Changes in sample size for MOTIVATE and REMIND trials if designed as adaptive trials under varying assumptions
| Scenario (outcome probabilities for each arm in the trial in order, with control listed first) | Mean (SD) sample size per arm | Mean trial duration (months) | |||||
| Control | Arm 1 | Arm 2 | Arm 3 | Arm 4 | Total | ||
| MOTIVATE trial | |||||||
| Original trial design | 114 002 (238) | 22 798 (143) | 22 798 (143) | 34 205 (169) | 34 198 (173) | 228 000 | 4 |
| Same outcome measurement window as original trial and two 30 day interim analyses | |||||||
| Scenario 1 (0.65, 0.75, 0.70, 0.70, 0.65) | 59 170 (1319) | 25 878 (458) | 25 876 (460) | 25 875 (458) | 7611 (87) | 144 411 (87) | 13 |
| Scenario 2 (0.65, 0.75, 0.65, 0.65, 0.65) | 39 457 (12 088) | 13 661 (11 969) | 7628 (150) | 7627 (140) | 7628 (135) | 76 001 (0) | 8.6 |
| Scenario 3 (0.65, 0.70, 0.70, 0.70, 0.70) | 74 464 (6424) | 38 382 (1614) | 38 385 (1615) | 38 383 (1614) | 38 385 (1617) | 228 000 (0) | 26 |
| Shorter outcome measurement window and one 30 day interim analysis | |||||||
| Scenario 1 (0.65, 0.75, 0.70, 0.70, 0.65) | 71 451 (7407) | 21 782 (2472) | 21 786 (2471) | 21 782 (2473) | 11 401 (95) | 148 201 (95) | 2.5 |
| Scenario 2 (0.65, 0.75, 0.65, 0.65, 0.65) | 59 187 (18 148) | 20 612 (18 195) | 11 401 (95) | 11 399 (95) | 11 402 (95) | 114 001 (0) | 3.2 |
| Scenario 3 (0.65, 0.70, 0.70, 0.70, 0.70) | 87 689 (11 893) | 35 074 (2978) | 35 080 (2978) | 35 078 (2976) | 35 079 (2979) | 228 000 (0) | 5 |
| REMIND trial | |||||||
| Original trial design | 3165 (53) | 6332 (66) | 6332 (67) | 6333 (67) | NA | 22 163 | 12 |
| Same outcome measurement window as original trial and two 30 day interim analyses | |||||||
| Scenario 1 (0.02, 0.10, 0.07, 0.02) | 5280 (673) | 3967 (844) | 2275 (785) | 202 (768) | NA | 12 724 (278) | 35.6 |
| Scenario 2 (0.02, 0.10, 0.02, 0.02) | 5258 (715) | 5002 (985) | 1202 (768) | 1203 (768) | NA | 12 665 (0) | 35.4 |
| Scenario 3 (0.02, 0.07, 0.07, 0.07) | 5346 (793) | 2554 (1034) | 2534 (1029) | 2543 (1027) | NA | 12 976 (1507) | 36.3 |
| Shorter outcome measurement window and one 30 day interim analysis | |||||||
| Scenario 1 (0.02, 0.10, 0.07, 0.02) | 6050 (1690) | 1940 (655) | 1932 (655) | 1753 (598) | NA | 11 675 (884) | 6.8 |
| Scenario 2 (0.02, 0.10, 0.02, 0.02) | 5821 (1791) | 1754 (598) | 1753 (598) | 1753 (598) | NA | 11 081 (0) | 6.5 |
| Scenario 3 (0.02, 0.07, 0.07, 0.07) | 6120 (1848) | 1946 (816) | 1949 (822) | 1950 (823) | NA | 11 964 (2316) | 7 |
MOTIVATE=Mail Outreach To Increase Vaccination Acceptance Through Engagement; REMIND=Randomised Evaluation to Measure Improvements in Nonadherence from low cost Devices; SD=standard deviation; NA=not applicable. For each scenario, mean sample sizes per arm and corresponding standard deviations across 10 000 simulations are reported, comparing the original MOTIVATE and REMIND trials’ study designs with a bayesian adaptive study design (see online supplemental section S1).