Baseline characteristics of patients and catheters
were described by group. The primary analysis was by
intention to treat, including all patients (and all
catheters) in their ran domised group. First, we
calculated relative incidence rates of phlebitis and
absolute rate diff erences per 100 catheters and per
100 patients, and we then used two-sided Fisher’s exact
test to assess equivalence in risk between groups.
Second, we calculated hazard rates per 1000 catheter
hours, with 95% CI, with a Cox proportional hazards
model (assumptions were checked) to sum marise the
eff ect of intervention per patient (includ ing all catheters
per patient). Finally, we used Kaplan-Meier survival
curves to compare rates of time until fi rst phlebitis per
patient between groups.
Baseline characteristics of patients and catheterswere described by group. The primary analysis was byintention to treat, including all patients (and allcatheters) in their ran domised group. First, wecalculated relative incidence rates of phlebitis andabsolute rate diff erences per 100 catheters and per100 patients, and we then used two-sided Fisher’s exacttest to assess equivalence in risk between groups.Second, we calculated hazard rates per 1000 catheterhours, with 95% CI, with a Cox proportional hazardsmodel (assumptions were checked) to sum marise theeff ect of intervention per patient (includ ing all cathetersper patient). Finally, we used Kaplan-Meier survivalcurves to compare rates of time until fi rst phlebitis perpatient between groups.
การแปล กรุณารอสักครู่..
