Bridging the evidence gap for implementing antibiotic stewardship in Norway: Interventions, process measures and patient outcomes related to antibiotic prescribing in hospitals

Abstract

Background: Antimicrobial resistance is a major challenge for patient safety worldwide, as a growing number of infections become difficult to treat and the advances made in modern medicine are threatened. Overuse and misuse of antibiotics accelerates the development of resistance. Optimizing treatment for the patients of today, while conserving effective antibiotics for future patients is therefore of great importance. Antimicrobial stewardship is a strategy and coherent set of actions which promote responsible use of antibiotics. Objectives: The main aim of this project was to contribute to the knowledge needed to implement antibiotic stewardship in Norwegian hospitals through a) determining the impact of audit with feedback and academic detailing with local target setting on antibiotic prescribing practice b) understanding how patient outcomes are associated with adherence to clinical guidelins on initiation of antibiotic treatment and c) determining whether targets for antimicrobial stewardship interventions can be identified through analysing the antibiotic prescribing process in Norwegian hospitals with patient-level data. Materials and methods: All three substudies were part of a combined multicentre study, performed within the specialties of pulmonary medicine, infectious diseases and gastroenterology at three hospitals in Western Norway. Study 1 included 1802 patients and was a randomised, controlled intervention study, assessing the impact of academic detailing, audit with feedback and local target setting on adherence to antibiotic guidelines and changes in locally defined targets. Study 2 and 3 were observational cohort studies, including 1756 patients and 1235 patients, respectively. Results: In study 1 there was an absolute increase in adherence to guidelines of 6% across all intervention wards (p=0.04). When analysed per specialty, pulmonary intervention wards had a 14% absolute increase in adherence (p=0.003), while other intervention wards had no observed impact of interventions on adherence. Intervention wards receiving audit with feedback decreased the use of broad-spectrum antibiotics (level and trend). Local target setting at one of the pulmonary wards led to a 30% increase in targeted prescribing behaviour (p<0.001). In study 2, guideline-adherent prescribing was associated with lower in-hospital (OR=0.46, p=0.003) and 30-day mortality (OR=0.48, p=0.001). There was also a trend towards shorter length of stay (-0.47 days) when guidelines were followed. Analysing the process of antibiotic prescribing in hospitals (study 3) identified 5 main targets for antimicrobial stewardship interventions: a) adherence to guidelines, b) prescribing in the emergency room, c) prescribing for patients admitted from other institution, d) understanding cultural and contextual drivers of antibiotic prescribing and e) duration of treatment. Conclusions: The impact of antimicrobial stewardship interventions on prescribing practice was dependent both on the context (e.g. specialty) in which interventions were implemented and how they were implemented. Pulmonary intervention wards increased adherence to guidelines with both audit with feedback and academic detailing, and additional impact was seen when locally defined targets were identified. We have shown that adherence to Norwegian antibiotic guidelines was associated with favourable patient outcomes across a range of common infectious diseases, both in terms of in-hospital- and 30-day mortality. Targets for antimicrobial stewardship interventions in hospitals were identified through analysis of patient-level, antibiotic prescribing data from admission to discharge and WHO AWaRe categories provided a useful system for analysing antibiotic regimens throughout the hospital stay. Although the studies were performed in Norwegian hospitals, we believe that the methods and findings will be applicable in other clinical settings where antibiotic prescribing and related patient outcomes is to be analysed and improved.