Background: The COVID-19 pandemic has disproportionately affected nursing home residents, and emerging evidence suggests quality, location, resident demographics, and staffing levels may be related to COVID-19 incidence within facilities. We describe the distribution of COVID-19 cases in Wisconsin nursing homes from January 2020 to October 2020, the effect of rural versus urban locations on COVID-19 incidence, and the temporal changes in COVID-19 incidence. Methods: We constructed a database using the Center for Medicaid and Medicare Services’ (CMS) publicly available data. Variables obtained per facility included location, number of beds, ownership type, average census, 5-star ratings (overall, quality, health, staffing, and nurse staffing categories), number of COVID-19 cases, resident Medicaid/Medicare share, area deprivation index, and social vulnerability index. Nursing homes were divided into tertiles based on total COVID-19 cases for descriptive analysis (zero cases, 1–7 cases, >7 cases). Demographic and clinical variables were reported as frequencies, mean (standard deviation) or median (interquartile range). We compared groups using the Pearson χ2 test and the Kruskal-Wallis test. COVID-19 incidence rates were calculated by dividing the number of COVID-19 cases by monthly occupied bed days and multiplied by 10,000. Results: From January 1, 2020, to November 1, 2020, in total, 3,133 SARS-CoV-2–confirmed cases were reported among 248 (70.5%) nursing homes. Urban location (P = .027), overall 5-star rating (P = .035), number of beds (p < 0.001), and average count of residents per day (p < 0.001) were associated with a greater number of COVID-19 cases. Temporal analysis showed that the highest incidence rates of COVID-19 in NHs were observed from January to May and in October 2020 (11.36 and 30.33 cases per 10,000 occupied-bed days, respectively). Urban NHs experienced higher incidence rates until September, then incidence rates among rural facilities surged (Fig.1A). In the first half of the year, NHs with lower quality scores (1-3 stars) had a higher COVID-19 incidence rate; however, in August this trend reversed, and facilities with higher quality scores (4-5 stars) showed the highest incidence rates (Fig.1B). Fig. 2 shows a temporal depiction of the shift from urban to rural settings. Conclusions: Higher COVID-19 incidence rates during the first 5 months of the pandemic were observed in urban, larger facilities with lower 5-star rating. By the end of the year, nursing homes in rural areas and those with higher quality ratings had the highest incidence rates.

Funding: No

Disclosures: None

Figure 1.

Figure 2.

Background: Healthcare-associated transmission of influenza leads to significant morbidity, mortality, and cost. Most studies classify healthcare-associated viral respiratory infections (HA-VRI) as those with a positive test result after the first 3 days following admission, which does not account for healthcare exposures prior to admission. Utilizing an expanded definition of healthcare-associated influenza, we aimed to improve the estimates of disease prevalence on a population level. Methods: This study included laboratory-confirmed cases of influenza in adult and pediatric patients admitted to any acute-care hospital in a catchment area of 8 counties Tennessee identified between October 1, 2012, and April 30, 2019. Surveillance information was abstracted from hospital and state laboratory databases, hospital infection control practitioner databases, reportable condition databases, and electronic health records as a part of the Influenza Hospitalization Surveillance Network (FluSurv-NET) by the Centers for Disease Control and Prevention (CDC) Emerging Infections Program (EIP). Cases were defined as healthcare-associated influenza laboratory confirmation of infection occurred (1) on or after hospital day 4 (“traditional definition”), or (2) between hospital days 0 and 3 in patients transferred from a chronic care facility or with a recent discharge from another acute-care facility in the 7 days preceding the current index admission (ie, enhanced definition). The proportion of laboratory-confirmed influenza designated as HA-VRI using both the traditional definition as well as with the added enhanced definition were compared. Data were imported into Stata software for analysis. Results: We identified 5,904 cases of laboratory-confirmed influenza in hospitalized patients over the study period. Using the traditional definition for HA-VRI, only 147 (2.5%, seasonal range 1.3%–3.4%) were deemed healthcare associated (Figure 1). Adding the cases identified using the enhanced definition, an additional 317 (5.4%, range 2.3%–6.7%) cases were noted in patients transferred from a chronic care facility for the current acute-care admission and 336 cases (5.7%; range, 4.1%–7.4%) were noted in patients with a prior acute-care facility admission in the preceding 7 days. Using our expanded definition, the total proportion of healthcare-associated influenza in this cohort was 772 of 5,904 (13.1%; range, 10.6%–14.8%). Conclusion: HA-VRI due to influenza is an underrecognized infection in hospitalized patients. Limiting surveillance assessment of this important outcome to just those patients with a positive influenza test after hospital day 3 captured only 19% of possible healthcare-associated influenza infections across 7 influenza seasons. These results suggest that the traditionally used definitions of healthcare-associated influenza underestimate the true burden of cases.

Funding: No

Disclosures: None

Figure 1.

In a survey of adult hospital providers regarding antibiotic use in the treatment of febrile neutropenia, clinical fellows, and pharmacists showed higher comfort levels with early antimicrobial de-escalation compared to hematology-oncology and transplant infectious diseases physicians. These frontline team members are ideal partners to champion antimicrobial stewardship interventions in febrile neutropenia.

Background: Antibiotics are frequently prescribed in nursing homes, often inappropriately. Data sources are needed to facilitate measurement and reporting of antibiotic use to inform antibiotic stewardship efforts. Previous analyses have shown that the type of nursing-home stay, that is, short stay (<100 days), is a strong predictor of high antibiotic use compared to longer nursing-home stays. The study objective was to compare 2 different data sources, electronic health record (EHR) and long-term care (LTC) pharmacy data, for surveillance of antibiotic use and type of nursing-home stay. Methods: EHR and pharmacy data during 2017 were included from 1,933 and 1,348 US-based nursing homes, respectively. We compared data elements available in each data source for antibiotic use reporting. In each data set, we attempted to describe antibiotic use as the proportion of residents on an antibiotic, days-of-therapy (DOT) per 1,000 resident days (RD), and distribution of antibiotic course duration, overall and at the facility level. Facility proportion of short-stay and long-stay (>100 days) nursing-home residents were calculated using admission dates and census data in the EHR data set and a payor variable in the pharmacy data set (Figure 1). The 2 data sources also provided antibiotic characteristics, including antibiotic class, agent, and route of administration. The deidentified nature of facility data prevented direct comparison of antibiotic use measures between facilities. Results: The EHR and pharmacy data sets contained 381,382 and 326,713 residents, respectively (Table 1). Within the EHR, 51% of residents were prescribed an antibiotic in 2017, at a median rate of 77 DOT per 1,000 RD. In the LTC pharmacy, 46% of residents were prescribed an antibiotic at a median rate of 79 DOT per 1,000 RD (Table 1). Short-stay residents contributed a smaller proportion of total RDs in the EHR relative to the pharmacy cohort (21% vs 50%, respectively). Conclusions: Nursing-home antibiotic use data obtained from EHR and pharmacy vendors can be used for calculating antibiotic use measures, which is important for antibiotic use reporting and facility-level tracking to identify opportunities for improving prescribing practices and provide facility-level benchmarks. Further validation of both data sources in the same facilities is needed to compare antibiotic use rates and to determine the most appropriate proxy for type of nursing-home stay for facility-level risk adjustment of antibiotic use rates.

Funding: No

Disclosures: None

Figure 1.

Table 1.

Background: Evidence-based hospital antimicrobial stewardship interventions, such as postprescription review with feedback, prior authorization, and handshake stewardship, involve communication between stewards and frontline prescribers. Hierarchy, asymmetric responsibility, prescribing etiquette, and autonomy can obstruct high-quality communication in stewardship. Little is known about the strategies that stewards use to overcome these barriers. The objective of this study was to identify how stewards navigate communication challenges when interacting with prescribers. Methods: We conducted semistructured interviews with antimicrobial stewards recruited from hospitals across the United States. Interviews were audio recorded, transcribed, and analyzed using a flexible coding approach and the framework method. Social identity theory and role theory were used to interpret framework matrices. Results: Interviews were conducted with 58 antimicrobial stewards (25 physicians and 33 pharmacists) from 10 hospitals (4 academic medical centers, 4 community hospitals, and 2 children’s hospitals). Respondents who felt empowered in their interactions with prescribers explicitly adopted a social identity that conceptualized stewards and prescribers as being on the “same team” with shared goals (in-group orientation). Drawing on the meaning conferred via this social role identity, respondents engaged in communication strategies to build and maintain common bonds with prescribers. These strategies included moderating language to minimize defensive recommendations when delivering stewardship recommendations, aligning the goals of stewardship with the goals of the clinical team, communicating with prescribers about things other than stewardship, compromising for the sake of future interactions, and engaging in strategic face-to-face interaction. Respondents who felt less empowered in their interactions thought of themselves as outsiders to the clinical team and experienced a heightened sense of “us versus them” mentality with the perception that stewards primarily serve a gate-keeping function (ie, outgroup orientation). These respondents expressed deference to hierarchy, a reluctance to engage in face-to-face interaction, a feeling of cynicism about the impact of stewardship, and a sense of low professional accomplishment within the role. Respondents who exhibited an in-group orientation were more likely than those who did not to describe the positive impact of stewardship mentors or colleagues on their social role identity. Conclusions: The way antimicrobial stewards perceive their role and identity within the social context of their healthcare organization influences how they approach communication with prescribers. Social role identity in stewardship is shaped by the influence of mentors and colleagues, indicating the importance of supportive relationships for the development of steward skill and confidence.

Funding: No

Disclosures: None

Background: Urine cultures are the most common microbiological tests in the outpatient setting and heavily influence treatment of suspected urinary tract infections (UTIs). Antibiotics for UTI are usually prescribed on an empiric basis in primary care before the urine culture results are available. However, culture results may be needed to confirm a UTI diagnosis and to verify that the correct antibiotic was prescribed. Although urine cultures are considered as the gold standard for diagnosis of UTI, cultures can easily become contaminated during collection. We determined the prevalence, predictors, and antibiotic use associated with contaminated urine cultures in 2 adult safety net primary care clinics. Methods: We conducted a retrospective chart review of visits with provider-suspected UTI in which a urine culture was ordered (November 2018–March 2020). Patient demographics, culture results, and prescription orders were captured for each visit. Culture results were defined as no culture growth, contaminated (ie, mixed flora, non-uropathogens, or ≥3 bacteria isolated on culture), low-count positive (growth between 100 and 100,000 CFU/mL), and high-count positive (>100,000 CFU/mL). A multivariable multinomial logistic regression model was used to identify factors associated with contaminated culture results. Results: There were 1,265 visits with urine cultures: 264 (20.9%) had no growth, 694 (54.9%) were contaminated, 159 (12.6%) were low counts, and 148 (11.7%) were high counts. Encounter-level factors are presented in Table 1. Female gender (adjusted odds ratio [aOR], 15.8; 95% confidence interval [CI], 10.21–23.46; P < .001), pregnancy (aOR, 13.98; 95% CI, 7.93–4.67; P < .001), and obesity (aOR, 1.9; 95% CI 1.31–2.77; P < .001) were independently associated with contaminated cultures. Of 264 patients whose urine cultures showed no growth, 36 (14%) were prescribed an antibiotic. Of 694 patients with contaminated cultures, 153 (22%) were prescribed an antibiotic (Figure 1). Conclusions: More than half of urine cultures were contaminated, and 1 in 5 patients were treated with antibiotics. Reduction of contamination should improve patient care by providing a more accurate record of the organism in the urine (if any) and its susceptibilities, which are relevant to managing future episodes of UTI in that patient. Optimizing urine collection represents a diagnostic stewardship opportunity in primary care.

Funding: This study was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (grant no. UM1AI104681). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Disclosures: None

Table 1.

Figure 1.

Training programs for infectious diseases fellows pursuing a career in infection prevention and control and hospital epidemiology are grounded in mentorship and organizational experience. In this commentary, we propose a competency-based framework for creating structured learning for infectious diseases fellows pursuing hospital epidemiology and related fields.

Efforts to improve antimicrobial prescribing are occurring within a changing healthcare landscape, which includes the expanded use of telehealth technology. The wider adoption of telehealth presents both challenges and opportunities for promoting antimicrobial stewardship. Telehealth provides 2 avenues for remote infectious disease (ID) specialists to improve inpatient antimicrobial prescribing: telehealth-supported antimicrobial stewardship and tele-ID consultations. Those 2 activities can work separately or synergistically. Studies on telehealth-supported antimicrobial stewardship have reported a reduction in inpatient antimicrobial prescribing, cost savings related to less antimicrobial use, a decrease in Clostridioides difficile infections, and improved antimicrobial susceptibility patterns for common organisms. Tele-ID consultation is associated with fewer hospital transfers, a shorter length of hospital stay, and decreased mortality. The implementation of these activities can be flexible depending on local needs and available resources, but several barriers may be encountered. Opportunities also exist to improve antimicrobial use in outpatient settings. Telehealth provides a more rapid mechanism for conducting outpatient ID consultations, and increasing use of telehealth for routine and urgent outpatient visits present new challenges for antimicrobial stewardship. In primary care, urgent care, and emergency care settings, unnecessary antimicrobial use for viral acute respiratory tract infections is common during telehealth encounters, as is the case for fact-to-face encounters. For some diagnoses, such as otitis media and pharyngitis, antimicrobials are further overprescribed via telehealth. Evidence is still lacking on the optimal stewardship strategies to improve antimicrobial prescribing during telehealth encounters in ambulatory care, but conventional outpatient stewardship strategies are likely transferable. Further work is warranted to fill this knowledge gap.

Background: Adherence to core elements of antimicrobial stewardship programs (ASPs) is increasing nationally but the robustness of programs and inclusion of pediatrics is poorly understood. We describe the details of ASP in Colorado hospitals and identify steps by which academic centers and public health departments can assist community ASPs. Methods: We invited ASP leaders at the 102 acute-care hospitals (ACHs) and critical-access hospitals (CAHs) in Colorado to participate in a web-based survey regarding their ASPs. Questions related to adherence to Centers for Disease Control and Prevention (CDC) core elements, barriers to improvement, desired resources, and extension to pediatrics. Enrollment began in August 2020. Hospital types were compared using the Fisher exact test. Results: As of January 1, 2021, 31 hospitals (30% of targeted hospitals) completed the web-based survey including 19 ACH and 12 CAH. Hospitals were distributed across the state. Median number of beds was 52 (range, 11–680). Of the responding hospitals, 87% were adherent to all CDC core elements. However, if action was defined as prospective audit and feedback or prior authorization, tracking was defined as measuring antibiotic use in days of therapy (DOT) or defined daily dose (DDD) quarterly, and reporting was defined as providing unit- or provider-specific antibiotic use reports annually. Overall adherence fell to 35% including 81% for action, 58% for tracking, and 58% for reporting. CAHs were less likely to adhere to these strict criteria than ACHs (Figure 1). In the 27 hospitals (87% of hospitals) caring for pediatric patients, adherence to a strict action for at least 1 pediatric population was 59%. Reported barriers to improved ASP were available time and personnel, information technology support, perceived concerns about provider attitudes, and education gaps (Figure 2). CAHs were less likely to use the NHSN antibiotic use or resistance modules or have a data analyst than ACHs (Figure 3). Pediatric pharmacy expertise and guidelines were often not available in hospitals caring for pediatric patients. Desired ASP resources included assistance with data analysis, access to stewardship expertise and education, and treatment guidelines, including for pediatrics. Conclusions: Adherence to CDC core elements of an ASP was excellent but fell dramatically when stricter criteria were used and was worse in pediatric patients. Academic centers and public health departments can assist community hospitals by providing educational resources, assistance in analyzing data including using the NHSN ED: /AR modules, and ASP expertise and clinical care guidelines including those for pediatrics.

Funding: No

Disclosures: None

Figure 1.

Figure 2.

Figure 3.

Background: The Centers for Disease Control and Prevention (CDC) estimates that outpatient settings account for 85%–90% of antibiotic prescriptions in the United States, and ~30% of those prescriptions are unnecessary. One of the most common examples of inappropriate prescribing is for viral upper respiratory infections (URIs). Up to 50% of prescriptions written for URIs are deemed inappropriate, making it an important focus for Antibiotic Stewardship programs. In this study, we evaluated the effect of a behaviorally enhanced quality improvement intervention in reducing inappropriate antibiotic prescribing for viral URIs. Methods: A quasi-experimental study assessed the effects of an Antibiotic Stewardship intervention on antibiotic prescribing for viral URIs. The outcome of interest was a change in the number of antibiotics prescribed at each participating clinic over a 1-year preimplementation period and a 2-year postimplementation period. Time trends were analyzed using segmented regression analysis, and a stepped wedge design was used to account for intervention roll-out across clinics. Results: From 2017 to 2020, there were 63,028 patient visits in 21 clinic locations. Antibiotics were prescribed an average of 11.5% and 5.8% of visits during the pre- and postimplementation periods, respectively. The most frequently prescribed antibiotic over the study period was azithromycin (n = 3,551), followed by amoxicillin (n = 924). Overall, the intervention was associated with a 46% reduction in antibiotic prescriptions or 0.54 times (P = .001) as many inappropriate antibiotics prescribed as before the intervention. There was no significant change in the month-to-month trend in inappropriate prescriptions after the intervention was implemented (P = .87). Conclusions: Our study demonstrates that a behaviorally enhanced quality improvement intervention to reduce inappropriate prescribing for URI in ambulatory care encounters was successful in reducing potentially inappropriate prescriptions for presumed viral respiratory infections.

Funding: No

Disclosures: None

Background: Bats are recognized as important vectors in disease transmission. Frequently, bats intrude into homes and buildings, increasing the risk to human health. We describe bat intrusions and exposure incidents in our hospital over a 3-year period. Methods: The University of Iowa Hospitals and Clinics (UIHC) is an 811-bed academic medical center in Iowa City, Iowa. Established in 1928, UIHC currently covers 209,031.84 m2 (~2,250,000 ft2) and contains 6 pavilions built between 1928 and 2017. We retrospectively obtained bat intrusion calls from the infection prevention and control program call database at UIHC during 2018–2020. We have also described the event management for intrusions potentially associated with patient exposures. Results: In total, 67 bat intrusions occurred during 2018–2020. The most frequent locations were hallways or lounges 28 (42%), nonclinical office spaces 19 (14%), and stairwells 8 (12%). Most bat intrusions (65%) occurred during the summer and fall (June–November). The number of events were 15 in 2018, 28 in 2019, and 24 in 2020. We observed that the number of intrusions increased with the age of each pavilion (Figure 1). Of 67 intrusions, 2 incidents (3%) were associated with potential exposure to patients. In the first incident, reported in 2019, the bat was captured in a patient care area and released before an investigation of exposures was completed and no rabies testing was available. Also, 10 patients were identified as having had potential exposure to the bat. Among them, 9 patients (90%) received rabies postexposure prophylaxis. In response to this serious event, we provided facility-wide education on our bat control policy, which includes the capture and safe handling of the bat, assessment of potential exposures, and potential need for rabies testing. We also implemented a bat exclusion project focused on the exterior of the oldest hospital buildings. The second event, 1 patient was identified to have potential exposure to the bat. The bat was captured, tested negative for rabies, no further action was needed. Conclusions: Bat intrusions can be an infection prevention and control challenge in facilities with older buildings. Hospitals may need animal intrusion surveillance systems, management protocols, and remediation efforts.

Funding: No

Disclosures: None

Figure 1.

Artificial intelligence (AI) refers to the performance of tasks by machines ordinarily associated with human intelligence. Machine learning (ML) is a subtype of AI; it refers to the ability of computers to draw conclusions (ie, learn) from data without being directly programmed. ML builds from traditional statistical methods and has drawn significant interest in healthcare epidemiology due to its potential for improving disease prediction and patient care. This review provides an overview of ML in healthcare epidemiology and practical examples of ML tools used to support healthcare decision making at 4 stages of hospital-based care: triage, diagnosis, treatment, and discharge. Examples include model-building efforts to assist emergency department triage, predicting time before septic shock onset, detecting community-acquired pneumonia, and classifying COVID-19 disposition risk level. Increasing availability and quality of electronic health record (EHR) data as well as computing power provides opportunities for ML to increase patient safety, improve the efficiency of clinical management, and reduce healthcare costs.

Background: Medical University of South Carolina Health began vaccinating care team members December 15, 2020, with the Pfizer-BioNTech SARS-CoV-2 mRNA vaccine. We report surveillance data for cases diagnosed following vaccination. Methods: Care team members (CTMs) diagnosed with COVID-19 following SARS-CoV-2 vaccination were self-identified during online electronic contact-tracing surveys. Demographic data, symptoms, and dates of symptoms were recorded. CTMs testing positive at MUSC were linked to viral burden data from nasopharyngeal swabs tested on Abbott PCR platforms. Results: As of January 31, 2021, 111 CTMs tested positive for SARS-CoV-2 following vaccination: 99 and 12 after 1 and 2 doses, respectively, at medians of 10 days (range, 1–22) and 5 days (range, 1–16), respectively, from vaccination to testing. Of 2 cases that tested positive >14 days from dose 2, CTMs had symptom onset at 4 and 12 days from dose 2. Among CTMs reporting symptoms, 104 did so after a median of 7 days (mean 6.3, range −23 to +22) from vaccination to symptom onset, with 8 reporting symptoms before vaccination, 9 on the day of vaccination, and 3 CTMs at 1 day after vaccination, 6 CTMs at 2 days after vaccination, and 11 CTMs at 3 days after vaccination. Overall, 86 (78%) of 111 were female and 95 (86%) of 111 were white. The median age was 44 years (range, 22–71). Clinical job roles were most frequently nurses (n = 31, 28%), physicians or physician extenders (n = 19, 17%), and CTMs with no patient contact (n = 21, 19%). Assessment by the contact-tracing team assigned sources as household clusters (n = 22, 23%), local transmission (n = 21, 22%), occupational acquisition from coworkers (n = 11, 12%), travel related (n = 9, 10%), and unknown (n = 32, 34%). Only 32 (32%) CTMs were compliant with physical distancing. Among 104 CTMs reporting symptoms, cough (75%), headache (71%), rhinorrhea (63%), myalgia (60%), sore throat (48%), anosmia (44%), and subjective fever (40%) were the most commonly reported symptoms. Among 54 symptomatic CTMs with available viral-load data, the median and mean cycle numbers (Cn) were 19.98 and 21.91, respectively, for samples tested a median 3 days from symptom onset. Asymptomatic and symptomatic CTMs had a median Cn of 30.1 vs 20.9, respectively (p <0.001) and overall 50% of vaccinated CTMs had Cn >20, with no significant effect seen by vaccine dose (Figure 1). Conclusions: Most COVID-19 cases following vaccination occurred in CTMs with infection incubating prior to vaccination. No significant attenuation of viral load is apparent among vaccinated CTMs with COVID-19, but asymptomatic CTMs diagnosed with COVID-19 following vaccination appear to have resolved infections. Our data reinforce the need to adhere to public health measures by people who have been vaccinated.

Funding: No

Disclosures: None

Figure 1.

Emergency preparedness programs have evolved over the last several decades as communities have responded to natural, intentional, and accidental disasters. This evolution has resulted in a comprehensive all-hazards approach centered around 4 fundamental phases spanning the entire disaster life cycle: mitigation, preparedness, response, and recovery. Increasing frequency of outbreaks and epidemics of emerging and reemerging infectious diseases in the last decade has emphasized the significance of healthcare emergency preparedness programs, but the coronavirus disease 2019 (COVID-19) pandemic has tested healthcare facilities’ emergency plans and exposed vulnerabilities in healthcare emergency preparedness on a scale unexperienced in recent history. We review the 4 phases of emergency management and explore the lessons to be learned from recent events in enhancing health systems capabilities and capacities to mitigate, prepare for, respond to, and recover from biological threats or events, whether it be a pandemic or a single case of an unknown infectious disease. A recurring cycle of assessing, planning, training, exercising, and revising is vital to maintaining healthcare system preparedness, even in absence of an immediate, high probability threat. Healthcare epidemiologists and infection preventionists must play a pivotal role in incorporating lessons learned from the pandemic into emergency preparedness programs and building more robust preparedness plans.

Contact precautions are used to prevent the spread of extended-spectrum β-lactamase (ESBL)–producing organisms in acute-care hospitals, but supporting data are lacking. We discontinued such precautions for ESBL Escherichia coli and Klebsiella spp and found no increased prevalence of these organisms with our change in practice.

Background: The disease caused by SARS-CoV-2, COVID-19, has caused a pandemic leading to strained healthcare systems worldwide and an unprecedented public health crisis. The hallmark of severe COVID-19 is lower respiratory tract infection (LRTI) and hypoxia requiring hospitalization. A paucity of data on bacterial coinfection and a lack of therapeutic options for COVID-19 during the first surge of cases has increased pressure on antimicrobial use and has challenged antimicrobial stewardship programs (ASPs). We implemented a multimodal approach to antimicrobial stewardship in an urban safety-net community hospital targeting selection and duration of therapy. Methods: Retrospective review of cases during the first wave of COVID-19 in a 151-bed urban safety-net community hospital from March to June 2020. EMR order sets (Figure 1) and prospective audit and feedback by ASPs targeting empiric antimicrobial selection and duration were implemented as part of the COVID-19 response. Hospitalized patients with COVID-19 were reviewed retrospectively. Demographic information was collected. Data on antimicrobial use were tabulated, including selection and duration of antimicrobials (Figure 1). Results: In total, 302 patients were reviewed, of whom 221 (73%) received empiric antimicrobials. The most commonly used antimicrobials were ceftriaxone and azithromycin (Figure 2). Days of therapy per 1,000 patient days (DOT/1,000 PD) for ceftriaxone increased from 71 in the quarter prior to 113 during the study period. Average duration of therapy was 6 days. In the ICU, average duration was 8 days compared to 5 days in non-ICU settings. Average durations of parenteral therapy were 5.54 days in the ICU and 3.36 days in non-ICU settings. Procalcitonin was obtained in 37 cases (17%) and ranged from 0.09 to 12.57 ng/mL, with an average of 1.21 ng/mL. No cases had documented bacterial coinfection (Figure 1). Conclusions: Antimicrobials were commonly prescribed during the first wave of COVID-19 in a safety-net community hospital. Procalcitonin did not guide therapy nor did lack of documented coinfection change physician behavior. With limited resources, ASP successfully guided clinicians toward IDSA guideline recommendations for selection and duration, as evidenced by antimicrobial use. During this unprecedented surge of LRTIs, a multimodal approach to antimicrobial stewardship was used and guided toward early transition to oral agents and shorter durations.

Funding: No

Disclosures: None

Figure 1.

Figure 2.