Discussion

Reducing inappropriate antimicrobial use can be viewed as a de-implementation challenge. Most implementation science models for de-implementation emphasize removal and replacement strategies to promote practice change. ^{Reference Norton and Chambers9} During our formative work, we found that ongoing antimicrobial use in the cardiac electrophysiology laboratory is a multi-level problem, driven not only by provider practice patterns but also by other clinical care providers and process elements that contribute to the sustainment of the guideline-discordant practice. ^{Reference Branch-Elliman, Gupta and Rani Elwy1} Thus, we sought to develop educational materials to address these barriers by giving providers other tools for promoting CIED infection prevention that would also engage patients in their own care. Our initial plan for this implementation strategy was to directly engage patients in antimicrobial stewardship activities, similar to work in de-prescribing of chronic medications demonstrating the effectiveness of patient-driven processes. ^{Reference Linsky, Kressin and Stolzmann3}

Based on traditional stewardship approaches, initial messaging in the materials focused on the harms of antibiotics. However, during the co-creation process, we found that negative messaging about the potential harms of inappropriate antibiotics and messaging related to patients asking providers about the necessity of post-procedural antibiotics were poorly received. The consistent feedback we received across key stakeholders about positive messaging (eg, what patients can do to prevent CIED infection) may suggest stronger support for replacement of processes (eg, focusing on effective interventions) compared to solely removal of processes (eg, focusing on harms of post-procedural antibiotics). ^{Reference Norton and Chambers9–Reference Helfrich, Rose and Hartmann11}

Reluctance to encourage questioning post-procedure antibiotic prescriptions was expressed by patients and stewardship providers and in the context of guideline-discordant use. ^{Reference Bratzler, Dellinger and Olsen8} This patient feedback was in contrast with other work that was found successful in de-prescribing using patient-targeted materials that included an emphasis on drug adverse effects. ^{Reference Linsky, Kressin and Stolzmann3} However, this successful study was completed in primary care and targeted long-term medications, which may be more conducive to shared decision-making discussions. While feedback in our study was elicited from key stakeholders (physicians and patients), study limitations include that this was from a small number of qualitative discussions rather than a standardized process, such as focus group or formal survey. However, participants in these discussions are the end-users of the products, and thus their buy-in and support are critical for adoption and dissemination of materials. The patients in this study were part of the Veterans Affairs (VA) hospital and perceptions may be different from non-VA populations.

In summary, positive messaging about the benefits of antibiotics and an emphasis on actionable infection prevention strategies were more acceptable to patients and providers than a focus on antibiotic harms in the creation of periprocedural patient-facing stewardship education materials, which can be widely used to expand antimicrobial stewardship practices and messages to clinical care settings with limited coverage, such as the electrophysiology laboratory (final materials available for download in the Supplementary Materials). Future work will focus on assessing the impact of the educational material in a de-implementation trial within real-world clinical settings. ^{Reference Branch-Elliman, Lamkin and Shin7}