Statistical analysis
We
described maternal characteristics of women who were dispensed at least
one of the selected medications during the 6-weeks diary period and
women who did not use medication according to the pharmacy records. For
each medication registered in the pharmacy records with theoretical use
in the diary period, we determined actual use with Wilson score 95%
confidence intervals (CIs), initiation, and implementation adherence for
the three predefined medication categories, medication groups, and
individual medications with ≥5 prescription dispensings. For
implementation adherence, we calculated the mean proportion of days with
at least the correct number of doses taken as well as the proportion of
users with an implementation adherence ≥80%. To evaluate whether the
maternal characteristics were associated with implementation adherence,
we used univariable linear regression with robust standard errors to
account for multiple prescriptions per participant in an exploratory
analysis. All statistical analyses were performed using Stata version
16.0 (StataCorp LLC, College Station, TX).