2.3-Power calculation
Several randomized, controlled studies comparing HFNCO and LFO have
assessed changes in HR, RR, and CRS over time in children with
bronchiolitis. However, we could not find a study comparing the time
that HR and RR returned to the normal range for age and, similarly, the
time for our CRS outcome mentioned. Therefore, we made a preliminary
analysis and sample calculation by evaluating the data of 24 patients
who met the study criteria. Sample size calculations were done using a
t-test on primary outcomes to obtain a power of 90% and an alpha of
.05. The sample size for HR and RR yielded 39 patients in each group,
while the sample size for CRS yielded 50 subjects. Unfortunately, we did
not reach the number of participants calculated with the power analysis
because of the financial problem. So instead, we figured the effect
sizes r of primary outcomes with the Mann-Whitney-U test interpreted as
0.1 small, 0.3 medium, and 0.5 large, according to the r = z/√n
formula.22,23 The effect size for each of the primary
outcomes was presented below the table.