Group-sequential analysis may allow for early trial termination: illustration by an intra-observer repeatability study

EJNMMI Research

Table 2 Sequential testing on population standard deviation σ

Number of interim analyses	Observed standard deviation, P value (sample size), respective significance level, and upper confidence limit at analysis time point
Number of interim analyses	Interim analysis 1	Interim analysis 2	Interim analysis 3	Final analysis
0	–	–	–	SD = 1.060 P = 0.0022 (N = 45) α ₁ = 0.05 95% uCL 1.288
1	SD = 1.415 P = 0.3900 (N = 23) α_0.5 = 0.025 97.5% uCL 2.002	–	–	SD = 1.060 P = 0.0022 (N = 45) α ₁ = 0.05 95% uCL 1.288
2	SD = 0.909 P = 0.0162 (N = 15) α _0.33 = 0.0167 98.33% uCL 1.495	SD = 1.255 P = 0.1162 (N = 30) α_0.67 = 0.0333 96.67% uCL 1.653	–	SD = 1.060 P = 0.0022 (N = 45) α ₁ = 0.05 95% uCL 1.288
3	SD = 1.044 P = 0.0984 (N = 11) α_0.25 = 0.0125 98.75% uCL 2.006	SD = 1.415 P = 0.3900 (N = 23) α_0.5 = 0.025 97.5% uCL 2.002	SD = 1.185 P = 0.0452 (N = 34) α_0.75 = 0.0375 96.25% uCL 1.519	SD = 1.060 P = 0.0022 (N = 45) α ₁ = 0.05 95% uCL 1.288

SD standard deviation, uCL upper confidence limit, italics rejection of null hypothesis