Background: According to the WHO, hand hygiene is the primary measure to reduce infections. It is a simple act, but the lack of compliance among healthcare workers has been a great concern for all healthcare facilities. Healthcare facilities can perform a situation analysis of hand hygiene promotion and practices according to a set of indicators designed by the WHO in the form of a hand hygiene self-assessment framework. Results can be used to identify areas of improvement and to develop an action plan and strategies accordingly. Low- or middle-income country (LMIC) initial scoring within this framework was 195 points (ie basic level); thus, we aimed to achieve the advanced level, with a score of > 375. Methods: The WHO hand hygiene self-assessment framework is a diagnostic tool to identify key issues requiring attention and improvement. Repeated assessments are done to document the progress over time, which allows a health-care facility to track their progress in hand hygiene resources, to conduct promotion activities, to plan their actions, and to achieve improvement and sustainability. We developed an action plan under each category of WHO framework that included: system change, training and education, evaluation and feedback, reminders in workplace, and institutional safety climate for hand hygiene. We implemented the following measures: point-of-care hand hygiene stations were made available at all bedsides; mandatory training was introduced for all healthcare workers, and consumption of hand rub or hand sanitizers and liquid soap was monitored as a consumption indicator. In addition, posters were placed in all wards and clinics, time was dedicated for HH promotion, and a May 5th plan was implemented. HH leaders, role models, and champions were identified from each discipline. Patients were involved in HH promotion; HH leaflets were given to patients, HH e-learning tools were implemented, and a system for personal accountability was initiated, as well as a buddy system for new employees. Results: After implementation of multiple strategies in each section of the WHO self-assessment framework, we our overall score increased from basic (ie, 195) to advanced (ie, 395). In addition, category score increased in system change from 60 to 85, in training and education from 35 to 100, in evaluation and feedback from 52.5 to 100, in reminders in workplace from 17.5 to 45, and in institutional safety climate from 30 to 65. Conclusions: The WHO hand hygiene self-assessment framework should be utilized by all the hospitals in LMICs as a guide to improve hand hygiene levels.

Funding: None

Disclosures: None

Background: Peritoneal dialysis is a type of dialysis performed by patients in their homes; patients receive training from dialysis clinic staff. Peritonitis is a serious complication of peritoneal dialysis, most commonly caused by gram-positive organisms. During March‒April 2019, a dialysis provider organization transitioned ~400 patients to a different manufacturer of peritoneal dialysis equipment and supplies (from product A to B). Shortly thereafter, patients experienced an increase in peritonitis episodes, caused predominantly by gram-negative organisms. In May 2019, we initiated an investigation to determine the source. Methods: We conducted case finding, reviewed medical records, observed peritoneal dialysis procedures and trainings, and performed patient home visits and interviews. A 1:1 matched case–control study was performed in 1 state. A case had ≥2 of the following: (1) positive peritoneal fluid culture, (2) high peritoneal fluid white cell count with ≥50% polymorphonuclear cells, or (3) cloudy peritoneal fluid and/or abdominal pain. Controls were matched to cases by week of clinic visit. Conditional logistic regression was used to estimate univariate matched odds ratios (mOR) and 95% confidence intervals (CIs). We conducted microbiological testing of peritoneal dialysis fluid bags to rule out product contamination. Results: During March‒September 2019, we identified 157 cases of peritonitis across 15 clinics in 2 states (attack rate≍39%). Staphylococcus spp (14%), Serratia spp (12%) and Klebsiella spp (6.3%) were the most common pathogens. Steps to perform peritoneal dialysis using product B differed from product A in several key areas; however, no common errors in practice were identified to explain the outbreak. Patient training on transitioning products was not standardized. Outcomes of the 73 cases in the case–control study included hospitalization (77%), peritoneal dialysis failure (40%), and death (7%). The median duration of training prior to product transition was 1 day for cases and controls (P = .86). Transitioning to product B (mOR, 18.00; 95% CI, 2.40‒134.83), using product B (mOR, 18.26; 95% CI, 3.86‒∞), drain-line reuse (mOR, 4.67; 95% CI, 1.34‒16.24) and performing daytime exchanges (mOR, 3.63; 95% CI, 1.71‒8.45) were associated with peritonitis. After several interventions, including transition of patients back to product A (Fig. 1), overall cases declined. Sterility testing of samples from 23 unopened product B peritoneal dialysis solution bags showed no contamination. Conclusions: Multiple factors may have contributed to this large outbreak, including a rapid transition in peritoneal dialysis products and potentially inadequate patient training. Efforts are needed to identify and incorporate best training practices, and product advances are desired to improve the safety of patient transitions between different types of peritoneal dialysis equipment.

Funding: None

Disclosures: None

Background:Pseudomonas aeruginosa is an important nosocomial pathogen associated with intrinsic and acquired resistance mechanisms to major classes of antibiotics. To better understand clinical risk factors for drug-resistant P. aeruginosa infection, decision-tree models for the prediction of fluoroquinolone and carbapenem-resistant P. aeruginosa were constructed and compared to multivariable logistic regression models using performance characteristics. Methods: In total, 5,636 patients admitted to 4 hospitals within a New York City healthcare system from 2010 to 2016 with blood, respiratory, wound, or urine cultures growing PA were included in the analysis. Presence or absence of drug-resistance was defined using the first culture of any source positive for P. aeruginosa during each hospitalization. To train and validate the prediction models, cases were randomly split (60 of 40) into training and validation datasets. Clinical decision-tree models for both fluoroquinolone and carbapenem resistance were built from the training dataset using 21 clinical variables of interest, and multivariable logistic regression models were built using the 16 clinical variables associated with resistance in bivariate analyses. Decision-tree models were optimized using K-fold cross validation, and performance characteristics between the 4 models were compared. Results: From 2010 through 2016, prevalence of fluoroquinolone and carbapenem resistance was 32% and 18%, respectively. For fluoroquinolone resistance, the logistic regression algorithm attained a positive predictive value (PPV) of 0.57 and a negative predictive value (NPV) of 0.73 (sensitivity, 0.27; specificity, 0.90) and the decision-tree algorithm attained a PPV of 0.65 and an NPV of 0.72 (sensitivity 0.21, specificity 0.95). For carbapenem resistance, the logistic regression algorithm attained a PPV of 0.53 and a NPV of 0.85 (sensitivity 0.20, specificity 0.96) and the decision-tree algorithm attained a PPV of 0.59 and an NPV of 0.84 (sensitivity 0.22, specificity 0.96). The decision-tree partitioning algorithm identified prior fluoroquinolone resistance, SNF stay, sex, and length-of-stay as variables of greatest importance for fluoroquinolone resistance compared to prior carbapenem resistance, age, and length-of-stay for carbapenem resistance. The highest-performing decision tree for fluoroquinolone resistance is illustrated in Fig. 1. Conclusions: Supervised machine-learning techniques may facilitate prediction of P. aeruginosa resistance and risk factors driving resistance patterns in hospitalized patients. Such techniques may be applied to readily available clinical information from hospital electronic health records to aid with clinical decision making.

Funding: None

Disclosures: None

Background: Australia has ~2,700 aged-care homes and 180 multipurpose services. The annual Aged Care National Antimicrobial Prescribing Survey (AC NAPS), first pilot tested in 2015, is a surveillance tool that can be used in these facilities to monitor infections and antimicrobial use. It assists in identifying priorities for local and national infection control and antimicrobial stewardship interventions. Methods: Nurses or pharmacists collect point prevalence data using standardized data collection forms: (1) A facility form, completed by each participating facility, includes resident-level data fields (eg, number of residents present on the survey day). (2) An infection form is completed for residents with signs and/or symptoms of infection. (3) An antimicrobial form is completed for residents who are prescribed an antimicrobial. Results: Regarding prevalence,for those 31 facilities that participated annually, there was no significant change in either prevalence rate (Table 1). Regarding priority areas for improvement (2018 data only), 64.6% of prescriptions were for residents who did not have signs and/or symptoms of a suspected infection in the week prior to the antimicrobial start date. The most common clinical indications for prescriptions were skin soft-tissue and mucosal infection (18.3%), cystitis (16.0%) and pneumonia (9.4%). Cefalexin (20.3%), clotrimazole (19.0%), and chloramphenicol (7.0%) were the most commonly prescribed antimicrobials. Review or stop dates were not documented for 58.9% of prescriptions. Only 39.2% of antimicrobials were prescribed in the 7 days prior to the survey day; 28.3% were prescribed >6 months prior. Furthermore, 36.3% of all prescriptions were for topical application. In addition, 19.0% of antimicrobials were prescribed for PRN (as needed) administration; most (94.4%) of these were for topical antimicrobials, most commonly clotrimazole (65.4%). Conclusions: The AC NAPS has identified infections and consistent patterns of antimicrobial use that may adversely affect the safety of care for Australian aged-care residents. Interventions are now being developed, implemented, and evaluated to address identified ‘priority areas for improvement.’

Funding: None

Disclosures: None

Background: Catheter-associated urinary tract infection (CAUTI) is considered a preventable healthcare-associated infection. Many local and national interventions using multimodal prevention measures have targeted CAUTI incidence as the primary outcome. Other undesirable events related to urinary catheters and infections such as overuse of urine culturing and antimicrobial prescribing for asymptomatic bacteriuria, are not captured by CAUTI surveillance, and may not be the targets of such interventions. The aim of this study was to assess the impact of expanded national surveillance targeting various aspects of urinary tract infections, culturing and treatment practices, and catheter use in internal medicine wards. Methods: The Israeli National Center for Infection Control (NCIC) issued CAUTI prevention guidelines and initiated in 2016 a urinary tract event surveillance system that targets the incidence of CAUTI, urinary catheter utilization ratio, and the proportion of urine cultures sent and patients treated in the absence of symptoms. The surveillance is conducted for 1 month 3 times per year. Hospitals are required to report all positive urine cultures (>100,000 CFU) collected in internal medicine wards, along with the following data: admission date, symptoms of infection, dates of urinary catheter use, and antibiotic treatment. These data enable the NCIC to validate hospital classifications of each event. In addition, during each surveillance month, hospitals conduct point-prevalence surveys of compliance with CAUTI prevention measures. An electronic data collection form with built-in algorithms supports the local teams during the surveillance process. Results: Between 2016 and 2019, a total of 3,028 positive urine cultures not present on admission were reported by internal medicine wards in 30 hospitals. A significant decrease was observed in the incidence of CAUTI (from 4.7 to 2.9; P < .001) and in the proportion of asymptomatic bacteriuria treated with antibiotics (from 31% to 20%; P = .02) (Table 1). The catheter utilization ratio decreased from 0.25 to 0.23 (P < .001). The rate of cultures sent from asymptomatic patients decreased from 1.5 to 1.1 (P < .01). Point-prevalence surveys in internal medicine wards detected a significant increase in the use of closed urinary drainage systems (from 79% to 97% in 2018, P < .001) and documentation of a daily nurse assessment of the need for a catheter (from 74% to 81%, P < .001). Conclusions: National surveillance of undesirable urinary tract events resulted in a significant reduction in CAUTI, antibiotic treatment for ASB, and the rate of cultures sent from asymptomatic patients. A small decrease was observed in catheter utilization ratio. CAUTI surveillance programs should include other undesirable urinary tract events.

Funding: None

Disclosures: None

Background: The genus Roseomonas, containing pink-pigmented glucose nonfermentative bacteria, has been associated with various primary and nosocomial human infections; however, to our knowledge, its nosocomial transmission has never been reported in the literature. Here, we report a nosocomial cluster of Roseomonas mucosa bacteremia. Methods: Two cases of R. mucosa bacteremia in 2018 are described. Clinical and epidemiological investigations were undertaken. Environmental surfaces prone to water contamination in the patient wards were sampled and cultured. The sampled surfaces included sinks, faucets, toilets, sewage, showerheads, refrigerators, exhaust vents, and washing machines. The 2 clinical isolates and all environmental isolates that showed growth of pink colonies were identified using matrix-assisted laser desorption/ionization time of flight mass spectrometry and 16S rRNA gene sequencing. Pulse-field gel electrophoresis (PFGE) was performed and fingerprinting software was used to analyze the DNA restriction patterns and determine their similarity. Results: Two patients who developed R. mucosa bacteremia had received care from the same treatment team. The patients were on different wards but had overlapping hospital stays. In addition to the treatment team, no other shared exposure was identified. Moreover, 126 environmental surfaces were sampled, of which 7 samples grew pink colonies. The 9 isolates from the patients and the environmental samples were examined using 16S rRNA gene sequencing. Overall, 7 isolates, including isolates from both patients, were identified as R. mucosa, and the other 2 isolates were identified as Roseomonas gilardii subsp. rosea (Fig. 1). With 80% similarity as a cutoff, PFGE analysis revealed that the R. mucosa isolates from 2 patients’ blood cultures and 3 environmental isolates (a washing machine in the ward, a sink in the shared washroom, and a sink in the patient room) belonged to the same clone (Fig. 2). Conclusions: The hospital water environment was contaminated with R. mucosa, and the same clone caused bacteremia in 2 separate patients, suggesting nosocomial transmission of R. mucosa possibly linked to contaminated water, environment, and/or patient care.

Funding: None

Disclosures: None

Background: In the setting of global warming, natural disasters are increasing in pace and scope. Although natural disasters themselves do not cause outbreaks, the breakdowns in sanitary infrastructure and the displacement of populations, often to crowded shelters, have caused outbreaks. On August 26, 2017, category 4 hurricane Harvey made landfall near Corpus Christi, Texas, causing catastrophic flooding and displacing >30,000 residents from the Southern Gulf Coast region. Dallas accepted >3,800 evacuees at the Kay Bailey Hutchison Convention Center mega-shelter for 23 days, where a medical clinic was erected in the convention center parking garage. The medical clinic uniquely included a dedicated infection prevention team composed of local volunteer infection preventionists, healthcare epidemiologists, infectious diseases providers, and health department personnel. Methods: Evacuees were housed at the Dallas mega-shelter from August 29 through September 20. The infection prevention team maintained a presence of 3–4 members during clinical operations in shifts. The team conducted an initial needs assessment upon opening of the shelter medical clinic, facilitated acquisition of adequate numbers of hand sanitizer stations, sinks with running water, portable hand-washing stations, portable toilets and showers, and cleaning products. The infection prevention team coordinated and oversaw environmental cleaning services (EVS) carried out by local hospital EVS staff. Protocols for cleaning, disinfection, communicable disease testing, isolation, and treatment were created. In addition, education and training materials for the implementation of these protocols were distributed to volunteer staff. The infection preventionists created and provided oversight of the designated isolation units for respiratory, gastrointestinal and dermatologic infections of outbreak potential. Infection prevention rounding tools were developed and executed daily in the clinic, at the on-site daycare center, dining area, and the general shelter dormitory. Vaccination for influenza was formalized under a protocol and administered at the clinic and via mobile vaccination teams in the chronic illness section of the dormitory. Results: In tota3,829 residents were housed at the mega-shelter for 23 days. Moreover, 1,560 patients were seen in 2,654 clinic visits at the shelter medical clinic. In total, 48 (19%) clinic visits were for respiratory symptoms, 228 (9%) were for dermatologic problems, and 215 (8%) were for gastrointestinal symptoms. Also, 32 patients were referred to the isolation unit within the clinic. Overall, 98 influenza vaccines were administered. There was 1 confirmed case of influenza and 1 confirmed case of norovirus. Conclusions: No known transmission of communicable diseases occurred in this long-term, natural disaster–related mega-shelter, likely attributed to having a comprehensive infection prevention team of on-site volunteers available throughout the shelter operation. This model should be considered in future large-scale shelter settings to prevent disease transmission.

Disclosures: None

Funding: None

Background: Twitter is used by officials to distribute public health messages and by the public to post information about ongoing afflictions. Because tweets originate from geographically and socially diverse sources, scholars have used this social media data to analyze the spread of diseases like flu [Alessio Signorini 2011], asthma [Philip Harber 2019] and mental health disorders [Chandler McClellan, 2017]. To our knowledge, no Twitter analysis has been performed for Valley fever. Valley fever is a fungal infection caused by the Coccidioides organism, mostly found in Arizona and California. Objective: We analyzed tweets concerning Valley fever to evaluate content, location, and timing. Methods: We collected tweets using the Twitter search application programming interface using the terms “Valley fever,” “valleyfever,” “cocci” or “‘Valleyfever” from August 6 to 16, 2019, and again from October 20 to 29, 2019. In total, 2,117 Tweets were retrieved. Tweets not focused on Valley fever were filtered out, including a tweet about “Rift valley fever” and tweets where “valley” and “fever” were separate and not one phrase. We excluded tweets not written in English. In total, 1,533 tweets remained; we grouped them into 3 categories: original tweets, hereafter labeled “normal” (N = 497), retweets (N = 811), and replies (N = 225). We converted all terms to lowercase, removed white space and punctuation, and tokenized the tweets. Informal messaging conventions (eg, hashtag, @user, RT, links) and stop words were removed, and terms were lemmatized. Finally, we analyzed the frequency of tweets by season, state, and co-occurring terms. Results: Tweet frequency was 228.5 per week in summer and 113.4 per week in the fall. Users tweeted from 40 different states; the most common were California (N = 401; 10.1 per 100,00 population) and Arizona (N = 216, 30.1 per 100,000 population), New York (N = 49), Florida (N = 21), and Washington, DC (N = 14). Term frequency analysis showed that for normal tweets, the 5 most frequent terms were “awareness,” “Arizona,” “disease,” “California,” and “people.” For retweets, the most common terms were “Gunner” (a dog name), “vet,” “prayer,” “cough,” and “family.” For replies, they were “dog,” “lung,” “vet,” “day,” and “result.” Several symptoms were mentioned: “cough” (normal: 8, retweets: 104, and replies: 7), “sick” (normal: 21, retweets: 42, replies: 7), “rash” (normal: 2, retweets: 6, replies: 1), and “headache” (normal: 1, retweets: 3, replies: 0). Conclusions: Valley fever tweets are potentially sufficient to track disease intensity, especially in Arizona and California. Data collection over longer intervals is needed to understand the utility of Twitter in this context.

Disclosures: None

Funding: None

Background: The CDC Core Elements of Antibiotic Stewardship (AS) include 7 evidence-based best practices adapted for a variety of healthcare settings, including nursing homes. We aimed to identify barriers and facilitators related to AS implementation in skilled nursing facilities (SNFs) within 18 months of the CMS mandate for AS implementation in SNFs, and to examine their relevance to the CDC’s Core Elements for Nursing Homes. Methods: We conducted 56 semistructured interviews with administrators, clinicians, and nonclinical staff at 10 SNFs in urban, suburban, rural, and border regions of Arizona. All interviews were recorded, transcribed, and imported into NVivo v12.0 software for constant comparative analysis by 3 researchers using a priori and emergent codes. After iterative coding, we confirmed high interrater reliability (κ = 0.8), finalized the code book, and used matrix coding queries to examine relationships and generate themes. Results: We identified 7 themes as “influencers” that were less (barrier) or more (facilitator) supportive of AS in SNFs. Intra- and interfacility communication were the most frequently described: respondents described stronger communication within the SNF and between the SNF and hospitals, labs, and pharmacies as critical to robust AS implementation. Other influencers included AS education, antibiotic tracking systems, SNF prescribing norms, human resources, and diagnostic resources. The Core Elements were reflected in all influencer themes except interfacility communication between SNFs and hospitals. Additionally, themes pertaining to systems emerged as critical to successful AS implementation, including the need to address: the interactions of multiple roles across the traditional SNF hierarchy, stewardship barriers from the lens of patient-level concerns (as opposed to population-level concerns), the distinction between antibiotic prescribing gatekeepers and stewardship gatekeepers, and care transition policies and practices. The Core Elements target many aspects of these systems themes—for example, they recognize the importance of creating a culture of stewardship. However, they do not address care transition policies or procedures beyond recommending that transfer-initiated antibiotics be tracked and verified. Conclusions: Because the interactions of various agents within and beyond the SNF can facilitate or inhibit stewardship in complex ways, our findings suggest the use of a systems approach to AS implementation that prioritizes communication within the SNF hierarchy, and between SNFs and hospitals, diagnostic facilities, and pharmacies. When followed, the CDC’s Core Elements can provide crucial guidance. However, SNFs need support to overcome the challenges of incorporating these elements into policy and practice. Additionally, more work is needed to understand and enhance stewardship-related care transition, which remains under-addressed by the CDC.

Disclosures: None

Funding: None

Background: Peripherally inserted central catheters (PICCs) are an increasingly common vascular access device. At our institution, >4,000 devices are placed per year by a trained team of vascular access nurses. Although PICCs are generally safe and effective, they do carry the risk of infection and thrombosis, and this risk increases exponentially with increasing number of lumens. As part of a multidisciplinary quality improvement effort to address rising CLABSI rates, we designed interventions to improve PICC utilization. Methods: The project team used 6-σ methodology, specifically following the DMAIC (define, measure, analyze, implement, control) framework to guide analysis and interventions. Process mapping, semistructured interviews with key stakeholders, electronic surveys, and audits were performed to identify gaps and inform interventions. The interventions consisted of 3 components: changes to the electronic ordering system, education (presentations to ordering providers and an online toolkit), and clinical decision support in the form of a team of vascular-access subject-matter experts who provided guidance on line selection. Results: In total, 4,655 PICCs and 434 midlines were inserted in the 12 months before the intervention, and 7,457 PICCs and 929 midlines were placed in the 24 months after the intervention. Following the implementation of the intervention, proportions of triple-lumen catheter utilization decreased from 31.9% to 22.3% (P < .0001). Concurrently, the proportion of single-lumen catheters has increased from 28.5% to 41.9% (P < .0001). Overall PICC utilization decreased in the postintervention period from an average of 387.9 PICCs placed per month to 310.7. The proportion of midline catheters increased from 8.5% of total lines inserted to 11.4% in the postintervention period (P < .001). Conclusions: Our intervention reduced overall PICC use and triple-lumen PICC use and increased relative utilization of single-lumen PICCs and midline catheters. Optimization of electronic orders, in conjunction with targeted education and decision support, can have a sustained impact on provider ordering behaviors and can shift the culture of utilization, even in a large academic medical center with frequent turnover of trainees.

Funding: None

Disclosures: Consulting fee- Merck (Priya Sampathkumar)

Background:Clostridiodes difficile is the leading healthcare-associated pathogen, with significant morbidity associated with acute C. difficile infection (CDI). However, polymerase chain reaction stool testing is unable to differentiate colonization from infection, leading to frequent overdiagnosis, unnecessary iatrogenesis, and additional costs. As a result, IDSA guidelines do not recommend C. difficile testing in patients with diarrheal symptoms attributed to other causes, including laxatives. Our group has previously investigated the use of a computerized clinical decision support (CCDS) tool to reduce inappropriate C. difficile testing in a single tertiary-care health system, with a subsequent 41% reduction in testing. We investigated the reduction in proportion of inappropriately completed tests with the randomized addition of a laxative alert to our existing CCDS. Methods: An existing electronic medical record-based CCDS tool was augmented by the addition of an automatic alert that notified the user if a patient received any of a set of identified laxative medications within 48 hours. During the 78-day pilot period, users encountered the existing CCDS or the CCDS with laxative alert (CCDS-LA), randomized by patient identification number. A proportional χ2 analysis was used to compare the proportion of aborted to completed tests among patients who met laxative criteria in the CCDS versus CCDS-LA groups. Results: In total, 187 test orders were attempted during the pilot period in 119 patients meeting the laxative alert criteria, with 43.3% order attempts randomized to the existing CCDS and 56.7% to the CCDS-LA. Of order attempts via the CCDS-LA, 50.0% were completed, compared to 64.2% of orders completed via the existing CCDS (22.1% relative reduction in test completion; P = .0525). Conclusions: We demonstrated substantially fewer completed C. difficile tests among patients receiving laxatives who were randomized to modified laxative-alert CCDS. Although our result did not reach statistical significance, the trend toward reduced inappropriate testing prompted the CCDS-LA alert to be adopted hospital-wide following completion of the test period. Further analyses of the pre- and postintervention periods are required to determine whether this intervention significantly impacts testing rates over time, as well as to determine the durability and safety of the CCDS-LA. Additional analyses are also needed to assess the impacts on hospital-onset CDI rate and the associated costs.

Funding: None

Disclosures: None

Background: Antibiotic stewardship programs (ASPs) have traditionally focused on inpatient prescribing, but they are now mandated to involve ambulatory settings. We developed and tested an educational tool in resident physicians to empower outpatient providers to perform self-reflection stewardship (SRS) to improve their antibiotic use. Results of the first SRS workshop are reported. Methods: A 90-minute SRS workshop focusing on the evaluation and management of sinusitis in ambulatory care was developed for PGY 2-3 internal medicine residents. Participants received a 15-minute didactic on the evaluation and management of adults with sinusitis, including typical microbiology, differentiation of bacterial sinusitis, and guideline recommendations on antibiotic treatment. In a computer lab, participants were instructed how to review charts of patients they had treated with antibiotics for sinusitis during the past year using the SlicerDicer application in Epic. Over 1 hour, they worked in pairs to complete and discuss a self-reflection inventory for 5 patients from each of their respective reviews. They evaluated pertinent history, comorbidities, presenting symptoms and signs, diagnostic testing performed, and a self-assessment of the subsequent antibiotic prescribing, including appropriateness of using an antibiotic, antibiotic choice and duration. In addition, they reflected on potential patient and prescriber challenges. Residents then identified common themes and developed a personal improvement plan for antibiotic prescribing for sinusitis. The last 15 minutes were spent debriefing with ASP faculty on reasons for overprescription of antibiotics for URIs and individual improvement plans. Residents completed workshop evaluations using a Likert scale and open-ended comments. Results: In total, 26 residents participated. All (100%) agreed or strongly agreed that the SRS workshop improved their understanding of how to obtain data on their own practice habits. Moreover, 23 (88%) agreed or strongly agreed that the workshop improved their understanding of when to prescribe antibiotics and how to practice antibiotic stewardship in the outpatient setting. Also, 20 participants (77%) agreed or strongly agreed that the SRS workshop helped them gain insight into reasons why they might overprescribe antibiotics in the outpatient setting. Furthermore, 25 (96%) agreed or strongly agreed that the SRS workshop helped them identify at least 1 way they could improve their antibiotic prescribing in the outpatient setting. Conclusions: The SRS workshop was well received by residents and offers a tool to empower primary care resident physicians to access their own antibiotic prescribing data, perform a structured self-reflection, and enhance their understanding of antibiotic stewardship in the ambulatory setting. SRS is a potential tool to improve ambulatory antibiotic use.

Disclosures: None

Funding: None

Background: A prolonged outbreak of carbapenemase-producing Serratia marcescens (CPSM) was identified in our quaternary healthcare center over a 2-year period from 2015 through 2017. A reservoir of IMP-4–producing S. marcescens in sink drains of clinical hand basins (CHB) was implicated in propagating transmission, supported by evidence from whole-genome sequencing (WGS). We assessed the impact of manual bioburden reduction intervention on further transmission of CPSM. Methods: Environmental sampling of frequently touched wet and dry areas around CPSM clinical cases was undertaken to identify potential reservoirs and transmission pathways. After identifying CHB as a source of CPSM, a widespread annual CHB cleaning intervention involving manual scrubbing of sink drains and the proximal pipes was implemented. Pre- and postintervention point prevalence surveys (PPS) of CHB drains performed to assess for CPSM colonization. Surveillance for subsequent transmission was conducted through weekly screening of patients and annual screening of CHB in transmission areas, and 6-monthly whole-hospital PPS of patients. All CPSM isolates were assessed by WGS. Results: In total, 6 patients were newly identified with CPSM from 2015 to 2017 (4.3 transmission events per 100,000 surveillance bed days [SBD]; 95% CI, 1.6–9.4). All clinical CPSM isolates were linked to CHB isolates by WGS. The CHB cleaning intervention resulted in a reduction in CHB colonization with CPSM in transmission areas from 72% colonization to 28% (ARR, 0.44; 95% CI, 0.25–0.63). A single further clinical case of CPSM linked to the CHB isolates was detected over 2 years of surveillance from 2017 to 2019 following the implementation of the annual CHB cleaning program (0.7 transmissions per 100,000 SBD; 95% CI, 0.0–3.9). No transmissions were linked to undertaking the cleaning intervention. Conclusions: A simple intervention targeted at reducing the biological burden of CPSM in CHB drains at regular intervals was effective in preventing transmission of carbapenemase-producing Enterobacterales from the hospital environment to patients over a prolonged period of intensive surveillance. These findings highlight the importance of detailed cleaning for controlling the spread of multidrug-resistant organisms from healthcare environments.

Funding: None

Disclosures: Jason Kwong, Austin Health

Background: Nipah encephalitis outbreaks mostly involve multiple patients. We report a case of Nipah virus encephalitis (NVE), which had no documented secondary cases in spite of many having prolonged and close contact with the patient. Methods: A 21-year-old male was admitted with NVE on May 30, 2019. Before the confirmatory report, there was close contact with multiple healthcare workers (HCWs), defined as exposure for >1 hour to the patient or his immediate environment and/or exposure to body fluids. We conducted extensive contact tracing of all HCWs who had come into close contact with the proven NVE case from the time of admission to the time of discharge. This contact tracing included those who had nursed him before the diagnosis with usual standard precautions and those who had nursed him after the diagnosis with full PPE. These HCWs were reviewed daily for fever and respiratory symptoms. All those who developed these symptoms within the 3 weeks of exposure where tested for NEV with a throat swab using RT-PCR. This testing was conducted twice over 3 days to confirm negative results. For the close family contacts that were asymptomatic, both throat swab and serum for Nipah IgM were tested. Results: In total, 169 HCW contacts were identified at our hospital. Of these, 94 were at high risk according the predetermined criteria and others were low-risk contacts. Moreover, 7 HCWs developed fever and respiratory symptoms within the defined surveillance period; 5 had symptoms before the diagnosis (using only standard precautions) and 2 were in contact with full PPE after the diagnosis. All of these symptomatic contacts were tested for NEV (throat swab and serology), and all were negative. The family members of the patient (his mother and aunt) who had cared for him throughout his illness period of 12 days before the diagnosis were also tested and were seronegative for NEV. Conclusions: This NEV case had very low transmission capability; even close family members who cared for him for 12 days without any precautions and had exposure to urine (which was positive for NEV) did not contract the disease. The absence of overt respiratory involvement and young age of the affected patient could have contributed to low transmissibility both prior to hospitalization and during the hospitalization.

Funding: None

Disclosures: None

Background: This review describes the epidemiology of carbapenemase-producing organisms (CPO) in both the community and hospitalized populations in the province of Alberta. Methods: Newly identified CPO-positive individuals from April 1, 2013, to March 31, 2018, were retrospectively reviewed from 3 data sources, which shared a common provincial CPO case definition: (1) positive CPO results from the Provincial Laboratory for Public Health, which provides all referral and confirmatory CPO testing, (2) CPO cases reported to Alberta Health, and (3) CPO surveillance from Alberta Health Services Infection Prevention and Control (IPC). The 3 data sources were collated, and initial CPO cases were classified according to their likely location of acquisition: hospital-acquired, hospital-identified, on admission, and community-identified. Risk factors and adverse outcomes were obtained from linkage to administrative data. Results: In total, 171 unique individuals were newly identified with a first-time CPO case. Also, 15% (25 of 171) were hospital-acquired (HA), 21% (36 of 171) were hospital-identified (HI), 33% (57 of 171) were on admission, and 31% (53 of 171) were community identified. Overall, 9% (5 of 171) resided in long-term care facilities. Of all patients in acute-care facilities, 30% (35 of 118) had infections and 70% were colonized. Overall, 38% (65 of 171) had an acute-care admission in the 1 year prior to CPO identification; 59% (63 of 106) of those who did not have a previous admission had received healthcare outside Alberta. A large proportion of on-admission cases (81%, 46 of 57) and community-identified (66%, 33 of 53) cases did not have any acute-care admissions in Alberta in the previous year. Overall, 10% (14 of 171) had ICU admissions in Alberta within 30 days of CPO identification, and 5% (8 of 171) died within 30 days. The most common carbapenemase gene identified was NDM-1 (53%, 90 of 171). Conclusions: These findings highlight the robust nature of Alberta’s provincial CPO surveillance network. We reviewed 3 different databases (laboratory, health ministry, IPC) to obtain comprehensive data to better understand the epidemiology of CPO in both the community and hospital settings. More than half of the individuals with CPO were initially identified in the community or on admission. Most had received healthcare outside Alberta, and no acute-care admissions occurred in Alberta in the previous year. It is important to be aware of the growing reservoir of CPO outside the hospital setting because it could impact future screening and management practices.

Funding: None

Disclosures: None

Background: Morbidity and mortality associated with invasive fungal infections and concerns of emerging antifungal resistance have highlighted the importance of optimizing antifungal therapy among hospitalized patients. Little is known about antifungal stewardship (AFS) practices among acute-care hospitals. We sought to assess AFS activities within Connecticut and to identify opportunities for improvement. Methods: An electronic survey assessing AFS practices was distributed to infectious disease physicians or pharmacy antibiotic stewardship program leaders in Connecticut hospitals. Survey questions evaluated AFS activities based on antibiotic stewardship principles, including several CDC Core Elements. Questions assessed antifungal restriction, prospective audit and feedback practices, antifungal utilization measurements, and the perceived utility of a local or statewide antifungal antibiogram. Results: Responses were received from 15 respondents, which represented 20 of 31 hospitals (65%); these hospitals made up the majority of the acute-care hospitals in Connecticut. Furthermore, 18 of these hospitals (58%) include antifungals in their stewardship programs. Also, 16 hospitals (52%) conduct routine review of antifungal ordering and provide feedback to providers for some antifungals, most commonly for amphotericin B, voriconazole, micafungin, isavuconazole, and flucytosine. All hospitals include guidance on intravenous (IV) to oral (PO) conversions, when appropriate. Only 14 of hospitals (45%) require practitioners to document indication(s) for systemic antifungal use. Most hospitals (17, 55%) provide recommendations for de-escalation of therapy in candidemia, though only 4 (13%) have institutional guidelines for candidemia treatment, and only 11 hospital mandates an infectious diseases consultation for candidemia. Assessing outcomes pertaining to antifungal utilization is uncommon; only 8 hospitals (26%) monitor days of therapy and 5 (16%) monitor antifungal expenditures. Antifungal susceptibility testing on Candida bloodstream isolates is performed routinely at 6 of the hospitals (19%). Most respondents (19, 95%) support developing an antibiogram for Candida bloodstream isolates at the statewide level. Conclusions: Although AFS interventions occur in Connecticut hospitals, there are opportunities for enhancement, such as providing institutional guidelines for candidemia treatment and mandating infectious diseases consultation for candidemia. The Connecticut Department of Public Health implemented statewide Candida bloodstream isolate surveillance in 2019, which includes antifungal susceptibility testing. The creation of a statewide antibiogram for Candida bloodstream infections is underway to support empiric antifungal therapy.

Funding: None

Disclosures: None

Background: Evidence-based best practices are available for the reduction and prevention of Clostridioides difficile infection (CDI). Often, these practices are not consistently followed in many inpatient care settings. A learning collaborative model resulted in a cost neutral, rapid, sustainable, statistically significant reduction in CDI events across an 88-hospital campus system without requiring hospitals to standardize laboratory methods, increase spending or increase staffing. Methods: In March 2018, a healthcare system with 88 critical access and community hospital campuses across 29 states participated in a harms-reduction learning collaborative. The collaborative format included educational webinars, gap analyses, action plans, and coaching calls facilitated by subject matter experts (SMEs). A collaborative cohort of 11 hospitals (55% rural*) was identified as having significant opportunity for improvement. These facilities participated in 3 monthly coaching calls. The coaching calls supported peer-to-peer sharing of practices and discussions of challenges and successes, and educational materials and presentations were provided by SMEs in pharmacy and infection prevention. Results: Statistically significant changes for the 88-hospital system as a whole: (1) 2018 compared to 2017: P < .001 (statistically significant); (2) 1H2018 compared to 2H2018 (before-and-after collaborative): P = .001; (3) 2019 compared to 2018: P < .001 (statistically significant). Statistically significant changes for the collaborative cohort: (1) 2018 compared to 2017: P < .001; (2) 1H2018 compared to 2H2018 (before-and-after collaborative): P = .002; and (3) 2019 compared to 2018: P < .001. We used 2-proportion, 2-tailed z-test for our analysis. Conclusions: Utilizing a learning collaborative model that included webinars, gap analyses, and interactive coaching calls, a cohort of 11 hospitals was able to induce rapid improvements to adherence of evidence-based practices resulting in a rapid, sustained, statistically significant improvement for both the cohort hospitals and the healthcare system.

*2018 American community survey, US Census.

Funding: None

Disclosures: None

Background: Hospital-acquired pneumonia (HAP) is one of the most common healthcare-associated infections (HAIs). Interventions based on the identification of patients at risk for aspiration with subsequent application of multidisciplinary measures, such as speech therapy follow-up, head elevation, oral hygiene, and patient and family education can be effective in reducing the incidence of HAP. In 2016, the step-down unit of our institution experienced an increase in the incidence of HAP with 21 cases. A root-cause analysis showed that most of them were related to comorbidities that increased aspiration risk. We conducted an study to decrease the incidence of HAP through a multidisciplinary bundled intervention. Methods: We conducted a quasi-experimental study in a 45-bed step-down unit from January 2016 to June 2019. In January 2017, we conducted an educational intervention with all the unit team, reinforcing practices of bed head elevation and oral hygiene. In June 2018, we observed inconsistencies in practice and conducted a second intervention with another round of educational training and a bundled intervention consisting of the following elements: identification of patients at risk for aspiration at admission by a speech therapy evaluation, bed-head elevation, oral hygiene, feeding guidance individualized to each patient by a nutritionist and a speech therapist, patient and family education with a printed material, signaling of aspiration risk in a care plan board within the room and development of a sialorrhea treatment protocol. HAP surveillance was conducted in accordance to CDC definitions and was reported as number of HAP cases per 1,000 patient days. Results: Our first intervention decreased the incidence of HAP in the first semester of 2017 from 1.03 to 0.29 (graph) but was not sustained. The incidence started to increase in the second semester of 2017 and reached a high incidence of 1.87 HAP per 1,000 patient days in the first semester of 2018. The second bundled intervention succeeded in decreasing HAP incidence to 0.57 in the second semester of 2018 and 0.23 in the first semester of 2019. Conclusions: An educational intervention combined with a bundled intervention focused on strategies to reduce the risk of aspiration succeeded in decreasing the incidence of HAP in a step-down unit. However, the sustainability of improvements remains challenging.

Funding: None

Disclosures: None

Background: Handwashing sinks in healthcare environments are reservoirs for healthcare pathogens and antibiotic-resistant microorganisms (ARO). We investigated the distribution of HCP and ARO within and among handwashing sinks in healthcare settings. To do this, we determined the differences in the number of ARO between samples within a sink (biofilm vs planktonic samples), between sink types (healthcare worker [HCW] vs patient room sinks), and between hospitals in the same city. Methods: Tap water, sink surface, drain cover, tail pipe, p-trap water and p-trap samples were collected from 2 patient room sinks and 2 HCW sinks over 11 months in 2 acute-care hospitals. Suspected pathogens were isolated from selective media (Pseudosel, Chromagar KPC, and MacConkey with 2 mg/L cefotaxime) and identified via MALDI-ToF. Isolates confirmed to be healthcare pathogens were characterized via disk diffusion to determine their antibiotic susceptibility according to CLSI guidelines. Isolates not susceptible to carbapenems (meropenem or ertapenem) were tested further via the modified carbapenem inactivation method to detect carbapenemase production. Results:Pseudomonas aeruginosa and Enterobacteriaceae (Enterobacter spp, Klebsiella spp, and Citrobacter spp) were the most frequently isolated pathogens. Among these isolates (195 P. aeruginosa and 42 Enterobacteriaceae isolates), 28.5% of P. aeruginosa and 85.7% of Enterobacteriaceae were nonsusceptible to 1 or more of the antibiotics tested. Of the isolates that were nonsusceptible to a carbapenem (46 of 237; 19%), none displayed phenotypic carbapenemase production. Other mechanisms of resistance have not been confirmed. There was no significant difference in the percentage of nonsusceptible HCP isolated from biofilm samples (from p-trap and tail pipe) compared to planktonic (p-trap water) samples (P > .05 for P. aeruginosa and Enterobacteriaceae). A greater percentage of resistant or intermediate isolates was recovered from patient room sinks than from HCW sinks (P < .05) for both P. aeruginosa and Enterobacteriaceae isolates (76.4 vs 32.9% for Enterobacteriaceae, 25.6 vs 0.3% for P. aeruginosa). We detected no significant difference in percentage of nonsusceptible isolates between the 2 hospitals sampled (P > .05). Conclusions: This survey of healthcare sinks supports previous work citing that they are reservoirs for HCP and ARO. This work further examines the distribution of HCP and ARO within and among sinks in these environments. Our findings thus far in the 2 hospitals studied reveal a higher percentage of ARO in patient sinks than in HCW sinks. This finding may suggest a higher input of ARO from patient use or greater selective pressure in patient room sinks.

Disclosures: None

Funding: Lauren Franco, Centers for Disease Control and Prevention

Background: Mount Sinai Beth Israel is a 350-bed, acute-care hospital located on Manhattan’s Lower East Side. In 2014, the hospital had reached a high (9.8 cases per 10,000 patient days ) hospital-onset (HO) C. difficile rate. By 2015, this rate had decreased to 5.6 cases per 10,000 patient days because of compliance with established C.difficile bundle practices performed by nursing and environmental services. Despite these interventions, HO C. difficile events continued to occur. We realized that more had to be done to gain control over our rates. To determine areas for further improvement, infection prevention held an RCA meeting for every positive hospital-onset result. We discovered from these RCAs that many C. difficile tests were ordered without a valid indication. We believed that measures could be taken to ensure that only C. difficile tests with a valid indication would be ordered. Methods: We used the Plan-Do-Study-Act (PDSA) model to look at what changes could be made to reduce our rate and to sustain this reduction. Multidisciplinary meetings of leaders and frontline staff were held to determine why patients were being tested for C. difficile. The following indications were revealed: repeat tests for same patient to “catch” a positive result after the first test was negative; inclusion as part of patient “pan-culturing”; testing patients who had diarrhea after receiving laxatives; and C. difficile cultures for patients who were asymptomatic. Starting in 2016, 3 consecutive interventions were implemented in fairly rapid succession. First, a C. difficile testing algorithm was developed. Second, a C. difficile test order protocol with a “hard stop” to prevent inappropriate indications was placed in the EMR. Last, a multidisciplinary form, called the C. difficile Team Huddle Form, was created for use by all members of the patient’s team. This form gave MDs, RNs, and PCAs a framework to decide together whether the test was indicated for the patient. If the team agreed to test, the ID physician on service was called for approval. Results: These 3 interventions yielded a sustained and statistically significant decrease (P = 0.0007) in the facility-wide hospital-onset C. difficile from a preintervention rate of 5.6 cases per 10,000 patient days in 2015 to 0.4 in 2019. Conclusions: Multidisciplinary use of the C. difficile testing interventions led to further reduction of the hospital-onset C. difficile infection rate. To sustain this rate reduction over time, infection prevention specialists must work with providers and frontline staff on an ongoing basis.

Funding: None

Disclosures: None