Implementing best practice for peripheral intravenous catheter use in Australian Emergency Departments: health economic analysis of a stepped-wedge cluster trial

This purpose of this project is to conduct a health economic analysis of an implementation science trial. The trial aims to reduce unnecessary peripheral intravenous catheter (PIVC) insertion and improve use of the devices on patients in emergency departments (EDs) in nine participating hospitals in Australia. A ‘within trial’ cost-benefit analysis will be conducted to calculate the net present value of the intervention. This will be done from each hospital's perspective that is enrolled in the funded, stepped-wedge cluster trial, and from a public hospital funder’s perspective. The cost of delivering the intervention, cost of PIVC insertion and background health service use will be considered for the intervention and control groups. The cost of delivering the intervention will be determined from trial data. Healthcare utilisation per participant will incorporate the cost of PIVC insertion (staff time, and equipment), as well as background health service utilisation of patients’ PIVCs during their ‘hospital episode’ (defined as the time from presentation to the ED through until discharge away from the ED). The background health service utilisation data will be determined through data linkage to routinely collected ED and admitted patient data collection records, and health service costing unit records for all patients at each participating site during the intervention and control periods. The costing unit records will be used to assign a cost to each patient’s episode of care, and the Independent Hospital Pricing Authority (IHPA) pricing reports will be used to assign a cost to public hospital funders for each episode of care. Generalised linear models will be used to compare the cost difference between the two groups (intervention and control), adjusting for any unbalanced confounders in patient characteristics. Bootstrapping will be used to estimate the distribution around costs. One-way and multi-way sensitivity analyses will be conducted for key variables. The net present value of the intervention will be the benefits of its provision (difference in cost between the intervention and control groups), minus the cost of providing it. Broader community and carbon costs will also be explored which are expected to be considerable.