2.0896 (0.071) |
| Site H | 0.6515 (0.002)* | 0.3139 (0.118) | 0.6935 (0.389) |
B is largest site, H is second largest. Wald test of B coefficient used in the model if significant at p < 0.1.
↵¥ Includes all study participants with measures of DAS28 at both visits. Linear mixed models.
↵‡ DAS28 < 2.6 = remission, includes only those study participants with baseline DAS28 > 2.6 (not in remission). Generalized linear mixed models, binary logistic with logit link.
↵* p < 0.05.
↵^ Reference category. Ever: patient was prescribed the treatment at any visit up to the 9-month visit; B: ß coefficient; MTX: methotrexate; DMARD: disease-modifying antirheumatic drug; SC subcutaneous; DAS28: 28-joint Disease Activity Score; HAQ: Health Assessment Questionnaire.