Table 3 Percentage of EPARs where the methods are applicable
From: Applicability and added value of novel methods to improve drug development in rare diseases
METHOD | Applicability in percentage of EPARs | |||
|---|---|---|---|---|
Step | Static step 1 (no adjustments) | Dynamic step 2 (adjustments) | ||
Statistic | Percentage of EPARs (n/26) | Percentage of clusters (n/6) | Percentage of EPARs (n/26) | Percentage of clusters (n/6) |
Extrapolation | 35% [9/26] | 83% | 46% [12/26] | 100% |
Heterogeneity estimators | 4% [1/26] | 17% | 4% [1/26] | 17% |
Prior distributions for variance parameters in sparse-event meta-analysis | 4% [1/26] | 17% | 4% [1/26] | 17% |
Delayed-start randomisation | 13% [3/26] | 50% | 12% [3/26] | 50% |
Sample size reassessment and hypothesis testing in adaptive survival trials | 35% [9/26] | 83% | 58% [15/26] | 100% |
Multi-arm group sequential designs with a simultaneous stopping rule | 23% [6/26] | 67% | 58% [15/26] | 100% |
Sequential designs for small samples | 31% [8/26] | 67% | 66% [17/26] | 100% |
Bayesian sample size re-estimation using power priors | 12% [3/26] | 33% | 50% [13/26] | 100% |
Dynamic borrowing through empirical power priors that control type I error | 15% [4/26] | 33% | 50% [13/26] | 100% |
Fallback tests for co-primary endpoints | 15% [4/26] | 50% | 50% [13/26] | 100% |
Optimal exact tests for multiple binary endpoints | 4% [1/26] | 17% | 31% [8/26] | 83% |
Simultaneous inference for multiple marginal GEE models | 19% [5/26] | 50% | 23% [6/26] | 67% |
Goal Attainment Scaling | 31% [8/26] | 67% | 31% [8/26] | 67% |