Risk factors for hospital-acquired infection during the SARS-CoV-2 pandemic.

OBJECTIVE: To evaluate risk factors for hospital-acquired infection (HAI) in patients during the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) pandemic, including historical and concurrent cohorts. DESIGN: Retrospective cohort. SETTING: Three Missouri hospitals, data from 1st January 2017 to 30th September 2020. PARTICIPANTS: Patients aged ≥18 years and admitted for ≥48 h. METHODS: Univariate and multi-variate Cox proportional hazards models incorporating the competing risk of death were used to determine risk factors for HAI. A-priori sensitivity analyses were performed to assess the robustness of the urine-, blood- and respiratory-culture-based HAI definition. RESULTS: The cohort included 254,792 admissions, with 7147 (2.8%) HAIs (1661 blood, 3407 urine, 2626 respiratory). Patients with SARS-CoV-2 had increased risk of HAI (adjusted hazards ratio 1.65, 95% confidence interval 1.38-1.96), and SARS-CoV-2 infection was one of the strongest risk factors for development of HAI. Other risk factors for HAI included certain admitting services, chronic comorbidities, intensive care unit stay during index admission, extremes of body mass index, hospital, and selected medications. Factors associated with lower risk of HAI included year of admission (declined over the course of the study), admitting service and medications. Risk factors for HAI were similar in sensitivity analyses restricted to patients with diagnostic codes for pneumonia/upper respiratory infection and urinary tract infection. CONCLUSIONS: SARS-CoV-2 was associated with significantly increased risk of HAI.

An Epidemiologic and Racial Disparity Evaluation of Continuous Infusion Sedation in the Icu

INTRODUCTION: Sedative medications are frequently utilized to relieve stress and anxiety, control agitation, and/or improve ventilator synchrony in patients requiring mechanical ventilation. However, many sedative agents have the potential to cause short- and long-term harm. Observational data have described associations between race and negative outcomes in patients with sepsis. It is unclear if race is associated with sedation administration practices. The purpose of this study was to investigate whether Black race is associated with differences in exposure to continuous infusion sedatives compared to White race. METHOD(S): A single-center, retrospective, cohort study was performed in adult patients admitted to the medical or surgical intensive care unit (ICU) with community-onset sepsis between June 1, 2018 and June 1, 2021. Data were collected from the electronic health record. The primary outcome was the cumulative dose of fentanyl received by continuous infusion between day 1-14 of ICU admission. The cumulative doses of other continuous infusion sedatives were described as secondary outcomes. A linear regression model assessed the association of race and the primary outcome while controlling for relevant confounding variables. RESULT(S): 772 patients were included. Black patients had a higher median modified APACHE II score (18 vs. 17, p< 0.001), Charlson Comorbidity Index (4 vs. 3, p=0.003), COVID-19 diagnosis incidence (18.5% vs. 7.1%, p< 0.001), and renal disease incidence (44.9% vs. 27.8%, p< 0.001) at baseline. No difference was observed in the amount of fentanyl received by continuous infusion between day 1-14 of ICU admission between Black and White patients (7085 mcg vs. 6426 mcg, p=0.37). Similarly, no association between Black race and the total amount of propofol (p=0.65), dexmedetomidine (p=0.67), or midazolam (p=0.08) via continuous infusion was observed. A linear regression model observed no significant association between race and the primary outcome. CONCLUSION(S): Black race was not associated with increased exposure to continuous infusion sedatives compared to White race. This is the first known study to investigate racial disparities in sedation strategies in mechanically ventilated adults in the ICU with sepsis.

Novel EHR-Derived SARS-CoV-2 Pneumonia Subphenotypes Show Divergent Clinical Outcomes with Differential Responses to Corticosteroids

Introduction: In COVID-19-related acute respiratory distress syndrome (ARDS), two distinct subphenotypes have been identified with differential outcomes and responses to corticosteroid therapy. We aimed to evaluate (1) whether clinical data can identify subgroups in a broader group of patients with SARS-CoV-2 pneumonia and (2) the extent to which corticosteroids demonstrate heterogeneity of treatment effect across such subgroups. Methods: We retrospectively studied all SARS-CoV-2 patients hospitalized for >24 hours and requiring oxygen support across 11 BJC HealthCare hospitals from June-December 2020. We excluded the initial surge (March-May 2020), as clinical care was heterogeneous and corticosteroid use low during this period. Using prespecified routinely-collected vital sign and laboratory indicator variables, we sought distinct clinical subphenotypes of SARS-CoV-2 pneumonia through latent class analysis (LCA). Across LCA subphenotypes, we evaluated the relationship between corticosteroid treatment and patient outcomes. We used multivariable logistic regression (dependent variable = composite of death/hospice) to explore treatment interaction between corticosteroid exposure and LCA subphenotype, adjusting for age and maximal SOFA score within 24 hours of admission as surrogates for indication. Results: The 3-class LCA model best fit the 1845-patient cohort (p=0.007). Class-1 (n=1456) had mean standardized values of all indicator variables near zero;Class-2 (n=235) manifested profound isolated hypoxemia;and Class-3 (n=154) displayed multiorgan failure, shock, and neutrophilia (Figure-1A). Despite representing <25% of the cohort, Classes 2 (n=109, 46%) and 3 (n=70, 46%) comprised >50% of the primary outcome (vs Class-1: n=151, 10%;p<0.001). Corticosteroids were more frequently administered in Class-2 (n=215, 91%) than in Class-1 (n=1071, 74%) or Class-3 (n=110, 71%, p<0.001;Figure-1B). Adjusted analyses demonstrated interaction between LCA class and corticosteroid treatment for the primary outcome (Class-1, p=0.003;Class-3, p=0.002). Corticosteroids were associated with increased adjusted odds for the primary outcome in Class-1 (aOR 2.11, 95% CI 1.33-3.50, p=0.002) and decreased adjusted odds for the primary outcome in Class-3 (aOR 0.44, 95% CI 0.19-0.98, p=0.048). Class-2 showed no outcome differences between the corticosteroid and noncorticosteroid groups (aOR 1.03, 95% CI 0.38-2.81, p=0.95). Conclusions: Three distinct subphenotypes of SARS-CoV-2 pneumonia demonstrate different clinical outcomes and corticosteroid response. No clear effect was seen among patients with isolated hypoxemia, whereas those with multiorgan failure appeared to benefit. Our findings suggest that among hospitalised patients in our healthcare system, corticosteroid therapy was associated with increased risk of harm. Prospective studies are needed to evaluate the efficacy of corticosteroids in SARS-CoV-2 pneumonia in predictively-enriched trials.

Using a Design Thinking Framework to Create User-Centered ICU-Ward Handoffs from Intensive Care Resident Teams to Academic Hospitalist Teams

RATIONALE: The discharge of patients from the ICU to the hospital ward is a high-risk period. Previously, we used Human-Centered Design - an iterative, collaborative process for user-focused solutions - to develop a prototype structured ICU-ward transfer communication tool for clinicians by focusing on resident-to-resident communication. We describe here an approach to rapid prototyping with the specific focus on meeting the needs academic Hospitalists. METHODS: In winter 2021, we conducted a voluntary focus group of Hospitalists at the University of Chicago to ensure that a recently developed ICU- ward transfer tool created through prior focus groups with resident also met the needs of hospital medicine clinicians. The focus group was conducted via videoconference due to the COVID-19 pandemic. With participants' consent, we recorded and transcribed the focus group discussions conducted over the Zoom platform. Following transcription, qualitative analysis on the transcript was used to look for common themes and concerns. Coding was performed using both a theory driven (deductive) and data driven (inductive) approach. RESULTS: The focus group identified several main themes around the role, content, and workflows related to an ideal ICU-wards handoff tool: (1) how the tool can communicate the patient course clearly through multiple verbal hand-offs;(2) that the ICU-ward handoff process must balance thoroughness, usability, and reducing the propagation of copy/paste errors;and (3) that design and implementation should provide hand-off communication between providers and patient families. Under these themes, participants identified specific attributes of an ideal handoff tool (Table 1), which coalesced around 3 main goals: (1) to synthesize key details and communicate the ICU team's thought process and follow up tasks (summarized in Table 1A);(2) to integrate the new tool into the EHR that minimizes documentation errors and communicates transfer status in the EHR to help handoffs between shifts (1B);and (3) to serve as a standardized process to ensure communication is bridged between teams and patient families (1C). CONCLUSIONS: Participants valued organization of handoff information, succinct information, EHR integration, and standardization of family communication when modifying a framework. One notable difference is our Hospitalist group focused on brevity while our resident group preferred a longer detailed course. Further work will need to be performed to find balance the needs between these two groups to ensure efficient handoffs.

Design thinking to create user-centered icu-ward handoffs at three academic hospitals

RATIONALE: Patient transfers from the intensive care unit (ICU) to the hospital ward are a vulnerable time for patients recovering from critical illness. Our prior work showed that ICU-ward transfers vary widely across sites. Previously, we used Human-Centered Design-an iterative, collaborative process for user-focused solutions-to develop a prototype structured ICU-ward transfer communication tool for clinicians at three academic Internal Medicine residency programs. We describe here an iterative approach to rapid prototyping and updating of this novel tool for integration into the electronic health record (EHR), based on structured focus groups using Design Thinking methodology. METHODS: In spring 2020, we conducted voluntary focus groups of PGY-2 and PGY-3 Internal Medicine Residents to co-create new iterations of an ICUward transfer tool created through prior focus groups. Each site conducted 2 focus groups with 4-10 participants per session. All focus groups were conducted via videoconference due to the COVID-19 pandemic. We used participant feedback to iteratively modify the ICU-ward transfer tool towards an ideal state. With participants' consent, we recorded and anonymously transcribed the focus group discussions conducted over the Zoom platform. We performed qualitative inquiry on transcripts for coding and thematic content analysis. Coding was performed independently by two coders (LS, BG) using both a theory-driven (deductive) and data-driven (inductive) approach;disparate coding was reconciled via in-depth discussion between coders. RESULTS: Focus groups identified several main themes around the role, content, and workflows related to an ideal ICU-wards handoff tool: (1) how the tool can best serve the needs of ICU and wards teams;(2) that the ICU-ward handoff process must balance thoroughness and usability;and (3) that design and implementation must consider user experience and training in best practices. Under these themes, participants identified specific attributes of an ideal handoff tool (Table 1), which coalesced around 3 main goals: (1) to synthesize key details and communicate the ICU team's thought process (summarized in Table 1A);(2) to integrate the new tool into the EHR within a customizable note template that minimizes click burden and redundancy (1B);and (3) to serve as a standardized outline to reduce errors of omission in written and verbal handoffs processes (1C). CONCLUSIONS: Participants valued organization of handoff information, EHR integration and user experience, and standardization and adaptability of workflows when modifying a framework based on prior resident user input to create a more user-friendly ICU-ward transfer tool using a Design Thinking Methodology.

Predictors of hospital outcomes and clinical subphenotypes differ between COVID-19 and influenza pneumonia

Introduction: Pneumonia due to SARS-CoV-2 (Coronavirus Disease 2019, COVID-19) has frequently been compared to other viral pneumonias, including influenza. While some data suggest significant differences in biological responses, dissimilarities in the clinical course and characteristics between SARS-COV-2 and influenza pneumonia remain unknown. We evaluated differences in clinical predictors of outcomes and early clinical subphenotypes in COVID-19 and influenza pneumonia. Methods: We performed a retrospective cohort study of all patients hospitalized for > 24 hours, requiring oxygen support, at Barnes-Jewish Hospital with COVID-19 (March-July 2020) or influenza (Jan 2012-Dec 2018). In-hospital mortality or hospice discharge was the primary outcome. First, supervised machine learning classifier models (XGBoost) were trained using bootstrap replications of each viral cohort to predict the primary outcome. 28 candidate predictor variables among the most extreme vital signs and laboratory values within 24 hours of hospitalization were preselected, excluding highly correlated variables. We compared each model's internal discrimination to its performance in the alternate cohort and evaluated differences in variable importance between the two viral pneumonia models. Next, we evaluated differences in clinical subphenotypes in two ways: 1) a previously-validated algorithm to group patients into four distinct subphenotypes based on temperature trajectories within 72 hours of hospitalization;2) latent class analysis (LCA) to identify unmeasured subgroups within each viral cohort based on the predictor variables described above. In both analyses, we compared frequency of subphenotype membership and each subphenotype's primary outcome between viral cohorts. Results: We evaluated 321 unique hospitalizations with COVID-19 and 535 with influenza. The primary outcome was experienced in 23% and 9.5% of patients, respectively. Influenza predictor model discriminated outcomes worse in COVID-19 than on internal evaluation (Panel A), suggesting prognostic variables differ between the viral pneumonias. Only one of the top five contributory variables was shared between the two models (Panel B). Prevalences of temperature trajectory subphenotype also differed significantly between viral pneumonias. All COVID-19 temperature trajectory subphenotypes experienced the primary outcome more frequently than their influenza counterparts (Panel C). LCA identified two distinct classes in each cohort, with each viral pneumonia's minority class experiencing worse outcomes than the majority class. Of each model's top 5 classdefining variables, only 2 were shared (Panel D). Conclusions: COVID-19 and influenza pneumonia differ markedly in predictors of outcome and in clinical subphenotypes. These findings emphasize observable pathogen-specific differential responses in viral pneumonias and suggest that distinct management approaches should be investigated for these diseases. (Table Presented).