Table 3 Number and percentage of trials stopping for futility when requiring a statistically significant result to consider a trial ‘successful’
From: An investigation of the impact of futility analysis in publicly funded trials
| Â | Number (percentage) of trials | True successful result | |||||||
|---|---|---|---|---|---|---|---|---|---|
CP stopping boundary 20% | CP stopping boundary 30% | CP stopping boundary 40% | |||||||
Successful trial | Successful trial | Successful trial | |||||||
Yes | No | Yes | No | Yes | No | ||||
Stop if futility analysis done | 50% of patients | Stopped early | Yes | 0 | 0 | 0 | 0 | 0 | 0 |
No | 9 (31%) | 20 (69%) | 9 (31%) | 20 (69%) | 9 (31%) | 20 (69%) | |||
60% of patients | Stopped early | Yes | 0 | 0 | 0 | 3 (12%) | 1 (4%) | 7 (28%) | |
No | 6 (24%) | 19 (76%) | 6 (24%) | 16 (64%) | 5 (20%) | 12 (48%) | |||
70% of patients | Stopped early | Yes | 0 | 5 (21%) | 0 | 11 (46%) | 0 | 12 (50%) | |
No | 5 (21%) | 14 (58%) | 5 (21%) | 8 (33%) | 5 (21%) | 7 (29%) | |||
75% of patients | Stopped early | Yes | 0 | 7 (30%) | 0 | 10 (43%) | 0 | 14 (61%) | |
No | 4 (17%) | 12 (52%) | 4 (17%) | 9 (39%) | 4 (17%) | 5 (22%) | |||
80% of patients | Stopped early | Yes | 0 | 9 (41%) | 0 | 11 (50%) | 0 | 15 (68%) | |
No | 4 (18%) | 9 (41%) | 4 (18%) | 7 (32%) | 4 (18%) | 3 (14%) | |||
90% of patients | Stopped early | Yes | 0 | 11 (55%) | 0 | 12 (60%) | 0 | 16 (80%) | |
| Â | Â | Â | No | 4 (20%) | 5 (25%) | 4 (20%) | 4 (20%) | 4 (20%) | 0 |