Impact of inter-hospital transfers on the prevalence of resistant pathogens in a hospital–community system

Piotrowska, Monika; Sakowski, Konrad; Lonc, Agata; Tahir, Hannan; Kretzschmar, Mirjam

doi:10.1016/j.epidem.2020.100408

“…Initially, to model the spread of Extended-spectrum lactamase-Producing Klebsiella pneumoniae (ESBL-Klebsiella) 20 – 22 and other common HAIs 4 , 5 , 11 – 17 , 28 , we set based on the assumption that the meantime of colonization was 365 days in the baseline scenario. We also studied other patient transfer patterns for different values of with , for sensitivity analyses.…”

Section: Methodsmentioning

confidence: 99%

Regional patient transfer patterns matter for the spread of hospital-acquired pathogens

Xia,

Horn,

Piotrowska

et al. 2024

Sci Rep

0

View full text Add to dashboard Cite

Pathogens typically responsible for hospital-acquired infections (HAIs) constitute a major threat to healthcare systems worldwide. They spread via hospital (or hospital-community) networks by readmissions or patient transfers. Therefore, knowledge of these networks is essential to develop and test strategies to mitigate and control the HAI spread. Until now, no methods for comparing healthcare networks across different systems were proposed. Based on healthcare insurance data from four German federal states (Bavaria, Lower Saxony, Saxony and Thuringia), we constructed hospital networks and compared them in a systematic approach regarding population, hospital characteristics, and patient transfer patterns. Direct patient transfers between hospitals had only a limited impact on HAI spread. Whereas, with low colonization clearance rates, readmissions to the same hospitals posed the biggest transmission risk of all inter-hospital transfers. We then generated hospital-community networks, in which patients either stay in communities or in hospitals. We found that network characteristics affect the final prevalence and the time to reach it. However, depending on the characteristics of the pathogen (colonization clearance rate and transmission rate or even the relationship between transmission rate in hospitals and in the community), the studied networks performed differently. The differences were not large, but justify further studies.

show abstract

“…They were weighted by p i j (w) = n i j (w) a i (w) with i ̸ = j, describing the probability of a patient being transferred from node i to j on a day of the week w, where n i j (w) with i ̸ = j denoted the average number of transferred patients from node i to j on days of the week w, and a i (w) represented the average number of patients in node i on one day of the week w, both calculated directly from federal scaled-up datasets. Similarly to the previous approach 27 , the probability of a patient staying in node i on one day of the week w can be calculated as p…”

Section: Network Constructionmentioning

confidence: 99%

Patient transfer patterns and the spread of hospital-acquired infections in subnetworks of a country: an example of federal states in Germany

Xia

¹

,

Horn

²

,

Piotrowska

³

et al. 2023

Preprint

0

View full text Add to dashboard Cite

Pathogens typically responsible for hospital-acquired infections (HAIs) constitute a major threat to healthcare systems worldwide. They spread via hospital (or hospital-community) networks by readmissions or patient transfers. Therefore, knowledge of these networks is essential to develop and test strategies to mitigate and control the HAI spread. Until now, no methods for comparing healthcare networks across different systems were proposed. Based on healthcare insurance data from four German federal states (Bavaria, Lower Saxony, Saxony and Thuringia), we constructed hospital networks and compared them in a systematic approach regarding population, hospital characteristics, and patient transfer patterns. Direct patient transfers between hospitals had only a limited impact on HAI spread. Whereas, with low colonization clearance rates, readmissions to the same hospitals posed the biggest transmission risk of all inter-hospital transfers. We then generated hospital-community networks, in which patients either stay in communities or in hospitals. We found that network characteristics affect the final prevalence and the time to reach it. However, depending on the characteristics of the pathogen (colonization clearance rate and transmission rate or even the relationship between transmission rate in hospitals and in the community), the studied networks performed differently. The differences were not large, but justify further studies.

show abstract

“…Various studies have attempted to address these two matters separately. For instance, authors of [12,14,57,69] discussed the transfer pattern associated with hospital acquired infections, and some also address hospital ward planning [44]. When it comes to capacity management, the more optimized and efficient the hospital operates (smooth patient flow), the less loss of time incurs.…”

Section: Contributions and Implicationsmentioning

confidence: 99%

Feature extraction from unequal length heterogeneous EHR time series via dynamic time warping and tensor decomposition

Zhang

¹

,

Fanaee-T

²

,

Thoresen

³

2021

Data Min Knowl Disc

View full text Add to dashboard Cite

Electronic Health Records (EHR) data is routinely generated patient data that can provide useful information for analytical tasks such as disease detection and clinical event prediction. However, temporal EHR data such as physiological vital signs and lab test results are particularly challenging. Temporal EHR features typically have different sampling frequencies; such examples include heart rate (measured almost continuously) and blood test results (a few times during a patient's entire stay). Different patients also have different length of stays. Existing approaches for temporal EHR sequence extraction either ignore the temporal pattern within features, or use a predefined window to select a section of the sequences without taking into account all the information. We propose a novel approach to tackle the issue of irregularly sampled, unequal length EHR time series using dynamic time warping and tensor decomposition. We use DTW to learn the pairwise distances for each temporal feature among the patient cohort and stack the distance matrices into a tensor. We then decompose the tensor to learn the latent structure, which is consequently used for patient representation. Finally, we use the patient representation for in-hospital mortality prediction. We illustrate our method on two cohorts from the MIMIC-III database: the sepsis and the acute kidney failure cohorts. We show that our method produces outstanding classification performance in terms of AUROC, AUPRC and accuracy compared with the baseline methods: LSTM and DTW-KNN. In the end we provide a detailed analysis on the feature importance for the interpretability of our method.

show abstract

Impact of inter-hospital transfers on the prevalence of resistant pathogens in a hospital–community system

Cited by 13 publications

References 19 publications

Regional patient transfer patterns matter for the spread of hospital-acquired pathogens

Regional patient transfer patterns matter for the spread of hospital-acquired pathogens

Patient transfer patterns and the spread of hospital-acquired infections in subnetworks of a country: an example of federal states in Germany

Feature extraction from unequal length heterogeneous EHR time series via dynamic time warping and tensor decomposition

Contact Info

Product

Resources

About