Outcomes of multisite antimicrobial stewardship programme implementation with a shared clinical decision support system.

Background: Studies evaluating antimicrobial stewardship programmes (ASPs) supported by computerized clinical decision support systems (CDSSs) have predominantly been conducted in single site metropolitan hospitals. Objectives: To examine outcomes of multisite ASP implementation supported by a centrally deployed CDSS. Methods: An interrupted time series study was conducted across five hospitals in New South Wales, Australia, from 2010 to 2014. Outcomes analysed were: effect of the intervention on targeted antimicrobial use, antimicrobial costs and healthcare-associated Clostridium difficile infection (HCA-CDI) rates. Infection-related length of stay (LOS) and standardized mortality ratios (SMRs) were also assessed. Results: Post-intervention, antimicrobials targeted for increased use rose from 223 to 293 defined daily doses (DDDs)/1000 occupied bed days (OBDs)/month (+32%, P < 0.01). Conversely, antimicrobials targeted for decreased use fell from 254 to 196 DDDs/1000 OBDs/month (-23%; P < 0.01). These effects diminished over time. Antimicrobial costs decreased initially (-AUD$64551/month; P < 0.01), then increased (+AUD$7273/month; P < 0.01). HCA-CDI rates decreased post-intervention (-0.2 cases/10 000 OBDs/month; P < 0.01). Proportional LOS reductions for key infections (respiratory from 4.8 to 4.3 days, P < 0.01; septicaemia 6.8 to 6.1 days, P < 0.01) were similar to background LOS reductions (2.1 to 1.9 days). Similarly, infection-related SMRs (observed/expected deaths) decreased (respiratory from 1.1 to 0.75; septicaemia 1.25 to 0.8; background rate 1.19 to 0.90. Conclusions: Implementation of a collaborative multisite ASP supported by a centrally deployed CDSS was associated with changes in targeted antimicrobial use, decreased antimicrobial costs, decreased HCA-CDI rates, and no observable increase in LOS or mortality. Ongoing targeted interventions are suggested to promote sustainability.

Design and Implementation of a Novel Web-Based E-Learning Tool for Education of Health Professionals on the Antibiotic Vancomycin.

BACKGROUND: Traditional approaches to health professional education are being challenged by increased clinical demands and decreased available time. Web-based e-learning tools offer a convenient and effective method of delivering education, particularly across multiple health care facilities. The effectiveness of this model for health professional education needs to be explored in context. OBJECTIVES: The study aimed to (1) determine health professionals' experience and knowledge of clinical use of vancomycin, an antibiotic used for treatment of serious infections caused by methicillin-resistant Staphylococcus aureus (MRSA) and (2) describe the design and implementation of a Web-based e-learning tool created to improve knowledge in this area. METHODS: We conducted a study on the design and implementation of a video-enhanced, Web-based e-learning tool between April 2014 and January 2016. A Web-based survey was developed to determine prior experience and knowledge of vancomycin use among nurses, doctors, and pharmacists. The Vancomycin Interactive (VI) involved a series of video clips interspersed with question and answer scenarios, where a correct response allowed for progression. Dramatic tension and humor were used as tools to engage users. Health professionals' knowledge of clinical vancomycin use was obtained from website data; qualitative participant feedback was also collected. RESULTS: From the 577 knowledge survey responses, pharmacists (n=70) answered the greatest number of questions correctly (median score 4/5), followed by doctors (n=271; 3/5) and nurses (n=236; 2/5; P<.001). Survey questions on target trough concentration (75.0%, 433/577) and rate of administration (64.9%, 375/577) were answered most correctly, followed by timing of first level (49%, 283/577), maintenance dose (41.9%, 242/577), and loading dose (38.0%, 219/577). Self-reported "very" and "reasonably" experienced health professionals were also more likely to achieve correct responses. The VI was completed by 163 participants during the study period. The rate of correctly answered VI questions on first attempt was 65% for nurses (n=63), 68% for doctors (n=86), and 82% for pharmacists (n=14; P<.001), reflecting a similar pattern to the knowledge survey. Knowledge gaps were identified for loading dose (39.2% correct on first attempt; 64/163), timing of first trough level (50.3%, 82/163), and subsequent trough levels (47.9%, 78/163). Of the 163 participants, we received qualitative user feedback from 51 participants following completion of the VI. Feedback was predominantly positive with themes of "entertaining," "engaging," and "fun" identified; however, there were some technical issues identified relating to accessibility from different operating systems and browsers. CONCLUSIONS: A novel Web-based e-learning tool was successfully developed combining game design principles and humor to improve user engagement. Knowledge gaps were identified that allowed for targeting of future education strategies. The VI provides an innovative model for delivering Web-based education to busy health professionals in different locations.

Evaluating the Effect of a Web-Based E-Learning Tool for Health Professional Education on Clinical Vancomycin Use: Comparative Study.

BACKGROUND: Internet-based learning for health professional education is increasing. It offers advantages over traditional learning approaches, as it enables learning to be completed at a time convenient to the user and improves access where facilities are geographically disparate. We developed and implemented the Vancomycin Interactive (VI) e-learning tool to improve knowledge on the clinical use of the antibiotic vancomycin, which is commonly used for treatment of infections caused by methicillin-resistant Staphylococcus aureus (MRSA). OBJECTIVE: The aims of this study were to evaluate the effect of the VI e-learning tool on (1) survey knowledge scores and (2) clinical use of vancomycin among health professionals. METHODS: We conducted a comparative pre-post intervention study across the 14 hospitals of two health districts in New South Wales, Australia. A knowledge survey was completed by nurses, doctors, and pharmacists before and after release of a Web-based e-learning tool. Survey scores were compared with those obtained following traditional education in the form of an email intervention. Survey questions related to dosing, administration, and monitoring of vancomycin. Outcome measures were survey knowledge scores among the three health professional groups, vancomycin plasma trough levels, and vancomycin approvals recorded on a computerized clinical decision support system. RESULTS: Survey response rates were low at 26.87% (577/2147) preintervention and 8.24% (177/2147) postintervention. The VI was associated with an increase in knowledge scores (maximum score=5) among nurses (median 2, IQR 1-2 to median 2, IQR 1-3; P<.001), but not among other professional groups. The comparator email intervention was associated with an increase in knowledge scores among doctors (median 3, IQR 2-4 to median 4, IQR 2-4; P=.04). Participants who referred to Web-based resources while completing the e-learning tool achieved higher overall scores than those who did not (P<.001). The e-learning tool was not shown to be significantly more effective than the comparator email in the clinical use of vancomycin, as measured by plasma levels within the therapeutic range. CONCLUSIONS: The e-learning tool was associated with improved knowledge scores among nurses, whereas the comparator email was associated with improved scores among doctors. This implies that different strategies may be required for optimizing the effectiveness of education among different health professional groups. Low survey response rates limited conclusions regarding the tool's effectiveness. Improvements to design and evaluation methodology may increase the likelihood of a demonstrable effect from e-learning tools in the future.