Audit and feedback

The most widely used strategy for tele-stewardship programs has been audit and feedback of antibiotic prescribing. ^{Reference Pierce and Stevens13} This intervention consists of tracking the amount and appropriateness of antibiotic use at a clinic or provider level and offering feedback to the prescriber regarding alternative courses of action if indicated. ^{Reference Shively, Moffa and Paul17} Audit and feedback can take multiple forms, including prospective audit and feedback that intervenes directly on new antimicrobial courses as they are prescribed as well as retrospective approaches that intermittently provide prescribers with feedback on their overall prescribing practices. This strategy has been used extensively in the inpatient setting where it is usually performed either by a pharmacist or ID physician. ^{Reference Mailig, Cookson and Schulz12} Audit and feedback has also been used successfully in outpatient settings. In a study by Gerber et al, quarterly audit and feedback sessions were performed with providers at selected pediatric primary care offices in non-rural parts of Pennsylvania and New Jersey, resulting in significant decreases in antibiotic prescribing compared to control sites where no intervention was performed. ^{Reference Gerber, Prasad and Fiks14} A follow-up study by the same group showed that once the audit and feedback intervention was removed, antibiotic prescribing increased back to baseline levels. ^{Reference Gerber, Prasad and Fiks15} Audit and feedback interventions align with many of the core elements outlined by the CDC, as they involve action for practice, tracking/reporting prescribing, leveraging antimicrobial prescribing expertise, and educating prescribers. This practice is often the core of a tele-stewardship program, has been effective at reducing antibiotic use, and has been implemented in some studies as the sole intervention. ^{Reference May, Veillette and Webb18,Reference Wilson, Banks and Crnich19} The specific methodology of audit and feedback varies to some degree and ranges from giving direct feedback on patients’ ongoing antimicrobial therapy (more common in inpatient implementation) to more intermittent feedback on prescribing trends (more common for outpatient). Studies examining audit and feedback as a sole intervention were often more limited in scope; one study was specific to decreasing antimicrobial prescriptions for COVID-19 pneumonia, ^{Reference May, Veillette and Webb18} though it did show decreased overall antibiotic use in the specific context studied. Another study that effectively demonstrated the efficacy of audit and feedback was conducted by Stenehjem et al using three sets of stewardship interventions of increasing intensity in 15 small hospitals in the Intermountain healthcare system. The first set included basic stewardship education, an ID hotline, and an antimicrobial use report. The second set of interventions added advanced stewardship education, local antibiotic restriction, and limited prospective audit and feedback involving pharmacist review of a few specific antibiotics (vancomycin, piperacillin/tazobactam, cefepime, and carbapenems) for opportunities for de-escalation, bug/drug mismatches, and indications for contacting an ID clinician. The third set expanded these interventions to include ID clinician approval for restricted antibiotics, ID clinician review of designated cultures, and expanded the audit and feedback initiated in the second set of interventions to a broader variety of antibiotics ordered for inpatients. Only the third set of interventions including the most intensive stewardship interventions like expanded audit and feedback was associated with a significant reduction in aggregate antimicrobial use. ^{Reference Stenehjem, Hersh and Buckel20} These results highlight the importance of audit and feedback as a component of a successful stewardship program.

The advantages of audit and feedback are likely as applicable to the rural outpatient context as they are to the inpatient setting. However, there are some features of rural outpatient settings that make audit and feedback more difficult to implement. As outpatient visits are brief and further contact with patients after the initial visit does not always occur, any prospective audit and feedback on ongoing courses of antimicrobials as is often done in inpatient programs will be difficult. An additional challenge making prospective audit and feedback impractical is the higher volume of patients seen in the outpatient setting compared to inpatient. The sort of personalized feedback required for this intervention is resource-intensive and impractical to implement across an outpatient healthcare network. These disadvantages would likely be more pronounced in rural settings where there are fewer providers available leading to higher individual patient volumes and less access to support from larger health systems. Instead, studies focused on tele-stewardship in the outpatient context tend to rely on more asynchronous methods of stewardship, such as prescriber education or distribution of antibiotic use reports, that do not require real-time clinical decision support from stewardship personnel. ^{Reference Stenehjem, Wallin and Willis5}

Antibiotic use reports

Antibiotic use reports are a less labor-intensive version of the audit and feedback intervention. Such interventions summarize antibiotic use in a particular context over time and often benchmark antibiotic use against other sites or national averages. These reports can also be individualized to the provider level. Antibiotic use reports have been included as a component of a variety of tele-stewardship studies based in diverse settings including urgent cares, community hospitals, and long term care hospitals. ^{Reference Stenehjem, Hersh and Buckel20} These studies showed improvement in some measures of antimicrobial prescribing, but used bundled interventions, so the degree that these results were due solely to the antibiotic use reports is unclear. Antibiotic use reports align with the “tracking and reporting” core element of antibiotic stewardship. In addition, such reports could also contain clinical guidance on how to improve prescribing practices which would also comply with the core element of education and expertise. Antibiotic use reports have the advantage of directly addressing antimicrobial use rates without requiring the sort of proactive, direct intervention and time investment needed for audit and feedback. Antibiotic utilization reports may be less likely to lead to fewer antibiotic prescriptions if additional interventions are not in place. ^{Reference Stenehjem, Hersh and Buckel20} Antibiotic reports are a retrospective rather than prospective intervention and therefore are likely to require a longer period to show improvement in antibiotic prescribing practices. Creation of antibiotic use reports also are difficult without a common EHR and data tracking capability that may not be available to many rural settings. That said, such reports are well suited to the outpatient rural context as they can easily be disseminated electronically, less labor intensive than audit and feedback, and are scalable to larger patient populations.

Peer comparison

Another variation of the audit and feedback approach is peer comparison. In such an intervention, individual providers receive a report comparing their antimicrobial prescribing practices to their peers, showing how frequently they prescribed antibiotics overall, how often antibiotics were prescribed in accordance with local guidelines, or other relevant metrics. Such reports would include comparison data for other providers in the program, clinic, or health system, allowing recipients of the reports to see how they benchmark. ^{Reference Gerber, Prasad and Fiks14} An example of such a report can be found in supplement 1. Peer comparison was a major component of a behavioral science-based stewardship intervention in the urgent care setting undertaken by Cummings et al in 2020; this intervention reduced inappropriate antibiotic prescriptions by 14.9 percentage points at three rural high-volume urgent care clinics. ^{Reference Cummings, Alajajian and May21} While this study was not specifically introduced as a tele-stewardship study, it illustrates the effectiveness of such an intervention. This approach has been used as a component of multiple successful stewardship interventions. ^{Reference Stenehjem, Wallin and Willis5,Reference Gerber, Prasad and Fiks14} An advantage of this approach is that providers may be incentivized to change their prescribing practices when they are engaged in friendly competition with their peers. Such interventions can also be generalized up to the clinic level, comparing locations instead of individuals, though this approach loses the advantages of individualized feedback. Drawbacks to peer comparison are that generating such individualized data is a time and labor-intensive process. It also may not be as applicable to very rural settings that do not have many comparator providers, though such situations would likely not be common.

Education

Educational interventions are another common feature of tele-stewardship programs. ^{Reference Klatt, Schulz and Fleischman4,Reference Stenehjem, Wallin and Willis5,Reference Cantey, Correa, Dugi, Huff, Olaya and Farner16,Reference Shively, Moffa and Paul17,Reference Stenehjem, Hersh and Buckel20,Reference Moehring, Yarrington and Davis22} These interventions focus on providing prescribers with up-to-date information regarding best practices associated with antimicrobial prescribing. Education could take a variety of forms including in person or virtual lectures or case-based discussions, ^{Reference Klatt, Schulz and Fleischman4,Reference Stenehjem, Wallin and Willis5,Reference Cantey, Correa, Dugi, Huff, Olaya and Farner16,Reference Stenehjem, Hersh and Buckel20,Reference Moehring, Yarrington and Davis22} written or electronic materials, ^{Reference Stenehjem, Wallin and Willis5,Reference Shively, Moffa and Paul17,Reference Stenehjem, Hersh and Buckel20} and even internal podcasts. ^{Reference Stenehjem, Wallin and Willis5} There are several advantages to educational interventions. Written materials and guides do not require an extensive time investment after their initial creation. Lecture or discussion-style interventions allow for direct engagement with prescribers. Successful educational interventions require fewer resources than audit and feedback. They are also more easily scaled to a large number of prescribers than audit and feedback, which is important when considering the rural outpatient context. There are limitations to educational interventions, however. The effectiveness of an educational intervention depends heavily on prescriber engagement. Lectures may not be attended by prescribers uninterested in antimicrobial stewardship, and written materials can be easily ignored by prescribers who feel their prescribing practices do not need to change. Strategies to increase provider engagement with antibiotic stewardship initiatives are needed, including offering continuing education credits, and requiring participation by clinic or health system leadership, emphasizing why leadership commitment was identified by the CDC as a pillar of any stewardship program. ⁶ Stenehjem’s 2018 comparative study of different levels of stewardship intervention revealed that programs that included only interventions that required prescriber initiative to engage with such as provider education were less effective at reducing overall antibiotic use than programs that included strategies like audit and feedback where the antibiotic prescribing data is “pushed” to the providers by the stewardship team. ^{Reference Stenehjem, Hersh and Buckel20} Educational stewardship interventions have the potential for significant impact in rural settings because they are well suited to the unique challenges of such sites. This limitation has been seen in educational interventions ranging from antibiotic prescribing to depression treatment. ^{Reference Stenehjem, Hersh and Buckel20,Reference Gilbody, Whitty, Grimshaw and Thomas23}

Local guideline creation

Creation of local guidelines regarding antimicrobial prescribing is a feature of several successful stewardship programs. ^{Reference Klatt, Schulz and Fleischman4,Reference Stenehjem, Wallin and Willis5,Reference Shively, Moffa and Paul17,Reference Stenehjem, Hersh and Buckel20} Such programs have developed syndrome-specific local guidelines designed to support antimicrobial stewardship minded decision making. Most of these studies focused on implementing stewardship in small community hospitals, though one focused on urgent care clinics. ^{Reference Stenehjem, Wallin and Willis5} These studies included adults and children across a variety of healthcare systems across the US. These interventions tend more toward action for policy and practice rather than education in terms of the CDC’s core elements. The advantages and disadvantages of local guidelines are similar to those associated with written educational interventions, however. Guideline creation has the additional advantage of carrying institutional approval- it is possible that providers will be more likely to follow stewardship recommendations when they know such recommendations have been approved by institutional authorities. This has the additional benefit of lending further credence to a provider’s recommendations when discussing antimicrobial choices with families, a benefit identified by Spencer et al in their 2022 qualitative assessment of pediatric antibiotic prescription decision making. ^{Reference Spencer, Katz, Staub, Audet and Banerjee24} The presence of local guidelines can facilitate further interventions such as implementation of clinical decision support tools into an EHR, as demonstrated by Khadem et al. ^{Reference Khadem, Ergen and Salata25} The main adaptation necessary for outpatient rural implementation of this strategy would be to ensure that the guidelines covered conditions such as URIs and community acquired pneumonia where antimicrobial stewardship is most needed in outpatient medicine.

Direct consultation with infectious disease experts

Another element sometimes included in tele-stewardship programs is direct consultation with ID experts, which is especially useful in rural areas that are often far from academic medical centers and lack access to subspecialists. ^{Reference Cantey, Correa, Dugi, Huff, Olaya and Farner16,Reference Cummings, Alajajian and May21,Reference Jelinski, Julien and Cook26} This intervention would allow patients or providers to consult with specialists regarding questions related to antimicrobial prescribing. This would allow for prescribing advice to be tailored to each unique situation, an advantage over more general educational interventions. Such direct consultation has been reported by prescribers to be easy to use and supportive of patient care. ^{Reference Jelinski, Julien and Cook26} Disadvantages include the need for prescribers to actively engage with the intervention for there to be any value added. Stenehjem’s 2018 study demonstrated no improvement in antibiotic use in interventions that included an ID hotline without more comprehensive audit and feedback interventions, though there was a decrease in the use of restricted antibiotics. ^{Reference Stenehjem, Hersh and Buckel20} This kind of intervention may be considered more of a telemedicine intervention rather than a tele-stewardship intervention considering its direct effect on clinical care, though it still bears mentioning as it is a common element bundled into tele-stewardship interventions and may help to increase provider “buy-in” to stewardship recommendations through relationship building with consultants.

Electronic health record nudges

Electronic decision support at the time of order entry has been shown to decrease inappropriate antibiotic prescribing. ^{Reference Klatt, Schulz and Fleischman4,Reference Stenehjem, Wallin and Willis5,Reference Shively, Moffa and Paul17,Reference Stenehjem, Hersh and Buckel20,Reference Moehring, Yarrington and Davis22} These EHR nudges usually were implemented in inpatient settings with some studies including both children and adults. EHR changes included alteration of order sets, quick links to local guidelines or antibiograms, restricting antibiotics and requiring co-signatures by approving providers, and requiring further information prior to prescribing a particular antibiotic. These kinds of interventions generally require IT expertise at the beginning but then require little in terms of ongoing support making them easy to scale to larger populations. Restriction of certain antibiotics unsurprisingly leads to less use, ^{Reference Shively, Moffa and Paul17,Reference Stenehjem, Hersh and Buckel20} although such interventions are unlikely to apply well to the outpatient rural context as the antibiotics restricted in hospitals are not used often in the outpatient context. Another limitation is that such interventions would also require a common EHR among the outpatient sites. Some rural sites may not use an EHR at all. Additionally, EHR modifications require substantial information technology (IT) resources and expertise, which may not always be available in all settings.