Table 1

Simulated scenarios for sample size determination in the design of a hypothetical study

Standard Scenario

Relaxed significance level

Relaxed Power

Baseline Mortality shifted away from 50%

Inflated delta

Significance level (two-sided)

0.05

0.1

0.05

0.05

0.05

Power

90%

90%

80%

90%

90%

Baseline (placebo) mortality rate

50%

50%

50%

40%

50%

Delta (ARR)

10%

10%

10%

10%

15%

Required sample size

1076

884

816

992

480

Inflation of delta has a substantially larger impact on required sample size than changes in the other variables. ARR, absolute risk reduction.

Aberegg et al. Critical Care 2010 14:R77   doi:10.1186/cc8990