Other clinical trials also investigated the effect of
iv PPI, bolus and continuous infusion, versus placebo [6,7]. However, we opted to rely on one study
only because all patients underwent endoscopic
therapy and the authors also reported the relative
risk of rebleeding following a repeat endoscopic
treatment [5]. To examine the impact of using the
benefit estimated from meta-analysis of all studies,
we conducted a sensitivity analysis by reestimating
the probabilistic model using the pooled measures
of relative risks for a first rebleed and for undergoing a surgical procedure.
To account for a potentially longer delay before
investigative endoscopy we also recalculated the
incremental cost of iv PPI by extending the waiting
time from 24 to 48 hours. Thus, instead of only
receiving iv PPI for the first 24 hours, subjects for
whom therapy would eventually be discontinued,
i.e., no bleeding peptic ulcers, were now assumed to
stay on therapy for 48 hours.