Sample size calculation
Post-hoc sample size estimation: Using the primary outcome measure of GA
at delivery, a sample size of 42 achieves 99.30% power to detect a mean
of paired difference of 3.5 with a significance level (alpha) of 0.05,
using a two-sided paired Wilcoxon signed-rank test. Using the secondary
outcome measure of sPTB at <32 weeks of gestation, in a paired
experimental design, this sample size allowed the detection of a
significant difference of 33.3% in the rate of PTB at <32
weeks. At the error level of α-value of 0.05, a total sample size of 84
women achieved a power of 93%.