Table 2. Primary and Secondary Outcome Measures.
| Outcome measure | Prednisolone (n = 134) | Placebo (n = 137) | Treatment effect (95% CI)a | P value |
|---|---|---|---|---|
| Primary outcome measure | ||||
| Proportion of patients experiencing an URTI-related relapse (n = 131 in both arms) | ||||
| No | 75 (57.3) | 73 (55.7) | RD, −0.024 (−0.142 to 0.095) | .70 |
| Yes | 56 (42.7) | 58 (44.3) | RD, 0.96 (0.74 to 1.26) | |
| Secondary outcome measures | ||||
| URTI-related relapse rate (n = 134 in prednisolone arm and 136 in placebo arm) | ||||
| 0 | 78 (58.2) | 78 (57.4) | IRR, 0.98 (0.72 to 1.33) | .88 |
| 1 | 36 (26.9) | 41 (30.1) | ||
| 2 | 15 (11.2) | 10 (7.4) | ||
| 3 | 4 (3.0) | 7 (5.1) | ||
| 4 | 1 (0.8) | 0 (0) | ||
| Proportion of patients experiencing any relapse (URTI and non–URTI related) (n = 132 in both arms) | ||||
| No | 41 (31.1) | 34 (25.8) | RD, −0.05 (−0.16 to 0.06) | .33 |
| Yes | 91 (68.9) | 98 (74.2) | RD, 0.93 (0.80 to 1.09) | |
| Relapse rate (n = 134 in prednisolone arm and 136 in placebo arm) | ||||
| 0 | 43 (32.1) | 38 (27.9) | IRR, 0.89 (0.74 to 1.07) | .23 |
| 1 | 28 (20.9) | 39 (28.7) | ||
| 2 | 24 (17.9) | 24 (17.7) | ||
| 3 | 22 (16.4) | 11 (8.1) | ||
| 4 | 11 (8.2) | 14 (10.3) | ||
| 5 | 6 (4.5) | 5 (3.7) | ||
| ≥6 | 0 (0) | 5 (3.7) | ||
| Proportion of patients who had escalation of background immunosuppressant therapy (n = 130 in prednisolone arm and 128 in placebo arm) | ||||
| No | 72 (55.4) | 71 (55.5) | RD, −0.003 (−0.12 to 0.12) | .96 |
| Yes | 58 (44.6) | 57 (44.5) | RD, 0.98 (0.75 to 1.29) | |
| Proportion of patients who had reduction of background immunosuppressant therapy (n = 128 in prednisolone arm and 129 in placebo arm) | ||||
| No | 73 (57.0) | 67 (51.9) | RD, −0.04 (−0.15 to 0.06) | .42 |
| Yes | 55 (43.0) | 62 (48.1) | 0.91 (0.71 to 1.16) | |
| Cumulative prednisolone dose, median (IQR), mg | 2060 (1128-3355) | 1880 (1115-3295) | Difference, 180 (−301.83 to 661.83) | .72 |
Abbreviations: IRR, incidence rate ratio (an offset was included in the model); RD, risk difference; RR, risk ratio; URTI, upper respiratory tract infection.
a
All treatment effects are from regression models that adjusted for the background therapy at baseline, except for cumulative prednisolone dose, which is based on an unadjusted analysis with the P value from a Wilcoxon rank sum test. An IRR less than 1 favors the prednisolone arm. An RR less than 1 favors the prednisolone arm. A negative RD favors the prednisolone arm.