Table 4 Final analyses based on the resulting sample collected from each design
From: Using Bayesian adaptive designs to improve phase III trials: a respiratory care example
| Â | OSCAR trial | Design 2 | Design 3 | Design 5 |
|---|---|---|---|---|
Primary outcome (control; HFOV) | 163/397 (41.1%); 166/398 (41.7%) | 154/377 (40.8%); 156/373 (41.8%) | 138/339 (40.7%); 134/331 (40.5%) | 156/380 (41.1%); 157/375 (41.9%) |
RR (95% CI) | 1.02 (0.86, 1.20) | 1.02 (0.86, 1.22) | 0.99 (0.83, 1.20) | 1.02 (0.86, 1.21) |
Posterior probability HFOV superior | 0.46 | 0.40 | 0.53 | 0.40 |
Number of deaths in trial | 329 | 310 | 272 | 313 |
Number randomiseda | 795 | 750 | 670 | 775 |
Recruitment savings from original sample size of N = 1006 | 211 | 256 | 336 | 231 |
Recruitment savings from achieved OSCAR sample size of N = 795 | NA | 45 | 125 | 20 |
Accrual Duration (weeks)b | 243 | 227 | 203 | 228 |