Table 1 Candidate Bayesian sequential designs generated for OSCAR
From: Using Bayesian adaptive designs to improve phase III trials: a respiratory care example
Design | Interim | Timing of interim (information fraction)a | Can stop for success | Can stop for futility | Success stopping boundariesb | Futility stopping boundariesc |
|---|---|---|---|---|---|---|
1 | NA (fixed design) | NA | NA | NA | NA | NA |
2 | 1 | 250 (1/4) | No | Yes | NA | F1 = 0.05 |
2 | 500 (1/2) | Yes | Yes | S2 = 0.99 | F2 = 0.1 | |
3 | 750 (3/4) | Yes | Yes | S3 = 0.98 | F3 = 0.15 | |
3 | 1 | 335 (1/3) | No | Yes | NA | F1 = 0.05 |
2 | 670 (2/3) | Yes | Yes | S2 = 0.99 | F2 = 0.1 | |
4 | 1 | 335 (1/3) | No | Yes | NA | F1 = 0.05 |
2 | 500 (1/2) | Yes | Yes | S2 = 0.99 | F2 = 0.1 | |
3 | 670 (2/3) | Yes | Yes | S3 = 0.98 | F3 = 0.15 | |
5 | 1 | 503 (1/2) | Yes | Yes | S1 = 0.99 | F1 = 0.05 |
2 | 755 (3/4) | Yes | Yes | S2 = 0.98 | F2 = 0.1 | |
6 | 1 | 503 (1/2) | Yes | Yes | S1 = 0.99 | F1 = 0.05 |
2 | 755 (3/4) | Yes | Yes | S2 = 0.98 | F2 = 0.1 | |
3 | 880 (7/8) | Yes | Yes | S3 = 0.98 | F3 = 0.15 |