Evaluating the association of TRPA1 gene polymorphisms with pain sensitivity: a protocol for an adaptive recall by genotype study

Table 2 Interim analyses

Interim analysis #	Futility effect size	Efficacy cut-off alpha	Effect size req. for statistical eff (ΔHPT)
1	0.57	0.0013	1.63 (Δ4.1 °C)
2	0.62	0.0023	1.52 (Δ3.8 °C)
3	0.68	0.0055	1.42 (Δ3.5 °C)

Each row indicates the study state at subsequent interims. Effect sizes are displayed as Cohen’s d \(\frac{{\left( {\mu_{minor} - \mu_{major} } \right)}}{\sigma }\) where µ is the mean and σ is the standard deviation. The futility effect size cut-off is the minimum effect size required to continue the group in the study. The futility effect size is calculated as the smallest effect size observable at the final analysis at 80% probability given an alpha cut-off of 0.05. The efficacy cut-off alpha is the alpha criteria for the interim analysis to determine efficacy. The efficacy cut-off alpha is calculated using an O’Brien-Fleming alpha spending through the gsdesign R package. With subsequent interims the alpha threshold is relaxed as the proportion of information known at interim increases due to the fixed sample size of 15. The effect size required for statistical efficacy (HPT) indicates the minimum effect size (cohen’s d) likely to be observed at 80% power at interim with the corresponding changes in HPT (Δ) indicated assuming σ = 2.5. HPT, heat pain threshold

ISSN: 1755-8794