To assess trends in antibiotic prescribing for patients hospitalized with COVID-19 with and without sepsis.

Retrospective cohort study using electronic health record (EHR) data.

Five hospitals in eastern Massachusetts.

Adults (≥18 years) hospitalized with community-onset SARS-CoV-2 infections between March 2020 and November 2022.

We assessed quarterly trends in the use of prolonged initial antibiotic therapy (≥4 antibiotic days within one week of admission, including discharge antibiotics) amongst COVID-19 patients with and without sepsis, defined using clinical signs of organ dysfunction before hospital day 3. Poisson regression models were used to adjust for baseline characteristics and severity of illness.

Of 431,017 hospitalizations in the study period, 21,563 (5.0%) had community-onset COVID-19. 4,769/21,563 (20.5%) presented with sepsis. Prolonged antibiotics were prescribed in 2,323/4,769 (48.7%) COVID-19 patients with sepsis and 2,866/16,794 (17.1%) without sepsis despite low rates of positive bacterial cultures on admission (15.0% vs 6.3%, respectively). Quarterly rates of prolonged antibiotics declined between the first and second pandemic quarters for both sepsis (66.8% to 43.9%) and no-sepsis (31.8% to 24.4%) groups. However, there was no significant change thereafter through November 2022 in either group (quarterly aORs 1.02, 95% CI 0.99–1.05 and 1.01, 95% CI 0.99–1.03, respectively).

Prolonged antibiotics were common in hospitalized COVID-19 patients with and without sepsis during the first 33 months of the pandemic despite low rates of proven bacterial infection. Decreases in antibiotic utilization occurred primarily between the first and second pandemic quarter with no further reduction thereafter.

To characterize the relationship between chlorhexidine gluconate (CHG) skin concentration and skin microbial colonization.

Serial cross-sectional study.

Adult patients in medical intensive care units (ICUs) from 7 hospitals; from 1 hospital, additional patients colonized with carbapenemase-producing Enterobacterales (CPE) from both ICU and non-ICU settings. All hospitals performed routine CHG bathing in the ICU.

Skin swab samples were collected from adjacent areas of the neck, axilla, and inguinal region for microbial culture and CHG skin concentration measurement using a semiquantitative colorimetric assay. We used linear mixed effects multilevel models to analyze the relationship between CHG concentration and microbial detection. We explored threshold effects using additional models.

We collected samples from 736 of 759 (97%) eligible ICU patients and 68 patients colonized with CPE. On skin, gram-positive bacteria were cultured most frequently (93% of patients), followed by Candida species (26%) and gram-negative bacteria (20%). The adjusted odds of microbial recovery for every twofold increase in CHG skin concentration were 0.84 (95% CI, 0.80–0.87; P < .001) for gram-positive bacteria, 0.93 (95% CI, 0.89–0.98; P = .008) for Candida species, 0.96 (95% CI, 0.91–1.02; P = .17) for gram-negative bacteria, and 0.94 (95% CI, 0.84–1.06; P = .33) for CPE. A threshold CHG skin concentration for reduced microbial detection was not observed.

On a cross-sectional basis, higher CHG skin concentrations were associated with less detection of gram-positive bacteria and Candida species on the skin, but not gram-negative bacteria, including CPE. For infection prevention, targeting higher CHG skin concentrations may improve control of certain pathogens.

Many providers use severe acute respiratory coronavirus virus 2 (SARS-CoV-2) cycle thresholds (Ct values) as approximate measures of viral burden in association with other clinical data to inform decisions about treatment and isolation. We characterized temporal changes in Ct values for non–SARS-CoV-2 respiratory viruses as a first step to determine whether cycle thresholds could play a similar role in the management of non–SARS-CoV-2 respiratory viruses.

Retrospective cohort study.

Brigham and Women’s Hospital, Boston.

We retrospectively identified all adult patients with positive nasopharyngeal PCRs for influenza, respiratory syncytial virus (RSV), parainfluenza, human metapneumovirus (HMPV), rhinovirus, or adenovirus between January 2022 and March 2023. We plotted Ct distributions relative to days since symptom onset, and we assessed whether distributions varied by immunosuppression and other comorbidities.

We analyzed 1,863 positive samples: 506 influenza, 502 rhinovirus, 430 RSV, 219 HMPV, 180 parainfluenza, 26 adenovirus. Ct values were generally 25–30 on the day of symptom onset, lower over the ensuing 1–3 days, and progressively higher thereafter with Ct values ≥30 after 1 week for most viruses. Ct values were generally higher and more stable over time for rhinovirus. There was no association between immunocompromised status and median intervals from symptom onset until Ct values were ≥30.

Ct values relative to symptom onset for influenza, RSV, and other non–SARS-CoV-2 respiratory viruses generally mirror patterns seen with SARS-CoV-2. Further data on associations between Ct values and viral viability, transmissibility, host characteristics, and response to treatment for non-SARS-CoV-2 respiratory viruses are needed to determine how clinicians and infection preventionists might integrate Ct values into treatment and isolation decisions.

To evaluate the prevalence, risk factors, and clinical impact of delays in second doses of antibiotics in patients with sepsis.

Single-center, retrospective, observational study.

Large teaching hospital.

Adult patients who triggered an electronic sepsis alert in the emergency department (ED), received ≥2 doses of vancomycin or an antipseudomonal beta-lactam, and were discharged with an ICD-10 sepsis code.

We assessed the prevalence of delays in second doses of antibiotics by ≥25% of the recommended dose interval and conducted multivariate regression analyses to assess for risk factors for delays and in-hospital mortality.

The cohort included 449 patients, of whom 123 (27.4%) had delays in second doses. In-hospital death occurred in 31 patients (25.2%) in the delayed group and 71 (21.8%) in the non-delayed group (p = 0.44). On multivariate analysis, only location in a non-ED unit at the time second doses were due was associated with delays (OR 2.75, 95% CI 1.20–6.32). In the mortality model, significant risk factors included malignant tumor, respiratory infection, and elevated Sequential Organ Failure Assessment (SOFA) score but not delayed second antibiotic doses (OR 1.19, 95% CI 0.69–2.05). In a subgroup analysis, delayed second doses were associated with higher mortality in patients admitted to non-intensive care units (ICUs) (OR 4.10, 95% CI 1.32–12.79).

Over a quarter of patients with sepsis experienced delays in second doses of antibiotics. Delays in second antibiotic doses were not associated with higher mortality overall, but an association was observed among patients admitted to non-ICUs.

To assess whether measurement and feedback of chlorhexidine gluconate (CHG) skin concentrations can improve CHG bathing practice across multiple intensive care units (ICUs).

A before-and-after quality improvement study measuring patient CHG skin concentrations during 6 point-prevalence surveys (3 surveys each during baseline and intervention periods).

The study was conducted across 7 geographically diverse ICUs with routine CHG bathing.

Adult patients in the medical ICU.

CHG skin concentrations were measured at the neck, axilla, and inguinal region using a semiquantitative colorimetric assay. Aggregate unit-level CHG skin concentration measurements from the baseline period and each intervention period survey were reported back to ICU leadership, which then used routine education and quality improvement activities to improve CHG bathing practice. We used multilevel linear models to assess the impact of intervention on CHG skin concentrations.

We enrolled 681 (93%) of 736 eligible patients; 92% received a CHG bath prior to survey. At baseline, CHG skin concentrations were lowest on the neck, compared to axillary or inguinal regions (P < .001). CHG was not detected on 33% of necks, 19% of axillae, and 18% of inguinal regions (P < .001 for differences in body sites). During the intervention period, ICUs that used CHG-impregnated cloths had a 3-fold increase in patient CHG skin concentrations as compared to baseline (P < .001).

Routine CHG bathing performance in the ICU varied across multiple hospitals. Measurement and feedback of CHG skin concentrations can be an important tool to improve CHG bathing practice.

Surveillance of non–ventilator-associated hospital-acquired pneumonia (NV-HAP) is complicated by subjectivity and variability in diagnosing pneumonia. We compared a fully automatable surveillance definition using routine electronic health record data to manual determinations of NV-HAP according to surveillance criteria and clinical diagnoses.

We retrospectively applied an electronic surveillance definition for NV-HAP to all adults admitted to Veterans’ Affairs (VA) hospitals from January 1, 2015, to November 30, 2020. We randomly selected 250 hospitalizations meeting NV-HAP surveillance criteria for independent review by 2 clinicians and calculated the percent of hospitalizations with (1) clinical deterioration, (2) CDC National Healthcare Safety Network (CDC-NHSN) criteria, (3) NV-HAP according to a reviewer, (4) NV-HAP according to a treating clinician, (5) pneumonia diagnosis in discharge summary; and (6) discharge diagnosis codes for HAP. We assessed interrater reliability by calculating simple agreement and the Cohen κ (kappa).

Among 3.1 million hospitalizations, 14,023 met NV-HAP electronic surveillance criteria. Among reviewed cases, 98% had a confirmed clinical deterioration; 67% met CDC-NHSN criteria; 71% had NV-HAP according to a reviewer; 60% had NV-HAP according to a treating clinician; 49% had a discharge summary diagnosis of pneumonia; and 82% had NV-HAP according to any definition according to at least 1 reviewer. Only 8% had diagnosis codes for HAP. Interrater agreement was 75% (κ = 0.50) for CDC-NHSN criteria and 78% (κ = 0.55) for reviewer diagnosis of NV-HAP.

Electronic NV-HAP surveillance criteria correlated moderately with existing manual surveillance criteria. Reviewer variability for all manual assessments was high. Electronic surveillance using clinical data may therefore allow for more consistent and efficient surveillance with similar accuracy compared to manual assessments or diagnosis codes.

To examine the impact of commonly used case definitions for coronavirus disease 2019 (COVID-19) hospitalizations on case counts and outcomes.

Retrospective analysis of all adults hospitalized between March 1, 2020, and March 1, 2022, at 5 Massachusetts acute-care hospitals.

We applied 6 commonly used definitions of COVID-19 hospitalization: positive severe acute respiratory coronavirus virus 2 (SARS-CoV-2) polymerase chain reaction (PCR) assay within 14 days of admission, PCR plus dexamethasone administration, PCR plus remdesivir, PCR plus hypoxemia, institutional COVID-19 flag, or COVID-19 International Classification of Disease, Tenth Revision (ICD-10) codes. Outcomes included case counts and in-hospital mortality. Overall, 100 PCR-positive cases were reviewed to determine each definition’s accuracy for distinguishing primary or contributing versus incidental COVID-19 hospitalizations.

Of 306,387 hospital encounters, 15,436 (5.0%) met the PCR-based definition. COVID-19 hospitalization counts varied substantially between definitions: 4,628 (1.5% of all encounters) for PCR plus dexamethasone, 5,757 (1.9%) for PCR plus remdesivir, 11,801 (3.9%) for PCR plus hypoxemia, 15,673 (5.1%) for institutional flags, and 15,868 (5.2%) for ICD-10 codes. Definitions requiring dexamethasone, hypoxemia, or remdesivir selected sicker patients compared to PCR alone (mortality rates 12.2%, 10.7%, and 8.8% vs 8.3%, respectively). Definitions requiring PCR plus remdesivir or dexamethasone did not detect a reduction in in-hospital mortality associated with the SARS-CoV-2 Omicron variant. ICD-10 codes had the highest sensitivity (98.4%) but low specificity (39.5%) for distinguishing primary or contributing versus incidental COVID-19 hospitalizations. PCR plus dexamethasone had the highest specificity (92.1%) but low sensitivity (35.5%).

Commonly used definitions for COVID-19 hospitalizations generate variable case counts and outcomes and differentiate poorly between primary or contributing versus incidental COVID-19 hospitalizations. Surveillance definitions that better capture and delineate COVID-19–associated hospitalizations are needed.

To assess coronavirus disease 2019 (COVID-19) infection policies at leading US medical centers in the context of the initial wave of the severe acute respiratory coronavirus virus 2 (SARS-CoV-2) omicron variant.

Electronic survey study eliciting hospital policies on masking, personal protective equipment, cohorting, airborne-infection isolation rooms (AIIRs), portable HEPA filters, and patient and employee testing.

“Hospital epidemiologists from U.S. News top 20 hospitals and 10 hospitals in the CDC Prevention Epicenters program.”  As it is currently written, it implies all 30 hospitals are from the CDC Prevention Epicenters program, but that only applies to 10 hospitals.  Alternatively, we could just say “Hospital epidemiologists from 30 leading US hospitals.”

Survey results were reported using descriptive statistics.

Of 30 hospital epidemiologists surveyed, 23 (77%) completed the survey between February 15 and March 3, 2022. Among the responding hospitals, 18 (78%) used medical masks for universal masking and 5 (22%) used N95 respirators. 16 hospitals (70%) required universal eye protection. 22 hospitals (96%) used N95s for routine COVID-19 care and 1 (4%) reserved N95s for aerosol-generating procedures. 2 responding hospitals (9%) utilized dedicated COVID-19 wards; 8 (35%) used mixed COVID-19 and non–COVID-19 units; and 13 (57%) used both dedicated and mixed units. 4 hospitals (17%) used AIIRs for all COVID-19 patients, 10 (43%) prioritized AIIRs for aerosol-generating procedures, 3 (13%) used alternate risk-stratification criteria (not based on aerosol-generating procedures), and 6 (26%) did not routinely use AIIRs. 9 hospitals (39%) did not use portable HEPA filters, but 14 (61%) used them for various indications, most commonly as substitutes for AIIRs when unavailable or for specific high-risk areas or situations. 21 hospitals (91%) tested asymptomatic patients on admission, but postadmission testing strategies and preferred specimen sites varied substantially. 5 hospitals (22%) required regular testing of unvaccinated employees and 1 hospital (4%) reported mandatory weekly testing even for vaccinated employees during the SARS-CoV-2 omicron surge.

COVID-19 infection control practices in leading hospitals vary substantially. Clearer public health guidance and transparency around hospital policies may facilitate more consistent national standards.

To assess preventability of hospital-onset bacteremia and fungemia (HOB), we developed and evaluated a structured rating guide accounting for intrinsic patient and extrinsic healthcare-related risks.

HOB preventability rating guide was compared against a reference standard expert panel.

A 10-member panel of clinical experts was assembled as the standard of preventability assessment, and 2 physician reviewers applied the rating guide for comparison.

The expert panel independently rated 82 hypothetical HOB scenarios using a 6-point Likert scale collapsed into 3 categories: preventable, uncertain, or not preventable. Consensus was defined as concurrence on the same category among ≥70% experts. Scenarios without consensus were deliberated and followed by a second round of rating.

Two reviewers independently applied the rating guide to adjudicate the same 82 scenarios in 2 rounds, with interim revisions. Interrater reliability was evaluated using the κ (kappa) statistic.

Expert panel consensus criteria were met for 52 scenarios (63%) after 2 rounds.

After 2 rounds, guide-based rating matched expert panel consensus in 40 of 52 (77%) and 39 of 52 (75%) cases for reviewers 1 and 2, respectively. Agreement rates between the 2 reviewers were 84% overall (κ, 0.76; 95% confidence interval [CI], 0.64–0.88]) and 87% (κ, 0.79; 95% CI, 0.65–0.94) for the 52 scenarios with expert consensus.

Preventability ratings of HOB scenarios by 2 reviewers using a rating guide matched expert consensus in most cases with moderately high interreviewer reliability. Although diversity of expert opinions and uncertainty of preventability merit further exploration, this is a step toward standardized assessment of HOB preventability.