Background: Antimicrobial stewardship programs (ASPs) seek to reduce the prevalence of antimicrobial-resistant and healthcare-associated infections. There are limited infectious disease (ID) physicians and pharmacists to support these ASPs, particularly in rural areas. The Veterans Health Administration has a robust telehealth program in place. Our previous work has demonstrated the feasibility of using telehealth modalities to support ASPs at rural Veterans Affairs medical centers (VAMCs) by pairing them with an ID expert from a larger, geographically distant, VAMC. This program, dubbed the Videoconference Antimicrobial Stewardship Team (VAST), emphasizes discussion of patients undergoing treatment for an active infection and additional relevant clinical topics with a multidisciplinary team at the rural VA. VAST implementation is ongoing at VAMCs. To understand and compare the qualitative differences in implementation, we used process maps to describe the VAST at 3 VAMC dyads. Methods: Team members from each dyad participated in interviews at 3, 6, and 9 months after beginning their VAST sessions. Questions addressed several aspects of VAST implementation and included identifying cases and topics to discuss; advance preparation for meetings; the frequency and general structure of VAST meetings; and documentation including workload capture. The research team used the responses to develop process maps to permit visual display and comparison of VAST implementation. Results: The first dyad began in January 2022 and the third in March 2022. The sessions had 3 phases: preparation, team meeting, and documentation of experts’ recommendations. Tasks were shared between VAST champions at the rural VAMC and the ID experts (Fig. 1). The preparation phase showed the most variation among the 3 dyads. In general, champions at the rural VA identified cases and topics for discussion that were sent to the ID expert for review. The approaches used to find cases and the type of preparatory work by the ID expert differed. Team meetings differed in both frequency and participation by professionals from the rural site. Documentation of expert recommendations processes appeared similar among the dyads. Discussion: Each of the 3 dyads implemented VAST differently. These results suggest that the overall structure of the VAST is readily adaptable and that each site tailored VAST to suit the clinical needs, workflow, and culture of their partner facility. Future work will seek to determine which aspects in the preparation, team meeting, or documentation phases are associated with successful ASPs, including assessment of quantitative and qualitative outcomes.

Disclosures: None

Background: Identification of predictors of antibiotic use can inform targeted antimicrobial stewardship initiatives and can account for sources of bias in before-and-after interventional stewardship studies. To date, no study has identified clinical predictors of antimicrobial use within intensive care units (ICUs), where antimicrobial resistance is most prevalent and problematic. Methods: As part of an ongoing prospective, single-arm, pilot feasibility trial of an ICU diagnostic stewardship intervention, we performed a nested retrospective cohort study to explore associations between patient clinical variables and ICUs antimicrobial use and resistance rates (AURs). We included all patients hospitalized in 3 ICUs (surgical, medical, and cardiac) from 2017 to 2021 at Michigan Medicine, a large, tertiary-care, academic medical center. Data were extracted from the electronic medical record using a structured query. Admission-level data were captured, including patient demographics, medical comorbidities, International Classification of Disease, Tenth Revision (ICD-10) admission diagnoses, as well as calendar day-level data including vital signs, clinical and microbiologic laboratory data, measures of acute severity of illness, ventilator–supplemental oxygen metrics, and procedural interventions using current procedural terminology (CPT) codes. ICU AURs were defined as total antibiotic days of therapy per patient per 100 ICU days. Associations between clinical variables and ICU AURs were calculated as rate ratios (RRs). Multiple imputation using fully conditional specification was performed to create 25 imputation data sets. Negative binomial regression models were constructed for each data set using backward selection. Variables retained in >50% of models were included in a final multivariate model. Results: In total, 15,177 ICU patient admissions were captured. Age, sex assigned at birth, and race did not independently associate with ICU AURs. Comorbidities, medical interventions, admission diagnoses, and laboratory data that independently associated with ICU-AURs are shown in Table 1. The clinical variables most strongly associated with increased ICU-AURs were pneumonia (RR, 1.55; 95% CI, 1.451.64), bacteremia (RR, 1.35; 95% CI, 1.25– 1.46), intraabdominal infection (RR, 1.35; 95% CI, 1.18–1.55), SOFA score (RR, 1.27; 95% CI, 1.14–1.42), abnormal WBC (RR, 1.26; 95% CI, 1.20–1.32), and immunocompromised status (RR, 1.20; 95% CI, 1.10–1.31). Clinical variables most strongly associated with decreased ICU-AURs were cardiac ICU (RR, 0.56; 95% CI, 0.52–0.60), COVID-19 (RR, 0.62; 95% CI, 0.56–0.70), and receipt of an invasive nonsurgical procedure (RR, 0.90; 95% CI, 0.82–0.98). Conclusions: In this single-center retrospective cohort study, several clinical variables were independently associated with ICU-AURs. These results may be used to identify patient subgroups for potentially high-yield ICU-based stewardship interventions and to account for sources of bias in before-and-after studies of ICU-based stewardship interventions.

Disclosures: None

Background: Antimicrobial stewardship programs (ASPs) often rely on International Classification of Diseases, Tenth Revision (ICD-10) codes to assess antibiotic appropriateness for provider feedback. Concordance between encounter ICD-10 codes and documented indication for antibiotics based on manual chart review varies greatly (74%–95%) in the inpatient setting. Data on concordance between documented indication and ICD-10 code in the outpatient setting are scarce. Methods: We conducted a retrospective cohort study of 650 randomly selected outpatient encounters with antibiotic prescriptions from walk-in and retail clinics between July 15 and September 15, 2021, at Vanderbilt University Medical Center. We performed chart review to compare documented antibiotic indication to the 3 most frequent encounter-associated ICD-10 codes. Also, 12 encounters were excluded due to insufficient available written documentation. The 95% CI for proportion of encounters with concordant antibiotic indications was calculated using Stata version 15.1 software. Results: Of the 638 antibiotic prescriptions with written documentation available for chart review, 204 (32%) were for amoxicillin, 102 (16%) were for amoxicillin-clavulanate, 61 (10%) were for cefdinir, and 56 (9%) were for azithromycin. Overall, 540 (84.6%; 95% CI, 81.6%–87.4%) of 638 encounters had concordant antibiotic indication based on documentation in the note and associated ICD-10 for the encounter. Of the 540 encounters with concordant ICD-10 and documented indications, 348 (64%), 130 (24%), and 35 (6%) were listed as the first, second, and third ICD-10 codes, respectively. An additional 27 (5%) had a concordant ICD-10 code listed beyond the third position. In total, 125 (19.6%) of 638 encounters did not have the intended antibiotic indication as documented in the note in the 3 most frequent encounter-associated ICD-10 codes (whether a lower position or incongruent ICD-10 code with documentation). Of those 125 encounters, 42 (34%) had a documented diagnosis of strep pharyngitis, 16 (13%) had a documented diagnosis of skin or soft-tissue infection, 11 (9%) had a documented diagnosis of urinary tract infection, and 11 (9%) had a documented diagnosis of acute otitis media. Conclusions: Our data suggest that outpatient antimicrobial prescriptions correlate relatively well with encounter ICD-10 codes. However, most ASP prescribing goals aim to reduce inappropriate prescribing to 10% or fewer of prescriptions based on indication. Therefore, providers may not trust individual prescribing feedback that is based on data that is only correct 85% of the time. For ASPs to accurately assess prescribing and provide trusted, meaningful recommendations and specific feedback to individual prescribers, more reliable and valid data are needed. We intend to evaluate whether requiring outpatient antibiotic indications on prescriptions increases data reliability and validity.

Disclosures: None

Background: Antibiotics are frequently prescribed in patients with COVID-19 infections to treat secondary bacterial pneumonia. The pneumonia panel (PNP) is a molecular diagnostic tool that rapidly detects 33 bacterial and viral targets. The utility of this panel in COVID-19 patients and how it may direct antibiotic use is unknown. We sought to understand the utilization of PNP in patients with COVID-19 pneumonia over time by comparing clinical parameters, microbiologic results, and antibiotic use between May–December 2020 and January–July 2022. Methods: We implemented the PNP in May 2020 with antimicrobial stewardship guidance, provider education, and order restriction to critical care and infectious disease clinicians. From February–July 2021 prescribers received regular structured antimicrobial stewardship feedback regarding PNP results; from August 2021 to January 2022, no antimicrobial stewardship feedback was provided; from February to July 2022, intermittent feedback was provided. We compared PNP and culture results from sputum or bronchoalveolar lavage samples and antibiotic use and modification within 24 hours of PNP result from patients with confirmed COVID-19 pneumonia between May–December 2020 and January–July 2022. Clinical data and antibiotic use were abstracted through chart review. We excluded patients who died within 72 hours of PNP, those who had concurrent nonpulmonary infections, and those whose COVID-19 test was >30 days prior. Results: We included 114 patients in 2020 and 71 patients in 2022. The overall median age was 61 years, 71% were male, and 66% were mechanically ventilated without statistical differences between the cohorts, including their comorbidities. Acute or worsening hypoxia remained the predominant indication for PNP (77% in 2020 vs 75% in 2022, NS). The median number of days between admission and PNP was 4 (IQR, 1–8) in 2020 versus 3 (IQR, 1–7), and the difference was not significant. PNP and culture results in Table 1 show that Staphylococcus aureus and Hemophilus influenzae were the pathogens most commonly identified. Table 2 describes empiric prescribing and modifications for commonly prescribed antibiotics. Prescribers used empiric cefepime and ceftriaxone more in 2020 and vancomycin more in the 2022 group; however, these were not statistically significant. Cefepime de-escalation was more common in 2022 (53% vs 28%; P = .03). Antibiotic modifications within 24 hours of PNP remained similar in 2020 vs 2022. Although vancomycin cessation was more common in 2020 (78%) versus 2022 (57%), the difference was not statistically significant. Conclusions: With ASP guidance, PNP may be a useful tool to stop or target antibiotics for secondary bacterial pneumonia in COVID-19 pneumonia. Early vancomycin cessation (prior to culture results) may be an enduring consequence of PNP implementation.

Disclosures: None

Background: Most antibiotic use occurs in ambulatory settings. Antibiotic prescribing for children living in the United States in medically underserved areas or in populations is not well understood.

Objective: To characterize antibiotic prescribing for children in a practice-based research network (PBRN).

Design and Methods: In this retrospective cohort study, we characterized oral antibiotic prescribing in a large PBRN. Patients aged 0–17 years with at least 1 in-person visit between January 1, 2014, and December 31, 2018, at 1 of 25 primary-care clinics located within the WWAMI (Washington, Wyoming, Alaska, Montana, and Idaho) region of the Practice and Research Network (WPRN) were included. Data were extracted from DataQUEST, a centralized data repository from included primary-care clinics. Encounters for wellness visits or those lacking a diagnosis code and patients with complex chronic conditions were excluded. Diagnoses were categorized using International Classification of Disease, Ninth Revision (ICD-9) and ICD-10 codes. Oral antibiotics prescribed within 3 days of an encounter were associated with that encounter. Demographic data included age, sex, race, and ethnicity. Antibiotic appropriateness was determined using a previously published 3-tiered classification system using diagnosis codes as always, sometimes, or never appropriate. Patient-level data (ZIP codes) were used to designate medically underserved areas (MUAs) and medically underserved populations (MUPs). Antibiotic prescribing was then analyzed within these groups. Results: In total, 37,314 patients across 206,845 encounters were included, of which 34,601 encounters (17%) resulted in antibiotic prescription (Table 1). Of those, appropriateness data were available for 34,286 (99%). Of the antibiotics prescribed, 14% were always appropriate, 57% were sometimes appropriate, and 27% were never appropriate (1% missing). In total, 64% and 35% of encounters occurred with patients from an MUA and MUP, respectively. Conclusions: Targets to improve oral antibiotic prescribing for children in a large PBRN include antibiotic prescribing for diagnoses that never require an antibiotic. Larger comparative studies may focus on the role (if any) that MUA/MUP has on antibiotic prescribing.

Disclosures: None

Background: Critical-access hospitals (CAHs) are required to meet the CDC 7 Core Elements of antimicrobial stewardship programs (ASPs). CAHs have lower adherence to the core elements than larger acute-care hospitals, and literature defining which core-element deficiencies exist within CAHs as well as barriers to adherence is lacking. Methods: We evaluated 21 CAH ASPs (5 in Nebraska and 15 in Iowa) that self-identified as potentially deficient in the Core Elements, via self-assessment followed by in-depth interviews with local ASP team members to assess adherence to the CDC Core Elements for ASPs. Core-element compliance was rated as either full (1 point), partial (0.5), or deficient (0), with a maximum score of 7 per ASP. High-priority recommendations to ensure core-element compliance were provided to facilities as written feedback. Self-reported barriers to implementation were thematically categorized. Results: Among the 21 CAH ASPs, none fully met all 7 core elements (range, 2.5–6.5), with a median of 5 full core elements met (Fig. 1). Only 6 ASPs (28.6%) had at least partial adherence to each of the 7 core elements. Action (21 of 21, 100%) and leadership commitment (16 of 21, 76.2%) were the core elements with the highest adherence, and accountability (4 of 21, 19%) and education (9 of 21, 42.9%) were the lowest. The most frequent high-priority recommendations were to provide physician and pharmacist leader ASP training (19 of 21, 90.5%), to track antimicrobial stewardship interventions (12 of 21, 57.1%), and to provide or track educational activities (12 of 21, 57.1%) (Fig. 2). One-third of programs were recommended to establish a physician leader. The most commonly self-identified barriers to establishing and maintaining an ASP were a lack of dedicated resources such as time of personnel (15 of 20, 75%), lack of infectious diseases expertise and training (8 of 20, 40%), and electronic medical record limitations (5 of 20, 25%) (Fig. 3). Conclusions: CAH ASPs demonstrate several critical gaps in achieving adherence to the CDC Core Elements, primarily in training for physician and pharmacist leaders and providing stewardship-focused education. Further resources and training customized to the issues present in CAH ASPs should be developed.

Disclosures: None

Background: The FIRST Trial is a 5-year study funded by the Agency for Healthcare Research and Quality. Our investigation is situated within a more extensive study to restrict fluoroquinolone antibiotics by requiring providers to obtain authorization from an infectious disease physician before prescribing fluoroquinolones. Our research team is performing a systematic evaluation to identify organizational characteristics and influencers of the fluoroquinolone preprescription authorization implementation process to understand variables that may facilitate or hinder implementation success. Methods: To address this critical gap, we present a qualitative analysis from our ongoing, multisite research project aimed at systematically assessing the adoption of an antimicrobial stewardship intervention in the form of an EHR-integrated best-practice alert (BPA) at each site to identify work system factors that impact uptake and variability in the implementation of the BPA at each location. The evaluation provides a detailed explanation of activities through the implementation process (eg, before implementation, during implementation, and after implementation) to assess how an organization effectively negotiates the phases and transitions, ultimately influencing the impact of the intervention. We have used a contextual determinant framework (CFIR) that has enabled us to perform a systematic and comprehensive exploration and identification of potential explanatory themes or variables to shed light on the complex social phenomenon of implementation. Results: Participants who will be a part of our poster presentation will learn about implementing a BPA, the potential barriers to implementation, and strategies for overcoming these barriers. Stakeholders within our study include site coordinators, medical doctors, nurses, pharmacists, and clinical informaticists. Our analysis synthesizes their experiences implementing and sustaining this evidence-based antimicrobial stewardship intervention. It includes (1) a detailed description of the process of change, (2) work-system factors (eg, inner setting and outer setting) that they believe influenced the success of the intervention, (3) barriers and facilitators (eg, CFIR constructs) within the implementation process; and (4) description of how these could have influenced the outcomes of interest (eg, implementation and intervention effectiveness). Conclusions: Our research is expected to advance patient safety research and initiatives by providing a more robust approach to performing systematic intervention evaluations. By outlining stakeholders’ experiences within our study, implementation leaders within healthcare systems will utilize our findings to aid them in their design and implementation process when designing and implementing similar types of healthcare interventions.

Disclosures: None

Background: COVID-19 is associated with symptoms, clinical findings, and laboratory abnormalities that raise concern for secondary infections. Excess antimicrobial use despite low rates of secondary infections has been reported and presents a continuing challenge for antimicrobial stewardship programs (ASPs), particularly during COVID-19 surges. The objective of this study was to analyze the appropriateness of antimicrobial use in patients with extended antimicrobial therapy during 2 distinct COVID-19 hospital surges. Methods: We conducted an observational, retrospective, cohort study of COVID-19 patients admitted to our 548-bed community teaching hospital between November and December 2021 (ie, the SARS-CoV-2 delta-variant predominant phase) and January–February 2022 (ie, the SARS-CoV-2 omicron-variant predominant phase) and who received antibiotics for >4 days without positive cultures. Demographic and clinical data were obtained from the institutional data warehouse. Infectious diseases–trained researchers evaluated the appropriateness of antimicrobials based on diagnostic and clinical reporting and institutional antimicrobial stewardship guidelines. Patients were considered to have probable secondary bacterial infection if they had 2 of the following symptoms: fever, unexplained leukocytosis, worsening secretions, or hypoxia and/or imaging. The outcomes of interest included confirmed infections and excess antimicrobial days. Categorical and continuous variables were analyzed using χ2 tests, Fisher exact tests, and Mann-Whitney U tests, respectively. Statistical significance was defined as P ≤ .05. Results: In total, 87 patients were included in the study. Moreover, 56 patients were identified in the SARS-CoV-2 delta-variant predominant phase and 31 patients were identified in the SARS-CoV-2 omicron-variant predominant phase. The groups were similar, with higher vaccination rates in the SARS-CoV-2 omicron-variant predominant group (37.5% vs 64.5%; P = .016). Patients in the SARS-CoV-2 omicron-variant predominant group required less mechanical ventilation (39.3% vs 16.1%; P = .025). There were no significant differences in infectious diseases consultation, immunomodulator or remdesivir use, antimicrobials classes prescribed, or antimicrobial days of therapy or duration between cohorts. There were no significant differences in length of stay, 30-day mortality, or 30-day readmissions. Infections were confirmed in 78.6% in the delta-variant group versus 83.9% in the omicron-variant group (P = .55). Pneumonias accounted for 60.7% in the delta-variant group and 40.9%, in the omicron-variant group. Excess antibiotic use occurred in 14.3% of patients in the delta-variant group and in 3.1% of patients in the omicron-variant group (P = .149). There was no significant difference in the duration of inappropriate antimicrobial use between groups in patients without infections: 5 days in the delta-variant group versus 5 days in the omicron-variant group (P = .24). Conclusions: Results demonstrated that most antimicrobial use was appropriate in a challenging patient population lacking positive cultures to guide therapy. Inappropriate antimicrobial utilization occurred demonstrating continued opportunities for our institutional ASP.

Disclosures: None

Background: Many clinical guidelines recommend that clinicians should use antibiograms to decide on empiric antimicrobial therapy. However, antibiograms aggregate epidemiologic data without consideration for any other factors that may affect the risk of antimicrobial resistance (AMR), and little is known about an antibiogram’s reliability in predicting antimicrobial susceptibility. We assessed the diagnostic accuracy of antibiograms as a prediction tool for E. coli clinical isolates in predicting the risk of AMR for individual patients. Methods: We extracted microbiologic and patient-level data from the nationwide clinical data warehouse of the Veterans Health Administration (VHA). We assessed the diagnostic accuracy of the antibiogram for 3 commonly used antimicrobial classes for E. coli: ceftriaxone, fluoroquinolones, and trimethoprim-sulfamethoxazole. First, we retrospectively generated facility-level antibiograms for all VHA facilities from 2000 to 2019 using all clinical culture specimens positive for E. coli, according to the latest Clinical & Laboratory Standards Institute guideline. Second, we created a patient-level data set by including only patients who did not have a positive culture for E. coli in the preceding 12 months. Then we assessed the diagnostic accuracy of an antibiogram for E. coli to predict resistance for the isolates in the following calendar year, using logistic regression models with percentages in the antibiogram as dependent variables. We also set 5 stepwise thresholds at 80%, 85%, 90%, 95%, and 98%, and we calculated sensitivity, specificity, and accuracy for each antimicrobial. Results: Among 127 VHA hospitals, 1,484,038 isolates from 704,779 patients were available for analysis. The area under the ROC curve (AU-ROC) was 0.686 for ceftriaxone, 0.637 for fluoroquinolones, and 0.578 for trimethoprim-sulfamethoxazole, suggesting their relatively poor prediction performances (Fig. 1). The sensitivity and specificity of the antibiogram widely varied by antimicrobial groups and thresholds, with substantial trade-offs. Along with AU-ROC, these metrics suggest poor prediction performances when antibiograms are used as the sole prediction tool (Fig. 2). Conclusions: Antibiograms for E. coli have poor performances in predicting the risk of AMR for individual patients when they are used as a sole tool, and their contribution to the clinical decision making may be limited. Clinicians should also consider other clinical and epidemiologic data when interpreting antibiograms, and guideline statements that suggest antibiogram as a valuable tool for decision making in empiric therapy may need to be reconsidered. Further studies are needed to evaluate the contribution of antibiograms when combined with other patient-level factors.

Disclosures: None

Background: Nationally, >30% of all outpatient antibiotics are unnecessary or inappropriate, and only 52% of outpatients with sinusitis, otitis media, or pharyngitis receive recommended first-line antibiotics. The Minnesota All Payer Claims Database (MN APCD) collects medical claims, pharmacy claims, and eligibility files from private and public healthcare payers. We analyzed claims to describe overall and firstline antibiotic prescribing for acute bronchitis, adult acute sinusitis, and pediatric patients. Results: We analyzed 3,502,013 respiratory events from 1,612,501 members. Acute bronchitis accounted for 179,723 events (5.1%), acute sinusitis accounted for 236,901 adult events (10%), and otitis media accounted for 232,226 pediatric events (19%). Also, 73,385 bronchitis diagnoses (~40%) had no associated antibiotic. Antibiotics were associated with 199,445 adult sinusitis events (84.2%), of which 89,386 (44.8%) were firstline antibiotics, and 190,962 pediatric otitis media events (82.2%), of which 126,859 (66.4%) were firstline antibiotics. Common antibiotic classes used when a firstline drug was not selected were macrolides (28.9%) and tetracyclines (26.8%) for adult acute sinusitis and cephalosporins (61.4%) and macrolides (30.6%) for pediatric otitis media. Compared to the least vulnerable quartile, the most vulnerable social vulnerability index (SVI) quartile had lower odds of receiving firstline antibiotics for adult acute sinusitis if antibiotics were prescribed (OR, 0.90; 95% CI, 0.87–0.94) and higher odds of receiving firstline antibiotics for pediatric otitis media if antibiotics were prescribed (OR, 1.16; 95% CI, 1.12–1.21). Conclusions: Improvement is needed in avoiding antibiotics for acute bronchitis and selecting firstline drugs for sinusitis and otitis media. Additional analyses adjusting for demographic, geographic, and prescriber factors are planned to better understand differences in prescribing appropriateness among Minnesotans.

Disclosures: None

Background: Patients receiving outpatient parenteral antimicrobial therapy (OPAT) are often medically complex and require carefully tailored treatments to address severe and often concomitant infections. Our objective was to illustrate the heterogeneity in antimicrobials used for patients in OPAT, within and across infection diagnosis groups. Methods: We abstracted electronic health record data regarding day-level treatment into a registry of 2,358 OPAT courses (n = 2,072 unique patients) treated in the University of North Carolina Medical Center OPAT program during 2015–2022 (total, 11,861 person weeks; average, 7 OPAT weeks per patient). We classified infection diagnoses into 10 hierarchical or mutually exclusive categories (eg, bacteremia only, diabetic foot infection (DFI) only, osteomyelitis only) (Fig., vertical axes). Accounting for 64 antimicrobial medications and 520 cocktails administered for at least 1 patient day in our OPAT registry, we also defined 18 hierarchical or mutually exclusive classifications of treatment (eg, “daptomycin alone” or “daptomycin and any other antibiotic(s)” (Fig. key). We conducted 2 stratified analyses to describe the heterogeneity across infection diagnoses with respect (1) to medications used at OPAT initiation (patient as unit of analysis) and (2) to medications used throughout OPAT (person time as unit of analysis, allowing for differential OPAT course to other treatment classifications during follow-up). We present stacked bar charts to visualize the intersection between infection diagnosis and treatment group. Results: Among patients in this OPAT registry, 34.6% had osteomyelitis and/or DFI, 4.8% had bacteremia, and 44.6% had multiple infections (Fig. 1). The most common medications in initial OPAT regimens were vancomycin (30.8% of OPAT patients), ceftriaxone (15.0%), and daptomycin (10.9%). We observed overall similarity between the distribution of treatment groups at initiation compared to cumulative person-time during the OPAT course (Figs. 1 and 2). However, we observed heterogeneity in medications by infection diagnosis (Figs. 1 and 2); for example, vancomycin was used in 39% of osteomyelitis cases but only 14% for endocarditis (Fig. 2). For several infection groups (eg, osteomyelitis, DFI, multiple infections, “other” single infections), no treatment classification exceeded 20% use (Figs. 1 and 2). Conclusions: Day-level data on medication use in this monitored registry of patients provided evidence of heterogeneity in the types of medications used throughout treatment in OPAT, which varies within and across infection diagnoses. These data highlight the need for multilayered ascertainment of medication exposure in this medically complex patient population to inform surveillance for adverse effects and guide comparative effectiveness research for postdischarge antibiotic treatment.

Disclosures: None

Background: Clostridioides difficile infection (CDI) poses a health burden to patients and a financial burden to hospital systems. Timely identification of CDI patients can reduce the impacts by allowing for prompt treatment and ensuring that proper isolation precautions are in place to prevent spread. It also ensures correct CDI event categorization according to the NHSN. Community-onset (CO) CDI cases are tested on or prior to hospital day 3, and hospital-onset (HO) CDI are tested on or after hospital day 4. The objective of this study was to determine the effectiveness of utilizing an electronic health record (EHR) report to reduce CDI standard infection ratios (SIRs) by identifying potential CDI cases prior to hospital day 4. Methods: From August of 2021 to September 2022, an EHR report was implemented in a 5-hospital healthcare system in the Midwest to identify patients with 3 or more type 6 or 7 stools in a 24-hour period based on Bristol stool chart classification. All inpatients with 3 or more type 6 or 7 stools in 24 hours without an active order for a Clostridioides difficile test were listed. Patients with a laxative in the previous 48 hours, tube feedings without fever or leukocytosis, or a known cause of diarrhea were excluded. The attending provider of the patients meeting criteria were notified with a recommendation to test for C. difficile or provide alternative reason for symptoms. Results: In total, 26 patients were tested for C. difficile using polymerase chain reaction testing. Of those tested, 5 (19.2%) tested positive for C. difficile. There were 13 HO-CDI cases for the healthcare system during this period, for an SIR of 0.351. If the early identified cases were not identified until after hospital day 3, the SIR had the potential to have been 35.6% greater at 0.476. Conclusions: We were able to identify 5 CDI cases prior to hospital day 4 using an early identification report during this 13-month period. Although these cases may have been identified without the use of the EHR report, we were able to obtain a timely CDI diagnosis, potentially limiting the spread of C. difficile and preventing an increase in the CDI SIR by 35.6%. An EHR report to identify patients meeting C. difficile testing criteria may be an effective way to identify CO-CDI prior to HD 4 and thus reduce CDI SIR

Disclosures: None

Background: Clostridioides difficile infection (CDI) is a serious healthcare-associated infection responsible for >12,000 US deaths annually. Overtesting can lead to antibiotic overuse and potential patient harm when patients are colonized with C. difficile, but not infected, yet treated. National guidelines recommend when testing is appropriate; occasionally, guideline-noncompliant testing (GNCT) may be warranted. A multidisciplinary group at UNC Medical Center (UNCMC) including the antimicrobial stewardship program (ASP) used a best-practice alert in 2020 to improve diagnostic stewardship, to no effect. Evidence supports use of hard stops for this purpose, though less is known about provider acceptance. Methods: Beginning in May 2022, UNCMC implemented a hard stop in its electronic medical record system (EMR) for C. difficile GNCT orders, with exceptions to be approved by an ASP attending physician. Requests were retrospectively reviewed May–November 2022 to monitor for adverse patient outcomes and provider hard-stop compliance. The team exported data from the EMR (Epic Systems) and generated descriptive statistics in Microsoft Excel. Results: There were 85 GNCT orders during the study period. Most tests (62%) were reviewed by the ASP, and 38% sought non-ASP or no approval. Of the tests reviewed by the ASP, 33 (62%) were approved and 20 (38%) were not. Among tests not approved by the ASP, no patients subsequently received CDI-directed antibiotics, and 1 patient (5%) warranted same-admission CDI testing (negative). Of tests that circumvented ASP review, 18 (56%) ordering providers received a follow-up email from an associate chief medical officer to determine the rationale. No single response type dominated: 3 (17%) were unaware of the ASP review requirement, 2 (11%) indicated their patient’s uncharted refusal of laxatives, 2 (11%) indicated another patient-specific reason. Provider avoidance of the ASP approval mechanism decreased 38%, from 53% of noncompliant tests in month 1 to 33% of tests in month 6. Total tests orders dropped 15.5% from 1,129 during the same period in 2021 to 954 during the study period (95% CI, 13.4%–17.7%). Compliance with the guideline component requiring at least a 48-hour laxative-free interval prior to CDI testing increased from 85% (95% CI, 83%–87%) to 95% (95% CI, 93%–96%). CDI incidence rates decreased from 0.52 per 1,000 patient days (95% CI, 0.41–0.65) to 0.41 (95% CI, 0.32–0.53), though the change was neither significant at P = .05 nor attributable to any 1 intervention. Conclusions: Over time and with feedback to providers circumventing the exception process, providers accepted and used the hard stop, improving diagnostic stewardship and avoiding unneeded treatment.

Disclosures: None

Background: Clostridioides difficile infection (CDI) often recurs in patients aged ≥65 years and those with comorbidities. Clinical trials often exclude patients with history of immunosuppression, malignancy, renal insufficiency, or other comorbidities. In a phase 3 trial (ECOSPOR III), SER-109 was superior to placebo in reducing recurrent CDI (rCDI) risk at week 8 and was well tolerated. We report integrated safety data for SER-109 in a broad patient population through week 24 from phase 3 studies: ECOSPOR III and ECOSPOR IV. Methods: ECOSPOR III was a double-blind, placebo-controlled trial conducted in participants with ≥2 CDI recurrences randomized 1:1 to placebo or SER-109. ECOSPOR IV was an open-label, single-arm study conducted in 263 patients with rCDI enrolled in 2 cohorts: (1) rollover participants from ECOSPOR III with on-study recurrence and (2) participants with ≥1 CDI recurrence, inclusive of the current episode. In both studies, the investigational product was administered as 4 oral capsules over 3 days. Treatment-emergent adverse events (TEAEs) were collected through week 8; serious TEAEs and TEAEs of special interest (ie, bacteremia, abscess, meningitis) were collected through week 24. Results: In total, 349 participants received SER-109 in ECOSPOR III and/or ECOSPOR IV (mean age 64.2; 68.8% female). Chronic diseases included cardiac disease (31.2%), immunocompromised or immunosuppressed (21.2%), diabetes (18.9% ), and renal impairment or failure (13.2%). Overall, 221 (63.3%) of 349 participants who received SER-109 experienced TEAEs through week 24. Most were mild to moderate and gastrointestinal. The most common (>5% of participants) treatment related TEAEs were flatulence, abdominal pain and distension, decreased appetite, constipation, nausea, fatigue, and diarrhea. No participants experienced a treatment-related TEAE leading to study withdrawal. Invasive infections were observed in 28 participants (8%); those with identified pathogens were unrelated to SER-109 species, and all were deemed unrelated to treatment by the investigators. There were 11 deaths (3.2%) and 48 participants (13.8%) with serious TEAEs, none of which were deemed treatment related. There were no clinically important differences in the safety profile across subgroups of sex, race, prior antibiotic regimen, or number of CDI recurrences. No safety signals were observed in participants with renal impairment or failure, diabetes, cardiac disease, or immunocompromised or immunosuppressed individuals. Conclusions: In this integrated analysis of phase 3 trials, SER-109, an investigational microbiome therapeutic, was well tolerated in this vulnerable patient population with prevalent comorbidities. No infections, nor those with identified pathogens, were attributed to SER-109 or product species. This safety profile might be expected because this purified product is composed of spore-forming Firmicutes normally abundant in the healthy microbiome.

Financial support: This study was funded by Seres Therapeutics.

Disclosures: None

Background: Pediatric patients often require central venous catheters (CVCs) for a variety of clinical indications, including medication administration, parenteral nutrition, and venous blood sampling. Patients with CVCs are at risk for central-line–associated bloodstream infections (CLABSI). These hospital-acquired infections are often preventable and may lead to increased morbidity and mortality. Clinicians at a 477-bed, freestanding pediatric academic hospital completed a quality improvement project to identify factors that place pediatric patients at increased risk for CLABSI and to outline strategies aimed at CLABSI reduction for our highest-risk patients. Methods: Project leaders completed a literature review to evaluate current research on the topic and then assembled a project team. The team completed a retrospective analysis and categorization of CLABSI cases and reviewed internal CLABSI root-cause analysis data. The group then completed a case–control analysis to identify risk factors in patients with CVCs who developed CLABSIs, compared to patients with CVCs who did not develop CLABSI. Following this analysis, the team created a CLABSI risk-factor tool for use by bedside nurses. This tool described patients with CLABSI risk factors and outlined best practices for CLABSI prevention. Results: Based upon literature review, root-cause analysis data, and retrospective CLABSI case review over the period from 2017 to 2021, an initial list of 9 potential CLABSI risk factors was compiled. A case–control analysis was performed comparing 97 CLABSI cases with 103 matched controls. Univariate, multivariate, and additional covariate analyses were employed to identify 3 factors placing pediatric patients at increased risk for CLABSI. These included (1) multiple enteral devices (ie, 2 or more devices, including gastrostomy tube, jejunostomy tube, gastrostomy or jejunostomy tube, ostomy, and peritoneal drain); (2) multiple CVC entries (ie, CVC used for medications and venous sampling); and (3) long-term CVC plus parenteral nutrition (CVC in place for >21 days and receiving parenteral nutrition and/or intralipids). Conclusions: Pediatric patients with central venous access are vulnerable to CLABSI, and certain patients may be at increased risk. Frontline clinicians may be able to identify these patients and adopt best practices to prevent infection. A tool for use by bedside nurses can be a useful adjunct to existing CLABSI prevention practices.

Disclosures: None

Background: Information regarding catheter-related bloodstream infections (CRBSIs) among patients on hemodialysis in the Philippines is lacking. Objective: In this study, we described the clinical profile, CRBSI incidence density, and outcomes of patients in a single-center hemodialysis unit. Methods: A retrospective review of patients receiving hemodialysis (HD) through a central venous catheter (CVC) from January 2016 to December 2020 in a tertiary-care, private hospital was performed. Baseline demographic data were recorded, and CRBSI incidence density rates (no. of CRBSIs per 1,000 catheter days) were calculated. Results: Of 868 hemodialysis patients (57%), 499 used a CVC and were followed for 182,135 catheter days. Half were male (248 of 499, 49.7%) with a median age of 62 years (range, 24–90). Only 48 (9.6%) of 499 developed CRBSI, with an overall CRBSI incidence of 2.63 per 1,000 catheter days. Of those with CRBSI, 31 (64.6%) of 48 were female. The median age was 74.5 years (range, 30–90). Hypertension (40 of 48, 83.3%) and diabetes mellitus (26 of 48, 54.2%) were frequent comorbidities. Fever with chills was the most common symptom, occurring in 30 (62.5%) of 48 patients. Both gram-positive (n = 24) and gram-negative (n = 25) organisms were isolated. Staphylococcus aureus was the most common gram-positive isolate (14 of 25, 56%); isolates from the order Enterobacterales (12 of 24, 50%) were the most common gram-negative organisms. More CRBSIs occurred among those with a nontunneled versus tunneled CVCs (28 vs 20). The median time to CRBSI occurrence was 7 weeks (range, 0.43–280) from CVC insertion. The most common empiric treatment was either vancomycin (n = 28) or piperacillin-tazobactam (n = 26), which were also used in combination (11 of 28, 39.3%). Treatment involved CVC removal in most patients (34 of 48, 70.8%), either alone (n = 1), or with systemic antibiotic therapy (SAT; n = 16), or SAT plus antibiotic lock therapy (ALT; n = 17). The remainder (14 of 48, 29.2%) retained their CVCs because of difficult access, and received both SAT and ALT. Attributable mortality (6 of 9, 33%) and overall mortality (9 of 48, 18.5%) were high. Mortality of those whose CVC was retained was lower compared to those whose line was removed: (3 of 9, 33%) versus (6 of 9, 66%). Conclusions: The overall CRBSI rate in our hemodialysis unit was low and occurred more commonly in the older age group with a nontunneled CVC. Both gram-positive and gram-negative pathogens were common. CRBSI was associated with high attributable mortality. Successful treatment often required CVC, SAT, and ALT. However, CVC retention was a viable option in some patients with specific limiting factors such as difficult access.

Disclosures: None

Background: Central-line–associated bloodstream infection (CLABSI) rates increased nationally during COVID-19, the drivers of which are still being characterized in the literature. CLABSI rates doubled during the SARS-CoV-2 omicron-variant surge at our rural academic medical center. We sought to identify potential drivers of CLABSIs by comparing period- and patient-specific characteristics of this COVID-19 surge to a historical control period. Methods: We defined the study period as the time of highest COVID-19 burden at our hospital (July 2021–June 2022) and the control period as the previous 2 years (July 2019–June 2021). We compared NHSN CLABSI standardized infection ratios (SIRs), central-line standardized utilization ratios (SURs), completion of practice evaluation tools (PETs) for monitoring of central-line bundle compliance, and proportions of traveling nurses. We performed chart reviews to determine patient-specific characteristics of NHSN CLABSIs during these periods, including demographics, comorbidities, central-line characteristics and care, and microbiology. Results: The CLABSI SIR was significantly higher during the study period than the control period (0.89 vs 0.52; P = .03); the SUR was significantly higher during the study period (1.08 vs 1.02; P < .01); the PET completion per 100 central-line days was significantly lower during the study period (23.0 vs 31.5; P < .01); and the proportion of traveling nurses was significantly higher during the study period (0.20 vs 0.08; P < .01) (Fig. 1). Patients with NHSN CLABSIs during the study period were more likely to have a history of COVID-19 (27% vs 3%; P = .01) and were more likely to receive a higher level of care (60% vs 27%; P = .02). During the study period, more patients had multilumen catheters (87% vs 61%; P = .04). The type of catheter, catheter care (ie, dressing changes and chlorhexidine bathing), catheter duration before CLABSI, and associated microbiology were similar between the study and control periods (Table 1). Conclusions: During the SARS-CoV-2 omicron-variant surge, the increase in CLABSIs at our hospital was significantly associated with increased central-line utilization, decreased PET completion, and increased proportion of traveling nurses. Critical illness and multilumen catheters were significant patient-specific factors that differed between CLABSIs from the study and control periods. We did not observe differences in catheter type, duration, or catheter care. Our study highlights key modifiable risk factors for CLABSI reduction. These findings may be surrogates for other difficult-to-measure challenges related to the culture of safety during a global pandemic, such as staff education related to infection prevention and daily review of central-line necessity.

Disclosures: None

Background: Central-line–associated bloodstream infections (CLABSIs) are an important public health issue. Recent data from the CDC have shown an increase in healthcare-associated infections (including CLABSI) during the COVID-19 pandemic. Therefore, the main aim of this project was to analyze the epidemiology of central-line–associated bloodstream infection during different periods at the Baystate Medical Center (Springfield, MA) before, during, and after COVID-19 peaks of infection. Methods: Two specific periods were considered during the year (quarter January–March and quarter July–September) to consider potential seasonal variations, and the incidence of CLABSI during those 2 quarters was analyzed for 4 different years: 2019 (prepandemic), 2020–2021 (intrapandemic), and 2022 (postpandemic). An analysis of the microbial pathogens causing line infections was also performed to investigate differences described by other authors. Results: In total, 97 CLABSI (all from different patients) were reported into the NHSN website during the 8 periods considered. The average age of the patients was 55 years, with a male:female ratio of 57%:43%, and 14 renal patients were on dialysis. The CLABSI rates ranged from a minimum of 1.11 in Q1 of 2020 (start of COVID) to a maximum of 2 in Q3 of 2021 (SARS-CoV-2 delta variant) (Table 1). A statistical comparison of the pre–COVID-19 period with the respective quarters during the pandemic years (2020, 2021, and 2022) did not show any significant differences (Table 2). In term of microbiological data, of the 97 patients with CLABSIs, most of the patients (n = 70) had only 1 pathogen isolated, 14 patients had 2 pathogens, and 3 patients had 3 pathogens, bringing the total number of bacteria cultured to 117. Candida spp and Enterococcus spp were the most frequently isolated pathogens at 19% and 13%, respectively (Fig. 1). There was no statistically significant difference between the pre–COVID-19 and intra–COVID-19 periods for Candida spp (rate ratio, 1.391; 95% CI, 0.5477–3.533; P = .48) or Enterococcus spp (rate ratio, 2.385; 95% CI, 0.8365–6.798; P = .09). Conclusions: The COVID-19 pandemic did not seem to have an impact on the local epidemiology at Baystate Medical Center in terms of CLABSI rates or type of pathogens causing infections, but the sample size taken into consideration may not have been powerful enough to detect statistical significance.

Note. This project was carried out as part of Dr Satta’s MPH requirements at UMass.

Disclosures: None

Background: Central-line–associated bloodstream infections (CLABSIs) significantly burden the US population and healthcare system. Reporting facilities in Tennessee consistently omit race and ethnicity data in the NHSN despite having the option to enter. Racial and ethnic disparities are well documented across many health outcomes, including patient safety. CLABSIs were compared among 3 racial groups to better understand the impact of race on CLABSI incidence in Tennessee. Methods: CLABSI data from NHSN were linked with records from the TN Hospital Discharge Data System (HDDS) for 2018–2021. A multivariable linear regression model was used to determine relative risk (RR) between racial groups for contracting a CLABSI after controlling for confounding variables including Charlson comorbidity index (CCI) and social vulnerability index (SVI) scores. Statistical significance was set at P < .05. Data linkage and statistical analyses were performed in SAS version 9.4 software. Results: In Tennessee between 2018 and 2021, 342 (17.2%) of the 1,980 CLABSI events had race documented, and no ethnicity variables exist in the NHSN. The data linkage process yielded a 72% match (1,426 CLABSIs). The remaining 28% were excluded from the analysis. Per 1,000 central-line days (CL days) for all races, white patients had the highest CLABSI rate (17.5), followed by Black patients (1.36), and Native American or Alaskan Native patients (0.68). Per 1,000 admissions by race, Black patients had a higher CLABSI rate (1.26) than Native American/Alaskan Native patients (0.85) and white patients (0.75). The risk of contracting a CLABSI was 79% higher in Black patients than in white patients (RR, 1.79; 95% CI, 1.55–2.07; P < .0001) when controlling for CCI, age group, and SVI. Conclusions: These results suggest that racial disparities between Black and white patients are present in Tennessee hospitals regarding CLABSIs. Although most CLABSI events were linked to HDDS patients, there were limitations in the ability to match all cases and calculate CL days by race. This study highlights the need for complete race and ethnicity data in the NHSN. Further studies should examine infection types at the regional and facility levels to target interventions for reducing HAI inequities in Tennessee.

Disclosures: None

Background: The Canadian Nosocomial Infection Surveillance Program (CNISP) observed increased mortality among neonatal intensive care unit (NICU) patients with central-line–associated bloodstream infection (CLABSI) starting in 2017. In this study, we compared NICU patients with CLABSIs before and after 2017, and quantified the impact of epidemiological factors on 30-day survival. Methods: We included 1,276 NICU patients from 8–16 participating CNISP hospitals from the pre-2017 period (2009–2016) and the post-2017 period (2017–2022) using standardized definitions and questionnaires. We used Cox regression modeling to assess the impact of age at date of positive culture, sex, birthweight, CLABSI microorganism, region of the country, and surveillance period (before 2017 vs after 2017) on time to 30-day all-cause mortality from date of positive culture. Gestational age was not available for this analysis. We reported model outputs as hazard ratios with 95% CIs. Results: In total, 769 (60%) NICU CLABSIs were reported in the pre-2017 period and 507 (40%) in the post-2017 period. The 30-day all-cause mortality rate was 8% (n = 100 of 1,276) overall, and significantly higher after 2017 (12%, n = 61 of 507) than before 2017 (5%, n = 39 of 769) (P < .001).

During the post-2017 period, cases were significantly younger: 16 days (IQR, 9–33) versus 21 days (IQR, 11–49) (P = .002). Median days from ICU admission to infection were shorter: 14 (IQR, 8–31) versus 19 (IQR, 10–41) (P < .001). More gram-negative CLABSIs were identified (29% vs 24%; P = .040) and fewer gram-positive CLABSIs were identified (64% vs 72%; P = .006) compared to the pre-2017 period. Mortality was higher in CLABSIs caused by gram-negative bacteria (15%, n = 50 of 328) than gram-positive bacteria (4.4%, n = 39 of 877) (P < .001), and mortality was higher in neonates with birthweight <1,000 g (11%, n = 71 of 673) compared to those weighing ≥1,000 g (5%, n = 28 of 560) (P < .001).

Adjusting for all other factors, survival modeling indicated that NICU CLABSIs identified in the post-2017 period had 2.12 (95% CI, 1.23–3.66) times the hazard ratio of 30-day all-cause mortality compared to those before 2017 (P < .006). Those identified with a gram-positive bacterium had a 0.28 hazard ratio (95% CI, 0.12–0.65) of 30-day mortality compared to those with a gram-negative bacterium or fungus (P = .003). In the fully adjusted model, age, sex, and birthweight were not significantly associated with NICU CLABSI survival. Conclusions: NICU patients with CLABSIs had significantly higher all-cause mortality between 2017–2022 compared to 2009–2016, and those who acquired gram-positive–associated CLABSIs had improved survival compared to other organisms. Further work is needed to identify and understand factors driving the increased mortality among NICU CLABSI patients from 2017–2022.

Disclosures: None