Background: Significant increases in healthcare-associated infections (HAIs) including central-line associated blood stream infections (CLABSIs) have been reported during the COVID-19 pandemic. Acute-care hospitals have faced staffing and personal protective equipment shortages, increased critical care capacity, and diversion of resources from traditional HAI surveillance and prevention efforts. In this study, we characterized CLABSIs among patients with COVID-19 and compared demographics, comorbidities, and outcomes between patients diagnosed with CLABSI with and without COVID-19. Methods: This is an observational retrospective cohort study of all patients diagnosed with CLABSI as defined by NHSN at William P. Clements Jr. University Hospital from April 1, 2020, through September 30, 2021. A retrospective chart review was conducted to identify demographics, comorbidities, and outcomes of hospitalized patients diagnosed with CLABSI. Patients hospitalized with and without COVID-19 were compared using the independent-sample t test for means and the χ2 test for proportions. Results: Overall, 82 patients diagnosed with CLABSI between April 1, 2020, to September 30, 2021, among whom 31 (38%) were hospitalized with COVID-19 and 51 (62%) were not hospitalized with COVID-19. Patients hospitalized with COVID-19 were significantly more likely to be obese (58% for COVID-19 positive vs 26% for COVID-19 negative; P = .01) and to require extracorporeal membrane oxygenation (19% vs 4%; P = .04). However, COVID-19 patients were significantly less likely to have hematologic malignancy (7% vs 28%; P = .03), undergone bone marrow transplantation (0% vs 18%; P = .01), or have neutropenia (3% vs 22%; P = .03). There were no significant differences in line type or organism identified. Gram-positive pathogens were identified in 16 patients (52%) hospitalized with COVID-19. Gram-negative pathogens were identified in 3 patients (10%); fungal organisms were identified in 10 patients (32%), and 2 cases (7%) were polymicrobial. Patients with COVID-19 were significantly more likely to require an ICU stay (84% vs 43%). p Conclusions: High device utilization as well as prolonged hospitalization and line days among patients with COVID-19 along are contributing risk factors for CLABSI among patients hospitalized with COVID-19. This finding highlights the need for ongoing HAI surveillance and prevention efforts in patients hospitalized with COVID-19 given their characteristics and increased risk for CLABSI. Reinforcing infection prevention efforts by accentuating the importance of optimal line care and regular feedback are crucial, especially among patients hospitalized with COVID-19.

Funding: None

Disclosures: None

Background: Patients requiring vascular catheters are at risk for bloodstream infections (BSIs), particularly those with central venous access devices (CVADs). Central-line–associated bloodstream infections (CLABSIs) may occur as a result of the introduction of pathogenic microbes during CVAD access procedures, including through the needleless connector. The use of an antiseptic scrub is recommended to disinfect the needleless connector before device access, and this procedure has been shown to reduce the risk for CLABSI. We identified perceived barriers and facilitators and assessed compliance with instructions for use of chlorhexidine or alcohol antisepsis products (CHG or IPA; 5-second scrub time plus 5-second dry time) and alcohol antisepsis products (IPA; facility protocol 15-second scrub time plus let dry) for needleless connector disinfection. Methods: We performed a multiple-methods study involving focus groups composed of a convenience sample of nurses and clinical observations of CVAD needleless-connector access procedures in 3 medical ICUs and 1 surgical ICU at 2 academic medical centers. We used open-ended questions to guide the focus-group discussions. We directly observed nursing staff performing needleless-connector disinfection following a time–motion paradigm using an electronic tool to document the observed needleless-connector access events and to measure needleless-connector antiseptic scrub times and dry times. Results: In total, 8 focus groups involving 28 nurses revealed access to the antiseptic product and lesser workload as best-practice facilitators of needleless-connector disinfection. Identified barriers were often the opposite of the facilitators, particularly the time required per needleless connector access using IPA and knowledge deficits regarding the need for disinfection between multiple needleless-connector accesses. From 36 observations, including a total of 48 access events, we determined that the mean scrub times were below the recommended times, especially for IPA (Table 1). Drying time after use of either antisepsis product was negligible. Conclusions: A lack of access to the disinfection product, emergency situations, and increased workload were perceived barriers to needleless-connector disinfection. Observed scrub times and drying times were shorter than recommended, much more so for IPA. These deficits in the performance of needleless-connector disinfection may increase the risk of CLABSI. Ongoing education and periodic competency evaluation of needleless-connector disinfection are needed to imbed and sustain best practices.

Funding: Professional Disposables, Inc.

Disclosures: None

Background: We conducted an anonymous survey in compliance with German data protections regulations among participants of the annual infectious disease and control meeting in Freiburg, Germany, in October 2021. Methods: In total, 391 surveys were returned: 188 from nurse infection control practitioners (ICPs) and 66 from specially infection control trained physicians (STPs). We report the results of these 2 subgroups regarding their perceptions and emotions during the pandemic. Descriptive statistics and χ2 test with P < .05 were used when applicable. Results: Shortages of medical masks or FFP2 masks during the first pandemic wave in 2020 were reported by 48.5% STPs and 57.4% ICPs. STPs and ICPs relied equally on information provided by the Robert Koch Institute, the WHO, the ECDC and the CDC. Occupational health information was sought significantly more often by ICPs; only 17% of ICPs never used this source versus 51.5% of STPs (P < .001). Most ICPs (58%) and STPs (51%) described their relationship to local authorities as good as well as communication with institutional leaders (69.7%). Fewer ICPs (36.1%) felt frequently appreciated during the pandemic compared to 45.5% of STPs and more ICPs (25%) reported frustration than STPs (18.2%). However, the differences were not statistically significant. Rarely, ICPs (2.1%) or STPs (1.5%) felt unsafe at work and only 1.6% of ICPs and no STPs reported loss of motivation. In addition, 13.8% of ICPs and 12.1% of STPs often felt overwhelmed, but only 3.2% of ICPs and no STPs felt hopeless. Their self-reported competency was rated as high by 75% of ICPs and 69.7% of STPs. The 5 most frequent free-text comments regarding “lessons learned” pertained to better crisis communication, better supply chain management, precise regulations, “less talking more doing,” and mandatory vaccination. The most frequent free-text general comments pertained to maintain basic hygiene measures in private and public life because of the pandemic. Conclusions: Our survey results indicate a high level of resilience among members of infections control teams in German medical institutions despite obvious shortcomings in supplies during the first wave of the pandemic. There were no significant differences between physician and nurse members of infection control teams regarding their perceptions and emotions, indicating a homogenous situation within the teams. The high level of self-perceived competency has likely helped deal with the pandemic and prevented the feeling of loss of control implied in the question items “feeling overwhelmed” and “hopeless.”

Funding: None

Disclosures: None

Background: COVID-19 occupational exposures have been examined using death certificates and employment data from the Bureau of Labor Statistics and the O*Net database in the United States. However, no studies have examined cases, hospitalizations, and deaths by occupation using hospital records.1 We analyzed COVID-19 cases using hospitalization data from a large, rural community hospital to fill this gap in the evidence base. Methods: A retrospective cross-sectional study design was used to examine patients with COVID-19 from March 1 through July 31, 2020. We examined demographic characteristics, such as age, sex, race or ethnicity, and length of stay (LOS), among those who tested positive for SARS-CoV-2. Epidemiological risk factors were also analyzed, including smoking status, body mass index (BMI), alcohol use, and occupation. Occupational data were processed using the National Institute for Occupational Safety and Health Industry and Occupation Computerized Coding System. Homemakers, disabled persons or retirees, students or minors, and listed occupations with insufficient information were excluded from the analysis. Occupations were categorized into 23 major occupation groups based on the 2018 Standard Occupational Classification System. To examine whether certain occupations are at a higher risk due to COVID-19, we stratified the analysis by overall cases, hospitalizations, and deaths. Microsoft Power BI Desktop and IBM SPSS version 28.0.0.0 software were used to analyze the data. This study was reviewed and approved by the local institutional review board. Results: In total, 2,132 COVID-19 diagnoses with 1,049 total hospitalizations were identified during the study period. Most cases were in the group aged 50–64 years, white race, and/or Hispanic ethnicity (Table 1). Most cases never or rarely drank alcohol, were nonsmokers, and had a BMI ≥30 (Table 2). The average LOS among those hospitalized for COVID-19 was 6.46 days. The occupational analysis revealed a higher frequency of cases among those in management (n = 95, 14%) and healthcare (n = 83, 12%), with those in management (n = 40, 14%) and sales (n = 29, 10%) having the highest frequency of being hospitalized. However, the highest frequency of deaths occurred among those in building and grounds cleaning and maintenance occupations (13%) (Table 3). Conclusions: This study describes the burden of COVID-19 in a rural area with a large aging population and highlights potential health disparities among severe cases and deaths in different occupational groups.

1. Baker MG, Peckham TK, Seixas NS. Estimating the burden of United States workers exposed to infection or disease: a key factor in containing risk of COVID-19 infection. PLoS One 2020;15:e0232452.

Funding: None

Disclosures: None

Background: Singapore General Hospital (SGH) is the largest acute tertiary-care hospital in Singapore. Healthcare workers (HCWs) are at risk of acquiring COVID-19 in both the community and workplaces. SGH has a robust exposure management process including prompt contact tracing, immediate ring fencing, lock down of affected cubicles or single room isolation for patient contacts, and home isolation orders for staff contacts of COVID-19 cases during the containment phase of the pandemic. Contacts were also placed on enhanced surveillance with PCR testing on days 1 and 4 as well as daily antigen rapid tests (ARTs) for 10 days after exposure. Here, we describe the characteristic of HCWs with COVID-19 during the third wave of the COVID-19 pandemic. Methods: This retrospective observational study included all SGH HCWs who acquired COVID-19 during the third wave (ie, the 18-week period from September 1 to December 31, 2021) of the COVID-19 pandemic. Univariate analysis was used to compare characteristics of work-associated infection (WAI) and community-acquired infection (CAI) among HCWs. Results: Among a workforce of >10,000 at SGH, 335 HCWs acquired COVID-19 during study period. CAI (exposure to known clusters or household contact) accounted for 111 HCW infections (33.1%). Also, 48 HCWs (14.3%) had a WAI (ie, acquired at their work places where there was no patient contact). Among WAsI, only 5 HCWs had hospital-acquired infection (confirmed by phylogenetic analysis). The sources of exposure for the remaining 176 HCWs were unknown. Weekly incidence of COVID-19 among HCWs was comparable to the epidemiology curve of all cases in Singapore (Fig. 1 and 2). The mean age of HCWs with COVID-19 was 39.6 years, and most were women. At the time of positive SARS-CoV-2 PCR test, 223 HCWs were symptomatic, and 67 (20.0%) of them had comorbidities. Only 16 HCWs (4.8%) required hospitalization, and all recovered fully with no mortality (Table 1). Being female was associated with community COVID-19 acquisition (OR, 4.6, P Conclusions: During the thrid wave of the COVID-19 pandemic, a higher percentage of HCWs at SGH acquired the infection from the community than from the workplace. Safe management measures, such as universal masking, social distancing, and robust exposure management processes including prompt contact tracing and environmental disinfection, can reduce the risk of COVID-19 in the hospital work environment.

Funding: None

Disclosures: None

Background: COVID-19 has had a disproportionate effect on nursing homes residents as well as people from racial and ethnic minorities. Whether differences in mortality due to COVID-19 exists for nursing-home residents from racial and ethnic minorities is less clear, with some previous studies reporting systems-level disparities. The Department of Veterans’ Affairs (VA) has nursing homes, termed community living centers (CLCs), across the United States. We hypothesized that differences in COVID-19–related mortality among racial and ethnic minorities would be less pronounced among residents of CLCs. Methods: Using data from the VA Corporate Data Warehouse, we conducted a retrospective cohort study from April 14, 2020 (implementation of population-based testing) to December 10, 2020 (availability of a COVID-19 vaccine). Inclusion criteria were residents with a positive SARS-CoV-2 test while residing in or <48 hours before admission to a CLC. Positive tests <180 days after a prior positive test were excluded. We assessed the cohort for demographics, including self-reported race or ethnicity, clinical characteristics, and survival probability including all-cause 30-day mortality. A multivariable logistic regression model was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for all-cause 30-day mortality that included race, ethnicity, age, and Charlson comorbidity index (CCI). Results: Among 14,759 CLC residents, 651 (4.4%) had a positive SARS-COV-2 test. Their mean age was 75.7 ± 11.3 years, and self-reported race or ethnicity was 68% White (445 of 651), 23% Black (152 of 651), and 4% Hispanic/Latinx (27 of 651). The mean CCI was lower among White residents than Black residents (4.15 ± 2.6 vs 4.61 ± 3.1, respectively). All-cause 30-day mortality for CLC residents following positive SARS-COV-2 test was 25% for White patients, 14% for Black patients, and 15% for Hispanic/Latinx patients (Fig. 1). Age (in years), but neither race or ethnicity nor CCI, was independently associated with all-cause 30-day mortality (OR, 1.07; 95% CI, 1.05–1.09) in CLC residents with COVID-19. Conclusions: Among VA CLC residents with a positive COVID-19 test, minority CLC residents did not have worse outcomes than white residents, suggesting that users of the VA healthcare system may enjoy abrogation of some aspects of health disparities.

Funding: None

Disclosures: None

Background: Throughout the COVID-19 pandemic, CDC Division of Healthcare Quality Promotion (DHQP) has provided technical assistance in support of state, tribal, local, and territorial health departments for COVID-19 healthcare outbreak management and infection prevention and control (IPC). We characterized the volume and trends of technical assistance provided during the pandemic to inform the future needs of health departments for COVID-19 healthcare IPC and DHQP resources required to meet these needs. Methods: In January 2020, DHQP began receiving COVID-19 IPC TA requests directly from health departments for remote assistance or from CDC staff on field deployments providing onsite support. DHQP subject-matter experts provided responses via e-mail or, for more complex inquiries, outbreaks, or field deployments, via phone consultations. Records of e-mail communications and phone consultations were entered into an inquiry database for tracking. We calculated the number, mean, and range of technical-assistance responses by jurisdiction and by month from January 2020 through December 2021. We designated months as high-volume periods for technical assistance if inquiries surpassed the 75th percentile. Results: In total, 1,869 IPC technical-assistance responses were provided. Of all technical-assistance responses, 1,725 (92%) were to state or local health departments, 115 (6%) were tribal nations, and 28 (2%) were US territories. IPC technical assistance was provided to all 50 states and the District of Columbia, 16 tribal nations, and 5 US territories. The average total number of technical assistance responses per site during the 24-month period was 34 to state and local HDs (range, 2–111), 6 to tribal nations (when tribal nation was specified; range, 1–17), and 6 to US territories (range, 1–15). E-mail communications comprised 1,164 responses (62%); phone consultations made up the remaining 705 responses (38%). Of phone consultations, 350 (50%) were with CDC field deployers providing onsite support to health departments. The average number of technical-assistance responses provided each month across all jurisdictions was 78 (range, 0–334); months with high volumes included April–August 2020 and January 2021. Conclusions: These findings highlight the high-level collaboration between federal and state, tribal, local, and territorial health department partners in remote and onsite support of COVID-19 prevention and response efforts in healthcare settings. Variations in monthly volumes of health-department COVID-19 healthcare IPC technical assistance requests may reflect factors such as fluctuations in community infection rates and changes in CDC IPC guidance. The ability to provide effective technical assistance during pandemic response depends on the CDC maintaining sufficient healthcare IPC staffing and expertise.

Funding: None

Disclosures: None

Background: In November 2020, the FDA issued an emergency use authorization (EUA) for monoclonal antibodies (mAbs) to be used in outpatients with COVID-19 infections who are at a high risk of progressing to severe disease. However, becuase the EUA had limited indications for inpatients, data on their use in hospitalized patients are limited. In this study, we have described the use of mAbs among hospitalized patients with nosocomial COVID-19. Methods: We retrospectively analyzed cases of nosocomial COVID-19 in 2 tertiary-care hospitals from November 1, 2020, to October 11, 2021, and we identified patients who received mAbs. The study period was prior to the ο (omicron) variant (B.1.1.529) being detected in the United States, and infections in the patients were likely primarily with the α alpha variant (B.1.1.7) and the δ (delta) variant (B.1.617.2) of SARS-CoV-2, which responded well to treatment with bamlanivimab and casirivimab–imdevimab. All patients had a negative SARS-CoV-2 PCR on admission. Data on clinical outcomes, including administration of medications for COVID-19, increases in oxygen requirements, ICU admission, mechanical ventilation, and death were collected by a review of the electronic medical record. The study was approved by the institutional review board with expedited approval. Descriptive statistics, such as means and standard deviations of continuous variables and proportions of categorical events or variables, were tabulated to describe patient characteristics and outcomes. Results: The 71 patients included in the study (age range, 39–89 years; median age, 70 years; 51% female) received either bamlanivimab (n = 31) or casirivimab–imdevimab (n = 40). The length of stay ranged from 6 to 242 days (median, 26 days). The comorbidities present included cardiovascular disease (56%), diabetes (45%), obesity (31%), autoimmune disease or immunosuppression (27%), kidney disease (23%), and pulmonary disease (20%). Most of the patients included in the study were incompletely vaccinated or unvaccinated (94%) and were negative for SARS-CoV-2 antibodies (81%). Prior to receiving the mAbs, 23% of patients required supplemental oxygen, including 3 patients who required mechanical ventilation. These patients required oxygen support due to non–COVID-19–related conditions. After mAb infusion, 72% of patients had no increase in their oxygen requirements, and 93% did not progress to mechanical ventilation. Overall, 7 deaths were attributed to COVID-19 among the studied patients (10%). Conclusions: Our study describes the use of mAbs in hospitalized patients with nosocomial COVID-19. Most of the patients who received mAbs had no progression to severe COVID-19, despite having significant comorbidities. The use of mAbs in nosocomial COVID-19 may be associated with beneficial outcomes.

Funding: None

Disclosures: None

Background: In the last 2 years of the COVID-19 pandemic, Singapore has been forced to explore alternative sites to quarantine persons or manage infected cases during surge periods in a national effort not to overwhelm the public healthcare facilities. External quarantine facilities were created at the EXPO and further extended to D’Resort and other hotels in May 2020. Infection prevention (IP) practices were implemented at these external facilities, where training non–healthcare staff to quickly learn and understand these required practices has been challenging. A team of staff from different clinical disciplines was formed to manage the COVID-19 patients at these facilities. The Infection Prevention and Epidemiology (IPE) department was invited to train all staff, including the clinical team, management agency, and security staff, regarding IP measures. We have described the system and approach used in the rapid training of all staff in IP measures where the goal is zero transmission while providing care to COVID-19 patients. Methods: Training materials were developed to facilitate rapid learning by all staff; medical jargon was avoided. Curriculum included precautions to be taken while performing terminal cleaning of patient rooms, serving meals, disinfecting phones and thermometers, as well as donning and doffing personal protective equipment (PPE). “Green” and “red” zones were created to assist staff in remembering appropriate PPE to be used. PPE training was provided using slides and video. Posters were created as a guide for staff at donning and doffing stations. Additionally, the IPE training team utilized an online data collection tool to capture staff completion on IP training and PPE competency for record keeping. We used a ‘soft’ approach because staff members were fearful of the unknown when caring for COVID-19 patients. Daily audits were conducted with immediate concurrent feedback to engage the relevant stakeholders. Infection prevention liaison officers (IPLOs) were appointed to assist in the daily audits. An electronic audit tool was used to facilitate audit and quick analysis. Conclusions: The experience gained in the last 2 years has been useful and may provide a template if new external sites are needed in the future because of the potential surge associated with the ο (omicron) variant.

Funding: None

Disclosures: None

Background: Carbapenem-resistant Acinetobacter baumannii (CRAb) is considered a public health threat, and this pathogen is typically associated with hospital infections. At 3 months after opening the hospital’s dedicated COVID-19 unit, 2 patients were identified with CRAb. Infection prevention staff collaborated with staff in the COVID-19 unit, hospital leadership, and health department partners to develop mitigation strategies and to prevent additional transmission. Methods: Admissions to the COVID-19 unit were stopped. Biweekly surveillance cultures were collected to identify any patients potentially colonized with CRAb. An infection control risk assessment was conducted to determine breaches in infection prevention practices. The risk assessment included environmental rounding of the area, epidemiological investigation, environmental testing, pulsed-field gel electrophoresis (PFGE) testing, and observing infection prevention practices. Results: The risk assessment identified multiple gaps in infection control practices, for example, gaps in hand and environmental hygiene practices. The extended use of personal protective equipment (PPE), staff shortages, fatigue, and staff taking on multiple roles and tasks outside their general job duties were other gaps identified by the team. Between June and September 2020, 43 additional CRAb cases were identified in the facility, with 4 (9.8%) cases outside the COVID-19 unit. Moreover, 29 cases (64%) were considered clinical infections and 16 (36%) were identified from surveillance efforts. Environmental cultures identified 1 positive surface with CRAb. PFGE testing was completed on 44 isolates; 42 isolates had identical PFGE patterns, and 2 isolates were unrelated to the COVID-19 unit; 2 isolates were closely related (with 3-band differences) but were not identified in the COVID-19 unit. Conclusions: The inability to definitively identify the source of transmission made it challenging to determine the best approach to eradicating the pathogen. Mitigation for outbreaks should focus on not deviating from core infection control practices.

Funding: None

Disclosures: None

Background: The Occupational Safety and Health Administration (OSHA) Respiratory Protection standard (29 CFR 1910.134) states that it is an employer’s responsibility to establish and maintain a respiratory protection program when a respirator is necessary to protect the health of employees, including annual assessment of adequate respirator fit. Prior to the COVID-19 pandemic, N95 respirators were rarely used in Philadelphia skilled-nursing facilities (SNFs), and many facilities did not have programs in place or materials to fit test their staff. Methods: The Philadelphia Department of Public Health’s (PDPH) Healthcare Associated Infections/Antimicrobial Resistance (HAI/AR) Program designed and pilot-tested 1.5-hour “train-the-trainer” sessions on OSHA-compliant fit-testing requirements and qualitative procedures. This training was offered to all 47 SNFs beginning May 2021. Training covered the role N95 respirators play in healthcare, proper donning and doffing, OSHA training requirements, medical clearances, record keeping, fit-testing procedure, and demonstrated competency to perform fit testing. Resources that were provided after training included templates of a respiratory protection policy for SNFs, a fit-test record, the OSHA medical clearance form, and a competency checklist. This bundle was designed to help SNFs establish self-sustaining respiratory protection programs. Post-training evaluations were administered on a 6-point Likert scale as well as qualitative, open-ended questions to evaluate the overall quality and effectiveness of the training session. Results: In total, 50 employees (clinical and nonclinical) from 13 Philadelphia SNFs received N95 fit-test training from June through December 2021. The average rating for the training overall was very high (5.9 of 6 points). On average, participants strongly agreed that content presented was directly applicable to their work (5.9 of 6 points), and most strongly agreed that information they learned would alter practices and procedures (5.79 of 6 points). When asked qualitatively what the participant would do differently in practice as a result of the training, the most frequent responses were fit test staff (58%) and educate staff on proper N95 use (60%). Conclusions: The PDPH HAI/AR program created a successful pilot fit-test training program for SNFs, demonstrated by program enrollment and high ratings by participants. This relatively low-cost intervention has provided tools to enhance respiratory protection during the COVID-19 pandemic and has increases the capacity of SNFs to provide essential services for their staff and residents. The PDPH will continue to offer these training sessions to SNFs, with plans to expand to other care settings, such as inpatient behavioral health facilities, outpatient clinics, and emergency medical services.

Funding: Funded by the CDC ELC Project Firstline

Disclosures: None

Background: SARS-CoV-2 N95 mask contamination in healthcare providers (HCPs) treating patients with COVID-19 is poorly understood. Method: We performed a prospective observational study of HCP N95 respirator SARS-CoV-2 contamination during aerosol-generating procedures (AGPs) on SARS-CoV-2–positive patients housed in a COVID-19–specific unit at an academic medical center. Medical masks were used as surrogates for N95 respirators to avoid waste and were worn on top of HCP N95 respirators during study AGPs. Study masks were provided to HCPs while donning PPE and were retrieved during doffing. Additionally, during doffing, face shields were swabbed with Floq swabs premoistened with viral transport media (VTM) prior to disinfection. Medical masks were cut into 9 position-based pieces, placed in VTM, vortexed, and centrifuged (Fig. 1). RNA extraction and RT-PCR were completed on all samples. RT-PCR–positive samples underwent cell culture infection to detect cytopathic effects (CPE). Contamination was characterized by mask location and front and back of face shields. Patient COVID-19 symptoms were collected from routine clinical documentation. Study HCPs completed HCP-role–specific routine care (eg, assessing, administering medications, and maintaining oxygen supplementation) while in patient rooms and were observed by study team members. Results: We enrolled 31 HCPs between September and December 2021. HCP and patient characteristics are presented in Table 1. In total, 330 individual samples were obtained from 31 masks and 26 face shields among 12 patient rooms. Of the 330 samples, 0 samples were positive for SARS-CoV-2 via RT-PCR. Positive controls were successfully performed in the laboratory setting to confirm that the virus was recoverable using these methods. Notably, all samples were collected from HCPs caring for COVID-19 patients on high-flow, high-humidity Optiflow (AGP), with an average of 960 seconds (IQR, 525–1,680) spent in each room. In addition to Optiflow and routine care, study speech pathologists completed an additional AGP of fiberoptic endoscopic evaluation of swallowing. Notably, 29 (94%) of 31 study HCP had physical contact with their patient. Conclusions: Overall, mask contamination in HCPs treating patients with COVID-19 undergoing AGPs was not detectable while wearing face shields, despite patient contact and performing AGP.

Funding: None

Disclosures: None

Background: Many healthcare facilities have faced the decision of conducting point prevalence testing (PPT) of healthcare workers (HCW) during COVID-19 outbreaks. As a containment strategy, PPT can identify asymptomatic or presymptomatic cases for isolation. It is less clear how useful testing asymptomatic HCW is in understanding the spread and possible routes of transmission in an outbreak. This study investigated HCW cases identified through PPT during acute-care outbreaks in Fraser Health (FH), British Columbia, incorporating both epidemiological and whole-genome sequencing (WGS) data to determine their epidemiological source. Methods: This study was a retrospective review of cases associated with COVID-19 acute-care outbreaks in FH occurring between December 2020 and June 2021, when most of these infections were of the alpha and gamma lineages. All patients and HCWs with a positive COVID-19 test and epidemiologically linked to the outbreaks were included in the study. WGS results supported determination of epidemiological source for cases. The proportion of patient and HCW cases related to the outbreak was compared. All analyses were conducted using SAS version 4.3 software with the PROC GLM package. Results: Between December 2020 and June 2021, 49 acute-care COVID-19 outbreaks were declared. Point-prevalence testing of HCWs, in addition to routine patient PPT, was conducted in 28 outbreaks (57%), with 2,167 eligible HCWs (63%) tested. Testing identified 14 previously unknown HCW cases, representing 12.96% of all HCW cases epidemiologically linked to the outbreaks. None of these HCWs were determined to be the index case for their associated outbreak. There was no statistically significant difference between HCWs and patients regarding WGS failure rate, and all failed samples were removed from further analysis. Patients were 3.8 times as likely as HCWs to present as symptomatic when testing positive. HCWs were 2.2 times as likely as patients to have WGS results unrelated to the outbreak. Discussion: Although point-prevalence testing of HCW identified previously unknown cases, these cases were more likely than patients to be unrelated to the outbreak and therefore less useful in understanding the epidemiology of the outbreak. It is difficult to determine whether HCW PPT was effective at preventing transmission, especially with robust infection prevention measures already in place. Patients were more likely than HCWs to present as asymptomatic, however this may be due to the attribution of symptoms to other conditions. Conclusions: Point prevalence testing of HCWs during COVID-19 outbreaks may assist with preventing transmission but is less likely to contribute meaningful information to the investigation.

Funding: None

Disclosures: None

Background: Transmission of SARS-CoV-2 in acute-care settings affects patients, healthcare workers, and the already-burdened healthcare system. An analysis of risk factors associated with outbreak severity was conducted to inform prevention strategies. Methods: This study was a cross-sectional analysis of COVID-19 outbreaks at Fraser Health (FH) acute-care sites between March 2020 and March 2021. Outbreak severity measures included COVID-19 attack rate, outbreak duration, and 30-day case mortality. Covariates at patient, outbreak, unit level, and facility level were included (Table 1). Generalized linear models with generalized estimation equations were used for all outcome measures, with outbreak duration and 30-day case mortality using multivariate negative binomial distributions, and attack rate using Gaussian distribution. A P value of .05 indicated statistical significance. Analyses

were performed using SAS version 3.8 software, R version 4.1.0 software, and Stata version 16.0 software. Results: Between March 2020 and March 2021, 54 COVID-19 outbreaks were declared in FH acute-care sites involving 455 SARS-CoV-2–positive patients. The average outbreak duration was 23 days, the average attack rate was 28%, and the average 30-day all-cause mortality per outbreak was 2 deaths. The results of the full models are shown in Table 1. Discussion: We identified an inverse relationship between increased hand hygiene compliance during outbreaks and all 3 severity measures. Paradoxically, hand hygiene rates in the year prior to the pandemic were positively associated with duration and mortality. Increased unit age was also associated with increases in each of the severity measures. Comorbidity total factor was correlated with outbreak attack rate and duration, demonstrating the importance of individual patient characteristics in an outbreak. Conclusions: Our findings highlight the importance of hand hygiene practices during an outbreak. Additionally, it is important to understand the difficulties faced by older facilities, many of which face infrastructural challenges. This study reinforces the need to incorporate infection control standards into healthcare planning and construction.

Funding: None

Disclosures: None

Background: The National University Hospital (NUH) is a 1,200 bed tertiary-care hospital with no documented nosocomial transmission of COVID-19 among patients for the first year and a half of the pandemic, despite 65% of the patients being housed in 4- to 8-bedded open cubicles with shared bathrooms. However, this arrangement changed in late September 2021 with large community clusters including in healthcare institutions nationally associated with the spread of the δ (delta) variant of SARS-CoV-2. We conducted a retrospective review of hospital epidemiology data to determine risk factors for SARS-COV-2 transmission during this period. Methods: Index patients were defined as the first patient in an open cubicle with a confirmed positive SARS-CoV-2 PCR test. Contacts were defined as being in the same cubicle as a patient before isolation from 2 days before symptom onset, up to 7 days from positive test if asymptomatic. Clinical and patient movement data were obtained manually from routine clinical records. Proximity of the contact from the index was classified as within, or more than, 2 m away, according to the prevailing definition from the Singapore Ministry of Health. A univariate analysis was performed to identify risk factors for nosocomial acquisition of SARS-CoV-2. The analysis was deemed exempt from ethics review (reference no. NHG-DSRB-2021/01026). Results: From October 1 to November 30, 2021, 30 index cases occurred in open cubicles identified (median, 9 days after admission; IQR, 19 days). Contact tracing yielded 211 contacts, of whom 10 (4.7%) were infected. Linear regression analysis found the duration of contact for each hour spent in the same room as the index case was the only statistically significant risk variable for contracting COVID-19, with an odds ratio 1.02 (Table 1). Conclusions: Patients in open cubicles are at risk for nosocomial transmission of COVID-19 and other infections. The duration of contact appeared to be more important than vaccination status of index or ward ventilation status. Larger multicentered studies are needed to validate this finding, which has significant implications for infection prevention strategies and pandemic planning.

Funding: None

Disclosures: None

Background: The Cleveland Clinic Main Campus is a multispecialty academic medical center with 1,200 adult patient beds, 58% of which are double occupancy. Our facility relies on double-occupancy rooms to provide needed care during the COVID-19 pandemic. Inherently, double occupancy poses a greater risk of exposure to SARS-CoV-2 despite mitigation efforts. We investigated the incidence of postexposure SARS-CoV-2 infection in double-occupancy rooms and evaluated risk factors for viral transmission. Methods: Early in the observation period patients were tested for SARS-CoV-2 based on clinical suspicion. By June 2020, all admitted patients were tested. Symptomatic patients were admitted with pending tests under transmission-based precautions. Asymptomatic patients were managed with standard precautions including patients admitted to double-occupancy rooms. A double-occupancy exposure event was defined as an uninfected patient sharing a room with a patient positive for SARS-CoV-2. All patient exposures were tracked and evaluated by the infection prevention (IP) team. The IP prospective review of source patients included determination of lowest cycle threshold (Ct) value of first COVID-19 test, and whether their infection was hospital or community onset. Review of exposed patients included sex, age, and exposure time (in hours) to the source patient. Postexposure infection was defined as a positive test for SARS-CoV-2 in the exposed population within 14 days of the defined exposure event. We fit a multivariable logistic regression model to estimate the effect of exposure time on the odds of postexposure infection in susceptible roommates. Results: From March 15 to December 15, 2020, 172 susceptible patients were exposed to a roommate with COVID-19. Also, 28 exposed patients met our definition for postexposure infection (attack rate, 16%). The frequency of postexposure infection was higher in patients for whom the source was hospital-onset versus community-onset disease (25% vs 10%; P = .01) and when the source patient’s Ct value was below the median value of 21.1 (26% vs 11% p Conclusions: We identified a postexposure infection attack rate of 16% for double-occupancy patients in the first 9 months of the pandemic. Time exposed to source patient was significantly associated with infection. Our experience demonstrates the potential benefit of asymptomatic admission testing with expedited turnaround time to mitigate viral transmission between patients in double-occupancy rooms.

Funding: None

Disclosures: None

Background: Face masks are a major tool for reducing the spread of COVID-19 has been the use of face masks because (1) they protect the wearer from aerosol laden virus in the environment and (2) they reduce aerosol emissions to the environment from infected individuals. Methods that quantify the fitted mask filtration efficiency for protection of the wearer are well established (eg, Sickbert-Bennett et al, JAMA Intern Med 2020;180:1607). In contrast, current methods for assessing face-mask containment efficiency are generally semiquantitative and rely on measurement of a very low concentration of aerosols emitted from a healthy or infected human, or the use of mannequins in which a high concentration of surrogate aerosols can be introduced inside the mask. Methods: Expanding on our standard methods used for fitted face-mask filtration efficiency, we designed a small-volume, low-ventilation chamber to accommodate a seated study participant. The study participant wore a ported face mask to allow introduction of a stream of 0.05 μm NaCl particles at a constant concentration (TSI 8026 particle generator) into the mask space. The ambient chamber concentration was continuously measured by a TSI 3775 condensation particle counter sampling 2 feet (~2 m) in front of the participant’s head over a series of three 3-minute periods: (1) resting, (2) reading out loud, and (3) repeated forceful coughing (2 × 10 coughs) (~450 L/min peak flows). Figure 1 shows a raw data sample for the coughing procedure. Containment efficiency (%) for each mask and procedure were determined as 100 × (1 – the average of all 1 − second ambient concentration values between 30 and 180 seconds divided by the same for the “no mask” condition). Results: Table 1 shows the average % containment efficiency for 2 study days with each mask or procedure in an adult male. The 2-ear-loop masks (KN95 and procedure) tested during coughing had the greatest reduction in % containment efficiency compared to that during resting breathing, likely owing to a decreased mask fit with transient pressure increase inside the mask associated with the coughs. The N95 was least affected by the introduction of reading and/or coughing, maintaining near 95% containment efficiency throughout. Conclusions: Our preliminary data on fitted containment efficiency of masks under different conditions suggest that the fitted containment efficiency closely mimics their performance for personal protection. This information that may aid in providing optimum source control in indoor environments.

Funding: None

Disclosures: None

Background: As of January 2022, more than 57 million cases of COVID-19 have been reported in the United States. Three primary COVID-19 vaccines are widely available: Pfizer (BNT162b2), Moderna (mRNA-1273), and Johnson & Johnson’s-Janssen (JNJ-78436735). The vaccines are effective but do not prevent all infections. We investigated trends in type of vaccine receipt, demographic characteristics, and disease outcomes in COVID-19 breakthrough infections among hospitalized patients. Methods: A breakthrough case is defined as the detection of SARS-CoV-2 ≥14 days after completion of all doses of an FDA-authorized COVID-19 vaccine. An electronic medical record report in EPIC EHR software identified 85 fully vaccinated patients with a documented positive SARS-CoV-2 result between February and September 2021 at 2 hospitals in southeastern Michigan. Demographic information and hospitalization characteristics, including length of stay and oxygen requirements, were collected from the report. Patients were classified into disease severity categories: nonsevere, severe, or critical. A case was considered severe if the patient’s oxygen saturation level (SpO2) was ≤94% on room air or if the patient required supplemental oxygen. Illness was considered critical if the patient developed respiratory failure, including mechanical ventilation or extracorporeal membrane oxygenation. All other cases were classified as nonsevere. Cycle threshold (Ct) values, the number of PCR cycles required to reach a threshold of SARS-CoV-2 genomic material, were collected from the hospital microbiology lab. Results: We identified 85 breakthrough infections (Fig. 1). The average patient age was 69.9±15.7 years, and 44 (51.8%) were female. Severe disease was most common (n = 73, 85.9%) followed by nonsevere disease (n = 7, 8.24%), and 9 patients (10.6%) in this cohort died. Most patients received either the Moderna (n = 35, 41.2%) or Pfizer (n = 38, 44.7%) vaccines. Pfizer vaccine receipt was most common among patients with severe illness (n = 33 of 73, 45.2%), and Moderna vaccine receipt was most common among patients with critical illness (n = 4 of 5, 80.0%). Average time from last vaccination to positive test was longest among Moderna vaccine recipients (181.9±43.1 days) and shortest among J&J vaccine recipients (91.0±61.1 days). The average Ct value was 23.8±7.5 and ranged from 13.0 to 41.3. There were no appreciable differences in the average Ct value by vaccine manufacturer. Conclusions: Breakthrough infections among hospitalized patients were uncommon, but incidence increased with time after vaccine receipt in all vaccines. Further study is needed to examine differences and severity in breakthrough infections by vaccine type and in individuals who completed booster vaccines.

Funding: None

Disclosures: None

Background: Vaccination against COVID-19 has demonstrated high efficacy in preventing illness severe enough to result in hospitalization. Despite these data, universal vaccine adoption by different population groups, including hospital employees, has been a challenging public health task. Vaccine-associated adverse events, the novelty of the vaccines, and the absence of long-term follow-up data have been reported as major contributors to COVID-19 vaccines mistrust. We sought to quantify postvaccination adverse events, to assess their correlation with unvaccinated status, and to evaluate other factors contributing to COVID-19 vaccination hesitancy. Methods: In a 240-bed community hospital located in a metropolitan area in the United States, we conducted a voluntary and anonymous online survey among contracted employees between September and November 2021. The study protocol was approved by the institutional review board at our facility. Results: Of all 185 responders, 143 (77%) were female, 95 (51%) were aged <51 years, and 146 were White (79%). Most (n = 100, 54%) reported no past medical history. Most common comorbidities included heart disease (n = 45, 24%), diabetes (n = 20, 11%), and chronic lung disease (n = 17, 9%). Among those surveyed, 178 were vaccinated either fully (n = 172, 93%) or partially (n = 6, 3%), and 7 (4%) were unvaccinated. Moderna was the most common vaccine received (n = 152, 85%). Those who received a 2-dose series reported experiencing more adverse events after the second dose than after the first dose (710 vs 451) of either Moderna or Pfizer vaccine. Adverse events included pain at the injection site (n = 257, 22%), fatigue (n = 178, 15%), chills (n = 133, 11%), muscle pain (n = 120, 10%), and headache (n = 117, 10%). Also, 2 responders reported omitting the second dose due to the severity of symptoms after the first dose of both Moderna and Pfizer vaccines. Concern for safety (n = 5, 71%) was the leading reason for vaccine refusal among unvaccinated followed by concern for efficacy (n = 3, 43%), lack of trust in government promoting vaccination (n = 3, 43%), religious reasons (n = 2, 28%), and immunity due to prior COVID-19 (n = 2, 28%). In addition, 3 responders reported intent to be vaccinated in the future. Conclusions: Most of the responders reported at least 1 adverse event related to COVID-19 vaccination. No severe adverse events were reported; however, a high prevalence of self-limited postvaccination adverse events might be misinterpreted as a concern for vaccine safety, as seen among surveyed unvaccinated individuals in our cohort. Targeted education is needed to limit knowledge gaps and address existing cognitive biases in COVID-19 vaccination among hospital employees.

Funding: None

Disclosures: None

Background: Postacute sequelae of SARS-CoV-2 (PASC) include fatigue, dyspnea, anxiety, and cognitive impairment. Few studies have explored the prevalence or presentation of PASC among nursing home (NH) residents. Method: A case–control study was conducted at 1 NH in Michigan in December 2021. Cases were defined as residents with SARS-CoV-2 infection between November 2, 2020, and October 8, 2021. Controls lived at the same NH during this interval and never tested positive for SARS CoV-2. Patient characteristics were compared between cases and controls using the Fisher exact test and Wilcoxon rank-sum test. Primary outcomes were functional decline, cognition, and adverse health outcomes. Outcomes were assessed by comparing measures on last observation to observations before COVID-19 diagnosis (cases) or to earliest observation (controls). Multivariable logistic regression assessed correlation between COVID-19 diagnosis and outcomes. Results: In total, 152 residents were identified for inclusion (147 included in final analyses, 76 cases, 71 controls); 5 were excluded due to insufficient data. We collected the following resident characteristics: 66% were aged ≥80 years; 73% were female; 95% were non-Hispanic white; 82% were long-stay residents; median of 3 comorbidities (IQR, 2–4). The mean number of follow-up observations was 2.60 (SD, 1.25). No significant differences in population characteristics were detected between cases and controls. Moreover, 106 patients (46 cases and 60 controls) had at least 1 follow-up visit and were thus included in the analyses to evaluate long-term outcomes. Among them, cases experienced significant declines in completing transfers (OR 5.65, p Conclusions: Nursing home residents with COVID-19 are more likely to enter hospice and have a higher mortality rate in the year following infection. Survivors experience significant functional decline in basic activities of daily living, specifically in the ability to transfer and dress. Larger studies are needed to further characterize our findings and to design interventions that can help overcome these long-term sequelae from COVID-19.

Funding: None

Disclosures: None