Table 3 Chi-square goodness of fit statistic for HARP2, ALTA, EDEN and SAILS
From: Re-analysis of ventilator-free days (VFD) in acute respiratory distress syndrome (ARDS) studies
Study | Treatment group | |||
|---|---|---|---|---|
HARP-2 | Placebo | Simvastatin | Estimate (95% CI) | p-value |
N | 279 | 258 | ||
Logit-Poisson hurdle model | ||||
Logistic sub-model (n (%))a | 170 (61.0%) | 180 (69.8%) | 1.47 (1.03 to 2.12) | 0.03 |
Count sub-model (Mean ± SD)b | 18.7 ± 6.4 | 17.8 ± 6.8 | 0.96 (0.91 to 1.01) | 0.09 |
ALTA | Placebo | Albuterol | Estimate (95% CI) | p-value |
N | 130 | 152 | ||
Logit-Poisson hurdle model | ||||
Logistic sub-model (n (%)) | 102 (78.5%) | 103 (67.8%) | 0.55 (0.32 to 0.95) | 0.03 |
Count sub-model (mean ± SD) | 21.1 ± 5.6 | 21.2 ± 6.0 | 1.01 (0.95 to 1.07) | 0.70 |
EDEN | Full | Tropic | Estimate (95% CI) | p-value |
N | 492 | 508 | ||
Logit-Poisson hurdle model | ||||
Logistic sub-model (n (%)) | 351 (71.3%) | 345 (67.9%) | 0.89 (0.68 to 1.17) | 0.40 |
Count sub-model (mean ± SD) | 21 ± 5.5 | 21.6 ± 5.4 | 1.02 (0.99 to 1.06) | 0.11 |
SAILS | Placebo | Rosuvastatin | Estimate (95% CI) | p-value |
N | 366 | 379 | ||
Logit-Poisson hurdle model | ||||
Logistic sub-model (n (%)) | 251 (68.6%) | 264 (69.7%) | 1.03 (0.75 to 1.40) | 0.88 |
Count sub-model (mean ± SD) | 21.9 ± 5.2 | 21.6 ± 5.1 | 0.99 (0.95 to 1.03) | 0.60 |