Compass for antibiotic stewardship: using a digital tool to improve guideline adherence and drive clinician behaviour for appendicitis treatment in the emergency department.

BACKGROUND: Antibiotic stewardship in the ED is important given the increasing prevalence of multidrug resistance associated with poorer patient outcomes. The use of broad-spectrum antibiotics in the ED for infections like appendicitis is common. At baseline, 75% of appendicitis cases at our institution received broad-spectrum ertapenem rather than the recommended narrower-spectrum ceftriaxone/metronidazole combination. We aimed to improve antibiotic stewardship by identifying barriers to guideline adherence and redesigning our appendicitis antibiotic guideline. METHODS: Using the 'Fit between Individuals, Task and Technology (FITT)' framework, we identified barriers that preventclinicians from adhering to guidelines. We reformatted a clinical guideline and disseminated it using our ED's clinical decision support system (CDSS), E*Drive. Next, we examined E*Drive's user data and clinician surveys to assess utilisation and satisfaction. Finally, we conducted a retrospective chart review to measure clinician behaviour change in antibiotic prescription for appendicitis treatment. RESULTS: Data demonstrated an upward trend in the number of monthly users of E*Drive from 1 April 2021 to 30 April 2022, with an average increase of 46 users per month. Our clinician survey results demonstrated that >95% of users strongly agree/agree that E*Drive improves access to clinical information, makes their job more efficient and that E*Drive is easy to access and navigate, with a Net Promoter Score increase from 26.0 to 78.3. 69.4% of patients treated for appendicitis in the post-intervention group received antibiotics concordant with our institutional guideline compared with 20.0% in the pre-intervention group (OR=9.07, 95% CI (3.84 to 21.41)). CONCLUSION: Antibiotic stewardship can be improved by ensuring clinicians have access to convenient and up-to-date guidelines through clinical decision support systems. The FITT model can help guide projects by identifying individual, task and technology barriers. Sustained adherence to clinical guidelines through simplification of guideline content is a potentially powerful tool to influence clinician behaviour in the ED.

Limb Salvage Does Not Predict Functional Limb Outcome after Revascularization for Traumatic Acute Limb Ischemia.

BACKGROUND: Traumatic vascular injury leading to acute limb ischemia (ALI) is an uncommon problem with a potential for high morbidity. We describe a contemporary series of patients with traumatic ALI managed primarily by vascular surgeons at a tertiary referral center and review factors associated with limb salvage and functional limb outcomes. METHODS: We conducted a single institution, retrospective review of all patients requiring revascularization for upper extremity (UE) and lower extremity (LE) ALI secondary to trauma from 2013 to 2016. Demographic data, transfer timing, injury severity score (ISS), Rutherford classification (RC), preoperative imaging, level of occlusion, procedural information, fasciotomy characteristics, and discharge disposition were reviewed. Outcome measures included limb salvage and functional limb outcomes. RESULTS: We identified 68 patients with traumatic ALI requiring revascularization. The majority of patients had moderate ISS scores, were RC 2a or 2b on presentation (65%), were transferred from another institution (53%), and underwent preoperative imaging (62%) with expeditious time to operation (median 4.5 hr). The most common location of vascular injury for UE was axillary-brachial (88%) and for LE was femoral-popliteal (69%). Open vascular procedures dominated the treatment strategy, and the median number of operations was 3. Fasciotomy was performed in 25% of UE and 58% of LE injuries. Shunts were utilized in only 2 patients. Overall LS was 94% for UE and 78% for LE. The median length of stay (LOS) was 11 days, with 25% of patients discharged to a skilled nursing facility. Follow-up was obtained for 59% of patients. For UE injuries, 57% of patients had no or minimal functional deficits, while 33% had major functional deficits and 10% underwent amputation. For LE injuries, 68% of patients had no or minimal functional deficits, while 6% had major functional deficits, and 26% had undergone amputation. Rutherford class and the number of operations performed were independent predictors of amputation and functional limb at follow-up in our logistic regression model (P < 0.05). CONCLUSIONS: Revascularization for traumatic ALI yields high limb salvage rates in patients with RC 1 and 2 ischemia and patients with UE injuries. However, limb salvage does not necessarily equate to good functional outcomes. This signifies the complex nature of injuries in this patient population, especially when multiple operations are required.

Designing clinical guidelines that improve access and satisfaction in the emergency department.

Clinical guidelines are evidence-based clinician decision-support tools that improve health outcomes, reduce patient harm, and decrease healthcare costs, but are often underused in emergency departments (EDs). This article describes a replicable, evidence-based design-thinking approach to developing best practices for guideline design that improves clinical satisfaction and usage. We used a 5-step process to enhance guideline usability in our ED. First, we conducted end-user interviews to identify barriers to guideline usage. Second, we reviewed the literature to identify key principles in guideline design. Third, we applied our findings to create a standardized guideline format, incorporating rapid cycle learning and iterative improvements. Fourth, we ensured the clinical validity of our updated guidelines by using a rigorous process for peer review. Lastly, we evaluated the impact of our guideline conversion process by tracking clinical guidelines access per day from October 2020 to January 2022. Our end-user interviews and review of the design literature revealed several barriers to guideline use, including lack of readability, design inconsistencies, and guideline complexity. Although our previous clinical guideline system averaged 0.13 users per day, >43 users per day accessed the clinical guidelines on our new digital platform in January 2022, representing an increase in access and use exceeding 33,000%. Our replicable process using open-access resources increased clinician access to and satisfaction with clinical guidelines in our ED. Design-thinking and use of low-cost technology can significantly improve clinical guideline visibility and has the potential to increase guideline use.

Discharge Navigator: Implementation and Cross-Sectional Evaluation of a Digital Decision Tool for Social Resources upon Emergency Department Discharge.

INTRODUCTION: Many patients have unaddressed social needs that significantly impact their health, yet navigating the landscape of available resources and eligibility requirements is complex for both patients and clinicians. METHODS: Using an iterative design-thinking approach, our multidisciplinary team built, tested, and deployed a digital decision tool called "Discharge Navigator" (edrive.ucsf.edu/dcnav) that helps emergency clinicians identify targeted social resources for patients upon discharge from the acute care setting. The tool uses each patient's clinical and demographic information to tailor recommended community resources, providing the clinician with action items, pandemic restrictions, and patient handouts for relevant resources in five languages. We implemented two modules at our urban, academic, Level I trauma center. RESULTS: Over the 10-week period following product launch, between 4-81 on-shift emergency clinicians used our tool each week. Anonymously surveyed clinicians (n = 53) reported a significant increase in awareness of homelessness resources (33% pre to 70% post, P<0.0001) and substance use resources (17% to 65%, P<0.0001); confidence in accessing resources (22% to 74%, P<0.0001); knowledge of eligibility criteria (13% to 75%, P<0.0001); and ability to refer patients always or most of the time (11% to 43%, P<0.0001). The average likelihood to recommend the tool was 7.8 of 10. CONCLUSION: Our design process and low-cost tool may be replicated at other institutions to improve knowledge and referrals to local community resources.