Table II.

Approximate statistical power (2-sided α = 0.01) among 100,000 participants randomized between statin and placebo to detect relative risks of 1.05, 1.1, 1.2, or 1.3 with absolute 5-year control rates of 2%, 5%, 10%, or 20%

5-y event rate in control group	Hypothetical relative risk associated with allocation to statin therapy
1.05	1.1	1.2	1.3
Power at 2-sided α = 0.01
2%	7%	36%	96%	>99%
5%	22%	83%	>99%	>99%
10%	51%	>99%	>99%	>99%
20%	91%	>99%	>99%	>99%

The above power estimates are based on a test of the observed odds ratio under each scenario, which will be slightly larger than the relative risks but will correspond exactly to the relative risks given the control event rate in each case. Odds ratios and relative risks are similar when outcomes are rare but become more different as outcome rates increase (eg, when the 5-year control event rate is 20%, the relative risks of 1.05, 1.1, 1.2, and 1.3 correspond to odd ratios of 1.06, 1.13, 1.26, and 1.41, respectively).