Efficacy and Cost-effectiveness of Hospital Antimicrobial Stewardship Programs

Context: There is a steady increase of the incidence of multidrug resistant bacteria, including some which have become extremely resistant, causing infections that are more and more difficult to treat with antibiotics thus dramatically impairing patient prognosis.

Several guidelines prompt hospitals to set up antimicrobial stewardship programs, with an antibiotic referrer or a dedicated team as a key component. The aim of the study is to encourage a better use of antibiotics, and to contain the emergence of resistance.

However, such programs, when they exist, cover multiple organisations and interventions to improve antimicrobial prescribing, and involve actors from different specialties. Furthermore, there is little information on how antimicrobial stewardship programs are actually implemented in hospitals and the impact of these programs has been difficult to measure.

Many studies, mainly conducted in England and in the United States, have evaluated interventions to control antibiotic prescription in hospitals. Two meta-analyses conclude that they are efficient but the variety of the interventions and the heterogeneity of the obtained results prevent any conclusions on which practices and strategies should be prioritized when conducting antimicrobial control programs.

The studies conducted so far have been mostly monocentric and have evaluated the impact of the interventions on a relatively short period of time. Finally, they rarely considered the costs and the cost-effectiveness of these programs.

More qualitative approaches have highlighted contextual and cultural factors that influence antibiotic prescribing habits and encourage deliverers to consider these factors when designing an intervention to promote appropriate antibiotic use. These studies have also emphasized that prescribers' adherence to guidelines and interventions is a fundamental factor in the success of any strategy. These aspects have not been explored much in the French national context.

Objective: The main goal is to evaluate the efficacy and the cost-effectiveness of strategies to promote appropriate use of antibiotics in hospitals and to identify the most efficient ones, on a short-term but also long-term scale.

The secondary objectives are 1) to map the different antimicrobial stewardship setups in French hospitals and determine the strategies to be evaluated by the model; 2) to measure prescribers' adherence to the different strategies used to control antibiotic prescription, and 3) to identify potential barriers and facilitators to the implementation of effective stewardship programs.

Methods: The project team will use modelling methods and techniques which combine a decision tree and a Markov model to estimate the impact of the different strategies, while considering the associated costs. To define the different strategies (restrictive, educational or structural strategies) and to measure clinicians' adherence to the strategy implemented in their setting, the project team will conduct a survey in a sample of 30 French hospitals (acute care). During this section, the project team will also collect a certain amount of quantitative data (antibiotic consumption per molecule, mean treatment length for selected infections…). These data will help us refine the model's parameters. To supply the model, other resources will be used : data from the relevant literature, computerized system of medical databases Outcome measurements: Main objective: A comparison of the different strategies based on the incremental cost-effectiveness ratio (ICER), which is defined by the difference in costs between interventions, divided by the difference in their measured impact; secondary objectives: accrue high-value information in the field of antibiotic stewardship, namely physicians' adherence to antibiotic counselling and interventions to control its use, as well as potential barriers and facilitators for the effective implementation of such programs.

Study population: Our study is based on modelling patients' hospitalisations and hospitalisation returns on a hypothetical cohort of patients, over a 10 year period.