Evaluation by simulation of clinical trial designs for evaluation of treatment during a viral haemorrhagic fever outbreak

Table 1 Proportion of trials showing significant improvement with the experimental treatment (“standard” and “changing with time” cases)

	Design	p_C = 0.50				p_C = 0.35		p_C = 0.75
	Design	∆ = 0.20	∆ = 0.10	∆ = 0	∆ = − 0.10	∆ = 0.20	∆ = 0	∆ = 0	∆ = − 0.10
Standard	F1	0.896	0.351	0.025	0.0002	0.120	0.000	0.985	0.662
	S1	0.893	0.339	0.024	0.0003	0.115	0.0001	0.982	0.657
	F2	0.897	0.347	0.024	0.0001	0.887	0.025	0.025	0.0001
	S2	0.904	0.339	0.025	0.0004	0.883	0.025	0.024	0.0002
Changing with time	F1	0.964	0.536	0.069	0.0011	0.241	0.0001	0.997	0.823
	S1 (^a)	0.944 (0)	0.520 (0)	0.062 (0)	0.0008 (0)	0.229 (0)	0 (0)	0.991 (0)	0.796 (0)
	F2	0.926	0.370	0.025	0.0002	0.875	0.023	0.025	0.0001
	S2 (^a)	0.958 (0.0007)	0.390 (0.0001)	0.024 (0)	0.0004 (0)	0.880 (0)	0.025 (0)	0.015 (0.11)	0 (0)

In bold, type-I-errors maintained in [0.024;0.026]
Abbreviations: p_C indicates control survival rate; ∆: simulated survival rate difference; F1: fixed single-arm design; S1: group-sequential single-arm design; F2: fixed two-arm design; S2: group-sequential two-arm design; Standard: specific case with a fixed control survival rate; Changing with time: specific case with an increase of the control survival rate over time
(^a): Proportion of significant tests with proportion of inconclusive group-sequential trials for “changing with time” case

ISSN: 1471-2288