The use of facility-specific clinical practice guidelines and clinical decision-making tools are recommended as part of an antimicrobial stewardship program (ASP). Reference Barlam, Cosgrove and Abbo1 Clinical decision-making tools are diverse and can include complex computer algorithms, checklists, and electronic order sets. Reference Cresswell, Mozaffar, Shah and Sheikh2 They have been shown to improve adherence to evidence-based antibiotic prescribing, reducing unnecessary antibiotic use Reference Cresswell, Mozaffar, Shah and Sheikh2-Reference Forrest, van Schooneveld and Kullar5 and improving appropriateness of therapy. Reference Hecker, Fox and Son3 Additionally, the use of order sets has been associated with reduced length of hospital stay. Reference Brown, Johnson and DeRonne4 An environmental scan of our institution’s numerous electronic order sets (EOSs) showed very poor use by clinicians of <10 times annually.
The primary objective of this study was to evaluate the impact of simplifying certain ASP EOSs with physician feedback and embedding these revised EOSs into the General Medicine Admission EOS, which is frequently used by our institution’s internists.
Methods
This study was conducted at a 400-bed community hospital in Toronto, Canada. An ASP was established at our institution in 2011, with a multimodal approach including prospective audit and feedback, development of guidelines and order sets, microbiology laboratory report optimization, and education. The study team consisted of a lead ASP physician and 2 full-time equivalent ASP-trained pharmacists during the study period. This study, including both pre- and postintervention periods, was conducted from January 1, 2016, to June 30, 2018.
The existing standalone ASP electronic order sets for community-acquired pneumonia (CAP), urinary tract infection (UTI), and skin and soft-tissue infection (SSTI) were reviewed prior to embedding by the ASP pharmacist, the ASP physician, a clinical informatics specialist, and the chief medical information officer, who is a practicing general internist. To reduce duplication with the general medicine admission EOS, only the antibiotic section was included from the standalone ASP EOS and other sections were excluded (ie, orders for cultures, imaging, and laboratory markers). The antibiotic section was optimized to align with local guidelines, then (1) antibiotic choices were rearranged to prioritize oral options, (2) minimum duration of therapy was used as the preset duration, and (3) high-risk antibiotics associated with Clostridioides difficile infection (CDI) were removed. For instance, the 7-day option for CAP, 14-day option for complicated pyelonephritis, clindamycin in the SSTI EOS and ciprofloxacin for cystitis were removed (later replaced by amoxicillin-clavulanate). General internists were informed of the embedded ASP EOS prior to its implementation on March 30, 2017.
We used a before-and-after study design; the preintervention period was January 1, 2016, to March 29, 2017, and the postintervention period was March 30, 2017, to June 30, 2018. The primary outcome was change in use of embedded ASP EOSs compared to the corresponding standalone ASP EOS preintervention using percentages. The use data for standalone ASP EOSs for the 3 conditions were collected, in addition to data for nonembedded ASP EOS (eg, CDI, hospital-acquired pneumonia) which were used as controls. This study was approved by the research ethics board at Unity Health Toronto on July 16, 2019.
Results
We observed a large increase in the use of embedded EOSs compared to the corresponding standalone EOS. The standalone control EOS use remained similar before and after the intervention, except for a 16-fold increase in CDI EOS use. An 11-fold (355 vs 32) increase in CAP EOS use occurred, as well as a 47-fold (94 vs 2) increase in UTI EOS use and a 24-fold (24 vs 1) increase in SSTI EOS use (Fig. 1).
Fig. 1. Electronic order set usage pre- and post-intervention.
Discussion
Although ‘nudging’ Reference Thaler and Sunstein6 has been previously applied to ASP principles, literature on successful and sustainable nudging interventions remains scarce. We embedded ASP EOS into a more frequently used EOS, which led to a large increase in EOS use and a decrease in fluoroquinolone use. We hypothesized that the stepwise simplification and prioritization of ASP-recommended antibiotic options with the embedded ASP EOS changed the prescribing habits of our general internists to some extent. Additionally, the inclusion of some clinician feedback in designing the embedded EOS likely contributed to ease of appropriate empiric antibiotic prescribing using these.
However, our study has several limitations. Given its retrospective before-and-after study design, several confounding factors likely contributed to changes in EOS use and antibiotic use over time. We tried to account for these by tracking changes in the corresponding standalone ASP EOS as well as nonembedded ASP EOS. We found minimal change in standalone ASP EOS use, which supported our hypothesis that embedding EOS increased its uptake. Other limitations of our study include not being able to assess the percentage use of embedded ASP EOSs in relation to the number of patients diagnosed with CAP, UTI, and cellulitis as well as not undertaking an audit on appropriateness of diagnosis and antibiotic selection for patients with embedded ASP EOS orders. There is a no specific discharge code for CAP nor codes for the UTI and cellulitis cases of interest, and we were unable to address these study limitations. Future studies on a similar intervention should consider the inclusion of these outcomes.
To fully optimize ASPs and to achieve a long-term sustainable impact on patient outcomes, information technology must be employed. The approach of embedding ASP EOSs into a more frequently used EOS has the potential to improve antibiotic prescribing using existing resources with minimal cost.
Acknowledgments
None.
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
