Hostname: page-component-6898cfcb55-cbk24 Total loading time: 0 Render date: 2026-11-12T02:58:31.557Z Has data issue: false hasContentIssue false
  • English
  • Français

Clinical characteristics associated with hospital-onset bacteremia and fungemia among cancer and transplant patients

Published online by Cambridge University Press:  23 October 2024

Kalvin C. YuOpen the ORCID record for Kalvin C. Yu[Opens in a new window],
John C. O’HoroOpen the ORCID record for John C. O’Horo[Opens in a new window],
ChinEn Ai,
Molly JungOpen the ORCID record for Molly Jung[Opens in a new window] and
Samantha BastowOpen the ORCID record for Samantha Bastow[Opens in a new window]
Kalvin C. Yu*
Affiliation:
Dept. of Medical Affairs, Becton Dickinson and Company, Franklin Lakes, NJ, USA
John C. O’Horo
Affiliation:
Mayo Clinic, Rochester, MN, USA
ChinEn Ai
Affiliation:
Dept. of Medical Affairs, Becton Dickinson and Company, Franklin Lakes, NJ, USA
Molly Jung
Affiliation:
Dept. of Medical Affairs, Becton Dickinson and Company, Franklin Lakes, NJ, USA
Samantha Bastow
Affiliation:
Dept. of Medical Affairs, Becton Dickinson and Company, Franklin Lakes, NJ, USA
*
Corresponding author: Kalvin C. Yu; Email: kalvin.yu@bd.com
Rights & Permissions [Opens in a new window]

Abstract

Objective:

This study quantified the burden of hospital-onset bacteremia and fungemia (HOB) among cancer and transplant patients compared to other patients.

Methods:

A retrospective cross-sectional study used data from 41 hospitals between October 2015 and June 2019. Hospitalizations were segmented into categories using diagnosis-related groups (DRG): myeloproliferative (MP) cancer, solid tumor cancer, transplant, and non-cancer/non-transplant (“reference group”). To quantify the association between DRG and HOB, multivariable adjusted Poisson regression models were fit. Analyses were stratified by length of stay (LOS).

Results:

Of 645,315 patients, 59% were female and the majority 41 years of age or older (76%). Hospitalizations with MP cancer and transplant demonstrated higher HOB burden compared to the reference group, regardless of LOS category. For all hospitalizations, the >30 days LOS category had a higher burden of HOB. The median time to reportable HOB was within 30 days regardless of duration of hospitalization (reference, 8 days; solid tumor cancer, 8 days; transplant, 12 days; MP cancer, 13 days).

Conclusion:

MP cancer and transplant patients had a higher burden of HOB compared to other hospitalized patients regardless of LOS. Whether these infections are preventable should be further evaluated to inform quality metrics involving reportable bacteremia and fungemia.

Information

Type
Original Article
Information
Infection Control & Hospital Epidemiology , Volume 45 , Issue 12 , December 2024, pp. 1391 - 1398
Check for updates
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of The Society for Healthcare Epidemiology of America

Introduction

Hospital-acquired infections (HAIs) are a significant driver of increased morbidity, mortality, and financial cost. Reference Yu, Jung and Ai1–Reference Liu, Spencer and Long4 HAI measures, in particular the National Healthcare Safety Network (NHSN)-defined central line-associated bloodstream infection (CLABSI) rates, are quality metrics evaluated by the Centers for Medicare and Medicaid Services (CMS) and private insurers as indicators of patient safety and can potentially affect reimbursement through CMS’ Hospital-acquired Conditions Reduction Program and the 2015 Value-Based Performance Program. Reference Rajaram, Barnard and Bilimoria5 Since 2011, CLABSIs have been included in the CMS national reporting programs but may be impacted by subjectivity and interrater variability. Reference DiGiorgio, Vinski and Bertin6 Decreasing national CLABSI rates have facilitated discussions on expanding the metric to a more encompassing measure beyond central lines to include broader sources of BSI in the form of hospital-onset bacteremia and fungemia (HOB). Reference Yu, Jung and Ai1

At the time of current writing, HOB reporting is unique in that a bacteremia or Candida species fungemia might be reportable irrespective of the original source, possibly including bacteremia and fungemia from other HAIs not specific to CLABSI. The provisional definition may also capture currently excluded but clinically significant entities, including midline-associated BSI, and peripheral intravenous (IV) associated infections. Reference Sato, Nakamura and Fujita7,Reference Chopra, Kaatz and Swaminathan8 In fact, the World Health Organization just released in May 2024 guidelines for preventing BSIs due to peripheral IVs. 9 HOB is objective, easier to electronically capture, automated, and removes the adjudication process as is currently required for catheter associated urinary tract infection (CAUTI) and CLABSI. Reference Dantes, Abbo and Anderson10,Reference Rock, Thom and Harris11 HOB is associated with significant incremental mortality, length of stay (LOS), and cost of care. Higher mortality rates among patients with noncentral line-associated BSI and $20,000 in additional costs have been reported. Reference Ridgway, Sun, Tabak, Johannes and Robicsek12 In a recent case-matched analysis, CLABSIs and non-CLABSI HOB were associated with significantly higher costs and longer LOS for both ICU and non-ICU patients; and a greater than 3.5-fold increased risk of mortality in ICU patients. Reference Yu, Jung and Ai1 Opinions from the Society for Healthcare Epidemiology of America (SHEA) Research Network suggest many hospital epidemiologists and infection prevention specialists (54%) view HOB as a measure of a hospital’s quality of care. Indeed, 29% of those surveyed reside in organizations implementing HOB testing. The majority of those surveyed (57%) favored publicly reporting HOB alone (22%) or in addition to CLABSI (35%), with 34% favoring CLABSI alone. Reference Dantes, Abbo and Anderson10 Therefore, HOB may represent a stand-alone or adjunct quality of care safety metric to aid in the improvement of infection prevention and patient outcomes.

HOB is thought to be preventable or at least partially preventable. Reference Dantes, Rock and Milstone13 Nevertheless, specific patient populations may be at a higher risk for HOB, especially those with nonmodifiable risk factors such as inherently immunocompromised patients (eg, myeloproliferative (MP) disease and cancer patients), and long-term surgical patients and/or prolonged exogenous immune suppression (eg, transplant recipients). Reference Battaglia and Hale14 Another clinical consideration based on the currently available HOB definition is that only the first positive blood culture determines HOB or community-onset bacteremia (COB) designation; therefore, subsequent or recurrent blood cultures during the HO period may not be considered for required reporting. While guidelines support more vigilant infection-control practices in these vulnerable populations, the development of HOB likely disproportionately impacts these patients. Reference Ariza-Heredia and Chemaly15,Reference Stoclin, Rotolo and Hicheri16 The current study primarily aims to quantify the burden of HOB among adult cancer (solid tumor cancer and MP cancer) and transplant (solid organ and bone marrow (BMT) recipient) patients compared to patients with other diagnosis-related groups (DRG). Secondary aims include the incidence of non-reportable HOB admissions (eg, subsequent HOB, commensal organisms, and other non-reported organisms), and describing the pathogens driving HOB infections.

Methods

This was a retrospective observational study that included data from inpatient adults age 18 years or older from 41 acute-care hospitals in the BD Insights and Research Database (Becton, Dickinson and Company, Franklin Lakes, NJ) between October 2015 and June 2019. Details of the data collection system have been previously described and include pharmacy, microbiology data and other laboratory measurements, administrative data, patient demographics, and admission, discharge, and transfer data feeds. Reference Yu, Ye and Edwards17,Reference Yu, Yamaga, Vankeepuram and Tabak18 The New England Institutional Review Board/WCG Human Subjects Research Committee (Wellesley, MA) approved the study as involving use of a limited retrospective data set for an epidemiology study and granted an exemption from consent.

Definitions

Exposure groups

Hospitalizations were categorized into four disease groups based on their corresponding DRG code description: solid tumor cancer, MP cancer, transplant (including solid organ and BMT), or non-cancer, non-transplant DRG code (here on described as the “reference”).

Hospital-onset bacteremia and fungemia: reportable and subsequent infections

Reportable HOB cases were defined by the currently available definition per the CDC (ie, a first positive blood culture and was collected in the hospital-onset period, on or after day 4 of hospitalization) for an eligible BSI organism as defined by the NHSN bloodstream pathogen list. Reference Yu, Jung and Ai1,Reference Schrank, Snyder and Leekha19,Reference Leekha, Robinson and Jacob20

“Subsequent HOB” was defined with 3 requisite qualifiers: (1) occurred after an index reportable HOB event; (2) the pathogen was not the same as the index HOB; and (3) contained an eligible BSI organism (as outlined earlier). A “non-duplicated pathogen for a subsequent HOB event” was defined with 2 mutually exclusive requirements, either: (1) a subsequent HOB event that was a different pathogen from the first (ie, Reportable) HOB or (2) a subsequent HOB event that had the same pathogen as the index HOB but occurred 30 days after the index HOB event and had a different antimicrobial susceptibility test result.

The prevalence of commensal bacteria or other pathogens (defined as not listed in either the NHSN pathogen list or commensal list) isolated in blood cultures were also evaluated in our analyses. To better understand the epidemiology of the microorganisms that are associated with HOB, species of organisms were grouped into pathogens categories based on previously published data. Reference Yu, Jung and Ai1

Other variables

Sex, age, LOS, DRG, 30-day readmissions, ICU during hospitalization, hospital cost per admission, in-hospital mortality, insurance payor, and hospital by demographics, staffed bed size, teaching status, and urban/rural location were variables collected in administrative data. ICU LOS was a derived variable from admission, discharge, and transfer data and quantified the number of days in the ICU. Severity of illness was measured by a quantified marker of risk for mortality during the same admission, the Acute Laboratory Risk of Mortality Score (ALaRMs) and was derived using laboratory data and administrative data with Agency of Healthcare Research and Quality (AHRQ)’s clinical classification software, the methods and use of which have been previously published. Reference Yu, Jung and Ai1,Reference Tabak, Sun, Nunez and Johannes21

Statistical analysis

The distribution of patient, clinical, and hospital characteristics were described using frequencies for categorical variables and medians with interquartile ranges for continuous variables overall and by DRGs.

Because hospitalizations tended to be longer in patients with MP cancer and transplant compared to other DRG groups, analyses were stratified by LOS (≤30 or >30 days). The rate of HOB was estimated per 1,000 admissions. We compared the rate of HOB by DRG using prevalence rate ratio (RR) with multivariate adjusted Poisson regression. In admissions with HOB, the association between DRG with subsequent HOB, commensal bacteria, and other pathogen infection was evaluated using frequencies and multivariable adjusted logistic regressions.

The regression models were adjusted for sex, age, ICU stay during the hospitalization, ALaRMS, payor, and hospital characteristics (staffed bed size, teaching status, and urbanicity) with hospital as random effect to account for within-cluster correlation of data. In hospitalizations with HOB, unadjusted Kaplan-Meier plots were generated to visualize the median time to reportable HOB by DRG group. The prevalence of pathogen categories at the reportable (first) HOB were estimated using frequencies by LOS and DRG. All analyses were conducted using R software version 4.1.2 software (R Foundation for Statistical Computing, Vienna, Austria) with R Studio (Boston, MA).

Results

Patient demographics

The present analysis included 645,315 hospitalizations (0.83% MP cancer, 1.61% solid tumor cancer, and 0.36% transplant, Table 1). Approximately 60% of the analytic cohort were female and the vast majority were 41 years or older (33% age 41–64 years and 43% age 65 years or older). LOS was higher in patients hospitalized for MP cancer or transplant compared to solid tumor cancer or non-cancer, non-transplant DRG. The median ALaRMS score was higher among all groups in comparison to the reference group (38.0). The median ALaRMS was highest for the transplant group (46.0), followed by solid tumor cancer (43.0), and MP cancer (41.0).

Table 1.

Patient, clinical, and hospital characteristics overall and by diagnosis related groups

Burden of HOB by LOS groups

The rate of HOB was 2.0 per 1,000 admissions in LOS ≤30 days and 82.1 per 1,000 admissions in LOS >30 days. In patients with LOS ≤30 days, the unadjusted rate was lowest in the reference group (2 per 1,000 admissions; Figure 1) and highest for patients hospitalized for transplant (22.0 per 1,000 admissions). In multivariable adjusted analyses, the risk of HOB in MP cancer was 7-fold higher (RR: 7.02, 95% CI: 5.41, 9.12) and fourfold higher in transplant (RR: 4.23; 95% CI: 3.14, 5.71) compared to the reference group. Despite the overall burden of HOB being higher in patients with longer hospitalizations, the higher risk for HOB in MP cancer and transplant vs the reference group were consistent. The risk of HOB was 2-fold higher for MP cancer and 57% higher for transplant compared to the reference group (RRMP cancer vs reference: 2.23; 95% 1.58, 3.14 and RRtransplant vs reference: 1.57; 95% CI: 1.07, 2.32). In both LOS groups, there were no statistically significant differences in the risk of HOB between solid tumor cancer and the reference.

Figure 1.

Prevalence rate ratio of HOB by DRG group and stratified by length of stay (≤30 days and >30 days). Models were adjusted for age, sex, diagnosis related groups (DRG) code, ICU status, Acute Laboratory Risk of Mortality Score (ALaRMS), payor type, and hospital-level variables (staffed bed size, teaching status, urbanicity).

Median time to reportable HOB

The median time between an admission and HOB event was within 30 days for all DRG groups (Figure 2). The median time to HOB was 8 days in solid tumor cancer and 8 days in the reference group. Patients hospitalized by MP cancer and transplant had a statistically significantly longer median compared to the reference group (13- and 12- days vs 8 days, both P < 0.001, respectively).

Figure 2.

Unadjusted Kaplan-Meier survival plot for days from admission to reportable (first) HOB by DRG. * Median days from admission start to reportable HOB.

HOB with other BSI events

In patients with HOB, the risk for subsequent HOB were generally similar by DRG. The exception were patients with LOS ≤30 days, MP cancer was associated with a 4.5-fold higher risk for subsequent HOB compared to the reference group (OR: 4.50, 95% CI: 1.81–11.22, Table 2).

Table 2.

Burden of Subsequent HOB by DRG and LOS

HOB, hospital-onset bacteremia and fungemia; DRG, diagnosis related groups; LOS, length of stay; OR, odds ratio; CI, confidence interval; n/a, not applicable.

Models were adjusted for age, sex, diagnosis related groups (DRG) code, Acute Laboratory Risk of Mortality Score (ALaRMS), payor type, and hospital-level variables (staffed bed size, teaching status, urbanicity).

The risk of commensal bacteria or infection with other pathogen was not statistically different by DRG or LOS in hospitalizations with reportable HOB (Supplemental Table 1).

Pathogen composition

Pathogen composition of reportable HOB by LOS and DRG group is presented in Figure 3. The top three prevalent pathogens were Enterobacteriaceae, Enterococcus spp., and Staphylococcus aureus in both LOS groups (Figure 3, Panel A) and DRG groups (Figure 3, Panel B). Enterobacteriaceae was the most prevalent pathogen for both LOS groups (34.6% in LOS ≤30 days and 28.5% in LOS >30 days). In LOS ≤30 days, S. aureus was the second most prevalent pathogen followed by Enterococcus spp. (25.2% and 13.7%, respectively. Whereas in the LOS >30 days, Enterococcus spp. was more prevalent than S. aureus (24.2% vs 15.8%).

Figure 3.

Prevalence of pathogen in reportable HOB (%) by LOS and DRG group. C. albicans and C. auris, Candida albicans and Candida auris; GNB, gram-negative bacteria.

The most prevalent pathogen for all DRG groups was Enterobacteriaceae. The second most prevalent pathogen for MP cancer, solid tumor cancer, and transplant groups was Enterococcus spp.; however, S. aureus was the second most prevalent for the reference group.

Discussion

The current report explored the demographics and clinical characteristics of hospitalizations at risk for HOB. When looking at initial and subsequent HOB events, a bimodal distribution was seen peaking at a time point within 30 days and >30 days of admission (data not shown). Most of these cases were in cancer and transplant patients which made clinical sense as both populations are at risk for prolonged immune suppression and/or often experience protracted and complicated hospitalizations. Therefore, this analysis focused on these patient populations compared to a reference group with further stratification of cancer patients into those with solid tumor vs MP cancer types. Key findings from this study include a higher burden of HOB and admission rates in MP cancer and transplant groups compared to the reference group.

MP Cancer and transplant patient populations have several clinical features in common that increase susceptibility to infection. Both groups are likely to be immunocompromised: transplant patients by intentional immune suppressive agents to enhance transplanted organ viability, and MP cancer by the underlying illness and/or antineoplastic chemotherapy. Solid tumor cancer patients likely have more admissions related to debulking and staging procedures and therefore may not be as chronically immune compromised until receiving antineoplastic or immunomodulating therapy. This may be why the HOB rates are more similar between not only the MP cancer and transplant patients but also between the solid tumor and reference group patients. Both groups can have protracted stays and are more prone to having invasive procedures with open or large post-surgical sites and frequently used vascular access devices for laboratory and therapeutic monitoring. These clinical realities could plausibly increase the risk for BSI, reportable HOB, and—for MP and transplant patients—subsequent HOB events.

A recent study evaluating HOB prevention in a tertiary care hospital setting determined that 56.0% of HOBs were not preventable by providers. The study attributed this to patients with HOB presenting with more complex disease states. Reference Stack, Dbeibo and Fadel22 Unadjusted HOB may be a marker for overall patient acuity rather than the quality of care. The goal of HOB is to provide an actionable metric for patient safety. To achieve this goal, an HOB-related quality metric ideally would standardize populations that account for acuity. Options include risk adjustment for specific patient-level risk factors to inform the HOB standardized infection ratio or excluding specific high-risk populations altogether, recognizing that prevention may not be realistic. These data would suggest that taking one of these approaches may be useful to evaluate quality of care fairly and accurately for hospitals with extensive transplant or cancer patient populations.

At the time of writing, the proposed HOB definition indexes on a first positive blood culture, which is agnostic to subsequent HOB events occurring in high-risk, prolonged stay populations. While highlighting a disproportionate risk for HOB, these data are insufficient to explore how preventable these events are in this high-risk population, which should drive the ultimate decision to risk adjust or exclude these populations altogether. This is a critical area of future research.

From a patient safety standpoint, visibility to the preferential burden of HOB (reportable or not) and in which at-risk populations, may help better prepare infection prevention programs and antimicrobial stewardship programs (ASP) for optimal HOB prevention, identification, and care. Understanding the pathogens associated with HOB in at-risk populations can further support efforts as certain organisms can originate from known reservoirs (ie, Enterobacteriaceae from GI sources, S. aureus from skin or soft tissue, etc.). Ostensibly, an infrastructure already exists to execute on focused infection prevention efforts and timely definitive therapy if a bloodstream infection event does occur given the CMS ruling that requires acute care hospitals to have an ASP as a condition of participation. Furthermore, ASP committees should have representation by infection prevention, microbiology lab and quality department leaders.

Conclusions

The results from this study demonstrated a higher burden of HOB in MP cancer and transplant hospitalizations, regardless of LOS. Across all DRG groups, the median time to first HOB event occurred within the first 30 days of admission; however, the MP cancer and transplant groups had a significantly longer time from admission to HOB event compared to the reference group. Finally, there was a higher risk of subsequent HOB in MP cancer hospitalizations. Should HOB become a mandatory metric, adjustment for hospitals caring for these patients is crucial to ensuring that the comparisons between facilities are meaningful.

Regardless as to whether HOB becomes a mandatory metric, delineating populations with a disproportionate burden of HOB and subsequent HOB represents a patient safety issue given the documented higher risk for mortality, LOS, readmission, and cost of care. Reference Yu, Jung and Ai1 These data identify a high priority focus area for future research in HOB prevention.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/ice.2024.160.

Acknowledgements

We thank Stephanie E. Tedford, PhD, of Pharmacologics, Inc, who, on the behalf of BD provided medical writing support.

Financial support

None.

Competing interests

KY, CA, MJ, and SB are employed by Becton Dickinson.

References

Yu, KC, Jung, M, Ai, C. Characteristics, costs, and outcomes associated with central-line-associated bloodstream infection and hospital-onset bacteremia and fungemia in US hospitals. Infect Control Hosp Epidemiol 2023; 44:1920–1926. doi: 10.1017/ice.2023.132.CrossRefGoogle ScholarPubMed
Allegranzi, B, Bagheri Nejad, S, Combescure, C, et al. Burden of endemic health-care-associated infection in developing countries: systematic review and meta-analysis. Lancet 2011;377:228–41. doi: 10.1016/s0140-6736(10)61458-4.CrossRefGoogle ScholarPubMed
Gidey, K, Gidey, MT, Hailu, BY, Gebreamlak, ZB, Niriayo, YL. Clinical and economic burden of healthcare-associated infections: a prospective cohort study. PLoS One 2023;18:e0282141. doi: 10.1371/journal.pone.0282141.CrossRefGoogle ScholarPubMed
Liu, X, Spencer, A, Long, Y, et al. A systematic review and meta-analysis of disease burden of healthcare-associated infections in China: an economic burden perspective from general hospitals. J Hosp Infect 2022;123:1–11. doi: 10.1016/j.jhin.2022.02.005.CrossRefGoogle Scholar
Rajaram, R, Barnard, C, Bilimoria, KY. Concerns about using the patient safety indicator-90 composite in pay-for-performance programs. JAMA 2015;313:897–8. doi: 10.1001/jama.2015.52.CrossRefGoogle ScholarPubMed
DiGiorgio, MJ, Vinski, J, Bertin, M, et al. Single-center study of interrater agreement in the identification of central line-associated bloodstream infection. Am J Infect Control 2014;42:638–42. doi: 10.1016/j.ajic.2014.02.027.CrossRefGoogle ScholarPubMed
Sato, A, Nakamura, I, Fujita, H, et al. Peripheral venous catheter-related bloodstream infection is associated with severe complications and potential death: a retrospective observational study. BMC Infect Dis 2017;17:434. doi: 10.1186/s12879-017-2536-0.CrossRefGoogle ScholarPubMed
Chopra, V, Kaatz, S, Swaminathan, L, et al. Variation in use and outcomes related to midline catheters: results from a multicentre pilot study. BMJ Qual Saf 2019;28:714–720. doi: 10.1136/bmjqs-2018-008554.CrossRefGoogle ScholarPubMed
Guidelines for the Prevention of Bloodstream Infections and Other Infections Associated with the Use of Intravascular Catheters. Part I: Peripheral Catheters. Geneva: World Health Organization; 2024.Google Scholar
Dantes, RB, Abbo, LM, Anderson, D, et al. Hospital epidemiologists’ and infection preventionists’ opinions regarding hospital-onset bacteremia and fungemia as a potential healthcare-associated infection metric. Infect Control Hosp Epidemiol 2019;40:536–540. doi: 10.1017/ice.2019.40.CrossRefGoogle ScholarPubMed
Rock, C, Thom, KA, Harris, AD, et al. A multicenter longitudinal study of hospital-onset bacteremia: time for a new quality outcome measure? Infect Control Hosp Epidemiol 2016;37:143–8. doi: 10.1017/ice.2015.261.CrossRefGoogle ScholarPubMed
Ridgway, JP, Sun, X, Tabak, YP, Johannes, RS, Robicsek, A. Performance characteristics and associated outcomes for an automated surveillance tool for bloodstream infection. Am J Infect Control 2016;44:567–71. doi: 10.1016/j.ajic.2015.12.044.CrossRefGoogle ScholarPubMed
Dantes, RB, Rock, C, Milstone, AM, et al. Preventability of hospital onset bacteremia and fungemia: A pilot study of a potential healthcare-associated infection outcome measure. Infect Control Hosp Epidemiol 2019;40:358–361. doi: 10.1017/ice.2018.339.CrossRefGoogle ScholarPubMed
Battaglia, CC, Hale, K. Hospital-acquired infections in critically ill patients with cancer. J Intensive Care Med 2019;34:523–536. doi: 10.1177/0885066618788019.CrossRefGoogle ScholarPubMed
Ariza-Heredia, EJ, Chemaly, RF. Update on infection control practices in cancer hospitals. CA Cancer J Clin 2018;68(5):340–355. doi: 10.3322/caac.21462 CrossRefGoogle ScholarPubMed
Stoclin, A, Rotolo, F, Hicheri, Y, et al. Ventilator-associated pneumonia and bloodstream infections in intensive care unit cancer patients: a retrospective 12-year study on 3388 prospectively monitored patients. Support Care Cancer 2020;28:193–200. doi: 10.1007/s00520-019-04800-6.CrossRefGoogle Scholar
Yu, KC, Ye, G, Edwards, JR, et al. Hospital-onset bacteremia and fungemia: An evaluation of predictors and feasibility of benchmarking comparing two risk-adjusted models among 267 hospitals. Infect Control Hosp Epidemiol 2022;43:1317–1325. doi: 10.1017/ice.2022.211.CrossRefGoogle ScholarPubMed
Yu, KC, Yamaga, C, Vankeepuram, L, Tabak, YP. Relationships between creatinine increase and mortality rates in patients given vancomycin in 76 hospitals: The increasing role of infectious disease pharmacists. Am J Health Syst Pharm 2021;78:2116–2125. doi: 10.1093/ajhp/zxab247.Google ScholarPubMed
Schrank, GM, Snyder, GM, Leekha, S. Hospital-onset bacteremia and fungemia: examining healthcare-associated infections prevention through a wider lens. Antimicrob Steward Healthc Epidemiol 2023;3:e198. doi: 10.1017/ash.2023.486.CrossRefGoogle ScholarPubMed
Leekha, S, Robinson, G, Jacob, JT, et al. Sources and preventability of hospital-onset bacteremia and fungemia in the united states: evaluation of a potential healthcare quality measure. Open Forum Infect Dis 2022;9:ofac492.132. doi: 10.1093/ofid/ofac492.132 CrossRefGoogle Scholar
Tabak, YP, Sun, X, Nunez, CM, Johannes, RS. Using electronic health record data to develop inpatient mortality predictive model: acute laboratory risk of mortality score (ALaRMS). J Am Med Inform Assoc 2014;21:455–63. doi: 10.1136/amiajnl-2013-001790.CrossRefGoogle ScholarPubMed
Stack, MA, Dbeibo, L, Fadel, W, et al. Etiology and utility of hospital-onset bacteremia as a safety metric for targeted harm reduction. Am J Infect Control 2024;52:195–199. doi: 10.1016/j.ajic.2023.06.002.CrossRefGoogle ScholarPubMed
Figure 0

Table 1.Patient, clinical, and hospital characteristics overall and by diagnosis related groups

Figure 1

Figure 1.Prevalence rate ratio of HOB by DRG group and stratified by length of stay (≤30 days and >30 days). Models were adjusted for age, sex, diagnosis related groups (DRG) code, ICU status, Acute Laboratory Risk of Mortality Score (ALaRMS), payor type, and hospital-level variables (staffed bed size, teaching status, urbanicity).

Figure 2

Figure 2.Unadjusted Kaplan-Meier survival plot for days from admission to reportable (first) HOB by DRG. * Median days from admission start to reportable HOB.

Figure 3

Table 2.Burden of Subsequent HOB by DRG and LOS

Figure 4

Figure 3.Prevalence of pathogen in reportable HOB (%) by LOS and DRG group. C. albicans and C. auris, Candida albicans and Candida auris; GNB, gram-negative bacteria.

Supplementary material: File

Yu et al. supplementary material

Yu et al. supplementary material
Download Yu et al. supplementary material(File)
File 15.6 KB