The goal was to recruit 20 infants per group. With this
sample size and alpha set to .05, a large effect size of f = .43
can be detected with a power of 84%. This effect sizewould
be present if the change in cortisol in the Yakson group was
1.0 standard deviations greater than in the control group,
and if the change in cortisol in the GHT group was 0.8
standard deviations greater than in the control group. The
subjects consisted of 59 preterm infants who were admitted
to NICUs in two general hospitals in SouthKorea (20 for the
Yakson group, 20 for the GHT group, and 19 for the control
group). Inclusion criteria were (1) a gestational age of 26–
34 weeks at birth, (2) no congenital anomalies, (3) not
having undergone surgery, and (4) no medical condition
contraindicating the administration of Yakson or GHT such
as symptomatic sepsis. The preterm infants were assigned
to the three groups based on the order in which they were
born