Sample size and statistical analysis
The target sample size for this trial was 40 (20 per group), based on
the weekly number of women attending who could be approached (15) and an
estimated 60% participation rate. There were no prior estimates of
standard deviations available for power estimation. All data was entered
into a secure database and anonymised using participant codes at the
point of data entry.
Statistical analysis was by intention-to-treat including all randomised
participants, using R software Version 3.5.1 (Feather Spray). Continuous
data were summarised as mean and standard deviation, and categorical
data as counts and percentages. Between-group differences were reported
with 95% confidence intervals (CI) and p-value (using t test to compare
normally distributed data). Cohen’s d, difference in scores measured on
a standard deviation scale, was used to determine effect size with
values above 0.7 considered to be large33. Linear
regression was used to estimate difference in continuous outcomes
between groups post-procedure, adjusting for stratification factors (of
menopausal status and parity) and baseline score.