Analytical approach:
A decision tree was constructed to combine data from a systematic review and costs from a US national database. The time horizon was 60 days, and the authors stated that they took a third-party payer perspective.
Effectiveness data:
A systematic review was undertaken to identify the relevant evidence. MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), and Web of Knowledge were searched, along with clinical trial databases and abstracts from Digestive Disease Week and United European Gastroenterology Week. The main clinical effectiveness estimates were the rates of bleeding from stress ulcer and ventilator-associated pneumonia. These rates were estimated from a published meta-analysis, by some of the authors of this study (see Other Publications of Related Interest).
Monetary benefit and utility valuations:
Not relevant.
Measure of benefit:
The benefit measure was the probability of no complication (bleeding or pneumonia).
Cost data:
The hospital costs were included, and these were assumed to include the drugs. The length of stay and per day costs were from the National Inpatient Sample for 2008. They were estimated by combining the average charges with a cost-to-charge ratio. They were presented in 2010 US $.
Analysis of uncertainty:
One-way sensitivity analyses were undertaken, by varying each parameter within its 95% confidence interval. The results were presented in a tornado diagram. A two-way sensitivity analysis was undertaken to assess the pneumonia incidence for the two treatment pathways, and threshold analysis was undertaken on the probabilities of ventilator-associated pneumonia and bleeding from stress ulcer, and the costs. Probabilistic sensitivity analysis was undertaken, using 10,000 Monte Carlo simulations; the results were presented in a scatterplot and cost-acceptability curves.