Capacity strain and risk of ICU-onset bloodstream infection Open Access
Niehaus, Emily (Spring 2019)
Abstract
Background:
Capacity strain refers to the concept of a mismatch between medical workload and patient care resources, which leads to variability in the ability to provide high-quality care. Capacity strain in the intensive care unit (ICU) has been associated with increased odds of mortality and changes in patient-care processes. Less is known about the impact of capacity strain on healthcare-associated infections, especially hospital-onset bloodstream infections (BSI), which occur in 4.4-10% of all ICU admissions and are associated with increased mortality rate, longer length of stay, and higher hospital costs.
Methods:
Using a retrospective cohort of all adult patient encounters admitted to any critical care unit for more than three days at four university-affiliated hospitals between 01/01/2014 and 12/29/2018, we classified all ICU onset-BSI, occurring after ICU day 3, as true pathogen or contaminant. We used multivariable logistic regression to evaluate the relationship between capacity strain at the ICU level and patient-level risk factors for ICU-onset BSI. Capacity strain, the primary exposure, was defined as the number of days within the first three days of the patient’s ICU stay that fell above the 90th percentile for the unit’s daily census in that year.
Results:
There were 24,786 patients included in the cohort, 387(1.6%) of which experienced a non-contaminant ICU-onset BSI. At the patient level, encounters that had ICU-onset BSI had higher SOFA scores at time of ICU admission (7 vs .5), were more likely to have a central line (83% vs. 57%) and require mechanical ventilation (62% vs. 39%), and had longer lengths of stay in ICU (18 vs. 6 days) than those who did not have ICU-onset BSI (p<0.0001 for all). While increased exposure to a strained unit was associated with risk of developing ICU-onset BSI when adjusting for patient risk factors (3 vs. 0 days of strain OR=1.53, 95% CI 1.11-2.11), there was no observed association after adjusting for unit characteristics.
Conclusion:
ICU-onset BSIs were infrequent (2%), but associated with higher acuity of illness and device utilization. ICU admission on a day with a census at >90th percentile was not associated with subsequently developing an ICU-onset bloodstream infection after adjusting for patient and unit level risk factors. Future studies should evaluate capacity strain and infection prevention processes of care in the ICU.
Table of Contents
Chapter I: Background.……...……………………………………….……………….1
Chapter II: Manuscript………………………………………………………………10
References…………………………………………………………………...20
Tables………………………………………………………………………..30
Figures………………………………………………………………….…...38
Chapter III: Public Health Implications…...……………………………………….. 39
Chapter IV: Appendix.……………………………………………….…………....41
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Keyword | |
| Committee Chair / Thesis Advisor |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Capacity strain and risk of ICU-onset bloodstream infection () | 2019-04-23 15:21:04 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|