Variable, unit (range) | Strategy I (n=107) | Strategy II (n=101) | Strategy III (n=105) | Dropouts (n=31) | Patients who were not randomly assigned (n=52) |
Female sex, % | 69 | 65 | 64 | 52 | 81 |
Rheumatoid factor positive, % | 67 | 56 | 65 | 55 | 69 |
Age, years | |||||
Mean | 56 | 55 | 57 | 671-151 | 55 |
10–90 Centiles | 37–74 | 35–73 | 37–73 | 38–79 | 30–73 |
Primary end points | |||||
Disability score (0–3) | |||||
Mean | 1.4 | 1.3 | 1.3 | 1.5 | 1.01-151 |
10–90 Centiles | 0.5–2.5 | 0.1–2.4 | 0.3–2.4 | 0.4–2.6 | 0.1–2.1 |
Pain score, mm (0–100) | |||||
Mean | 46 | 43 | 44 | 43 | 311-151 |
10–90 Centiles | 9–86 | 3–82 | 9–92 | 1–96 | 1–72 |
Joint score (0–534) | |||||
Mean | 141 | 147 | 142 | 152 | 110 |
10–90 Centiles | 35–306 | 38–280 | 44–279 | 47–357 | 20–226 |
ESR, mm/1st h (0–140) | |||||
Mean | 42 | 41 | 43 | 41 | 39 |
10–90 Centiles | 11–86 | 9–88 | 14–88 | 14–87 | 8–92 |
Radiological damage (0–448) | |||||
Median | 2 | 2 | 2 | 2 | 2 |
10–90 Centiles | 0–12 | 0–11 | 0–13 | 0–6 | 0–10 |
Secondary end points | |||||
Wellbeing, mm (0–100) | |||||
Mean | 49 | 50 | 49 | 48 | 39 |
10–90 Centiles | 6–90 | 6–82 | 12–92 | 12–92 | 2–80 |
Grip strength, kPa (0–120) | |||||
Mean | 30 | 33 | 31 | 24 | 35 |
10–90 Centiles | 4–58 | 6–67 | 5–58 | 1–43 | 9–59 |
Haemoglobin, mmol/l | |||||
Mean | 7.8 | 8.0 | 8.0 | 8.1 | 7.8 |
10–90 Centiles | 6.5–9.0 | 6.9–9.1 | 6.8–9.0 | 7.1–9.2 | 6.7–9.0 |
Morning stiffness, min (0–720) | |||||
Median | 60 | 60 | 60 | 60 | 30 |
10–90 Centiles | 10–360 | 5–180 | 3–360 | 1–288 | 0–162 |
C reactive protein, mg/l | |||||
Median | 26 | 20 | 16 | 18 | 18 |
10–90 Centiles | 0–94 | 0–85 | 0–72 | 0–79 | 0–70 |
Platelet count × 109/l | |||||
Median | 321 | 332 | 317 | 330 | 317 |
10–90 Centiles | 215–503 | 220–482 | 219–483 | 218–436 | 243–522 |
↵1-150 Higher values indicate more active disease, with the exception of values for grip strength and haemoglobin.
Mean was presented for normal distribution of the variables, and median for a skew distribution. Presented ranges are theoretical ranges.
↵1-151 Dropouts had a significantly higher age than all other groups (analysis of variance (ANOVA), p = 0.002).
Patients who were not randomly assigned differed significantly from all other groups: less disability (ANOVA, p = 0.05) and lower pain score (ANOVA, p = 0.04).