Act 4 · Apply
Simulate twice
What changes when local behaviour replaces global defaults.
Sample values. Synthetic study, not real patients.
The same brief, sized two ways and simulated under both sets of inputs
Each figure is the median of simulated trials with an 80% spread, or a proportion with its 95% Monte Carlo interval.
| Scenario | Enrolled | Power | Months to full enrolment | Months to last visit | Retained | Deviations |
|---|---|---|---|---|---|---|
| Design A under global defaults | 330 (214 to 552, 90% sensitivity range) Assumed | 88.2% (86.1% to 90.1%, 95% Monte Carlo) Assumed | 14.8 months (10.7 to 20.8, 80% spread across simulated trials) Assumed | 26.8 months (22.7 to 32.8, 80% spread across simulated trials) Assumed | 80.3% (72.4% to 87.0%, 80% spread across simulated trials) Assumed | 184 (109 to 313, 80% spread across simulated trials) Assumed |
| Design A under the local profile | 330 (214 to 552, 90% sensitivity range) Assumed | 73.7% (70.9% to 76.3%, 95% Monte Carlo) Assumed | 13.6 months (10.5 to 18.1, 80% spread across simulated trials) Assumed | 25.6 months (22.5 to 30.1, 80% spread across simulated trials) Assumed | 80.0% (73.3% to 85.5%, 80% spread across simulated trials) Assumed | 175 (130 to 234, 80% spread across simulated trials) Assumed |
| Design B under the local profile | 524 (370 to 750, 90% sensitivity range) Assumed | 87.4% (85.2% to 89.3%, 95% Monte Carlo) Assumed | 17.4 months (13.6 to 22.7, 80% spread across simulated trials) Assumed | 29.4 months (25.6 to 34.7, 80% spread across simulated trials) Assumed | 79.8% (73.7% to 84.9%, 80% spread across simulated trials) Assumed | 279 (206 to 372, 80% spread across simulated trials) Assumed |
| Recommended design, effect at the weak end | 524 (370 to 750, 90% sensitivity range) Assumed | 53.2% (50.1% to 56.3%, 95% Monte Carlo) Assumed | 17.4 months (13.6 to 22.7, 80% spread across simulated trials) Assumed | 29.4 months (25.6 to 34.7, 80% spread across simulated trials) Assumed | 79.8% (73.7% to 84.9%, 80% spread across simulated trials) Assumed | 279 (206 to 372, 80% spread across simulated trials) Assumed |
Inputs that fell back to a global default
The local profile has no value for these, so the simulator used the global default. Uncited defaults are flagged.
| Input | Source | Citation |
|---|---|---|
| site_activation_days | cited | 31.4 weeks to site activation, Tufts CSDD site initiation benchmark, 2018 (PRD Appendix A). The range is assumed. |
Enrolment forecast for design B, by site type
Median cumulative enrolment: total (ink), central (green), provincial (amber, dashed). Grey band: 80% spread of the total. Dotted line: target. Weakest input label assumed.
Code check: re-running the ingested study's own design
F3.4. The observed dropout curve should sit inside the band simulated from the study's own estimates. This checks the code, not the model.
| Month | Observed survival | 90% simulated band | |
|---|---|---|---|
| 0 | 1.000 | 0.997 to 1.000 | inside |
| 2 | 0.960 | 0.947 to 0.980 | inside |
| 4 | 0.923 | 0.903 to 0.953 | inside |
| 7 | 0.870 | 0.847 to 0.910 | inside |
| 12 | 0.802 | 0.766 to 0.840 | inside |
Run records
Every run reproduces exactly from its seed, profile version and code version.
| Run | Inputs | Design | Trials | Seed | Profile | Code |
|---|---|---|---|---|---|---|
| run-e50f82e6f538 | global defaults | A, sized on global defaults | 1000 | 7 | global-defaults | 0.1.0+g089e218.src.52797a48 |
| run-3d0da899393e | local profile v1 | A, sized on global defaults | 1000 | 7 | v1 | 0.1.0+g089e218.src.52797a48 |
| run-c2c8f49fe438 | local profile v1 | B, sized on the local profile | 1000 | 7 | v1 | 0.1.0+g089e218.src.52797a48 |
| run-f8b301d10a6a | local profile v1 | B, sized on the local profile | 10000 | 7 | v1 | 0.1.0+g089e218.src.52797a48 |
| run-a31d7f0e67eb | local profile v1 | B, sized on the local profile | 1000 | 7 | v1 | 0.1.0+g089e218.src.52797a48 |