[Elements for a cost/utility analysis of long term intravenous devices]

Three long-term intravenous infusion strategies were compared, namely an implantable port (IP), a tunneled central catheter (TCC), and repeated peripheral catheterization (RPC). A decision analysis model was used in which the sequence of decisions and their possible consequences was described as a decision tree for each of the three strategies. The likelihood of each event occurring was determined based on a literature review. Each event was assigned a cost and a utility. Direct medical costs for the society include the cost of the material, the cost of implantation and removal of the device in the operating room, and the cost of treating complications directly ascribable to the strategy used. Utility is a combination of efficacy (as evaluated based on the likelihood of use of the strategy for one year) and quality of life (as evaluated by experts). In the basic analysis, RPC was the most cost-saving method but had an adverse effect on quality of life (0.82). The cost of using IC or TCC for one year was higher, with the main contributors to the excess cost being the insertion procedure and the management of complications (primarily deep vein thrombosis and infection). Quality of life was better with IC (0.98) than with TCC (0.93). IC also had a higher cost-utility ratio (11,738 French francs [FF]) versus 17,393 FF). A one-way sensitivity comparison of IC and TCC showed that the only realistic change capable of reversing the order between these two methods was a decrease by one-third in the risk of infection with TCCs. This model, used here for the first time, establishes that IC is superior over TCC.