AUTHOR=Duan Juntao , Li Hanmo , Ma Xiaoran , Zhang Hanjie , Lasky Rachel , Monaghan Caitlin K. , Chaudhuri Sheetal , Usvyat Len A. , Gu Mengyang , Guo Wensheng , Kotanko Peter , Wang Yuedong 

TITLE=Predicting SARS-CoV-2 infection among hemodialysis patients using multimodal data

JOURNAL=Frontiers in Nephrology

VOLUME=Volume 3 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/nephrology/articles/10.3389/fneph.2023.1179342

DOI=10.3389/fneph.2023.1179342

ISSN=2813-0626

ABSTRACT=Background 
The COVID-19 pandemic has created more devastation to dialysis patients
than to the general population. Patient-level prediction models for SARS-CoV-2
infection are crucial for the early identification of patients to prevent and mitigate
outbreaks within dialysis clinics. As the COVID-19 pandemic evolves, it is unclear
whether previously built prediction models are still sufficiently effective.

Methods
We developed a machine learning (XGBoost) model to predict during the incubation
period a SARS-CoV-2 infection that is subsequently diagnosed after three or
more days. We used data from multiple sources, including demographic, clinical, treatment,
laboratory, and vaccination information from a national network of hemodialysis
clinics, socioeconomic information from the Census Bureau, and county-level COVID-
19 infection and mortality information from state and local health agencies. We created
prediction models and evaluated their performances on a rolling basis to investigate
the evolution of prediction power and risk factors.

Result 
From April 2020 to August 2020, our machine learning model achieved an
AUC of 0.75, an improvement over 0.07 from a previously developed machine learning
model published on Kidney360 in 2021. As the pandemic evolved, the prediction performance
deteriorated and fluctuated more, with the lowest AUC of 0.6 in December 2021 and January 2022. Over the whole study period from April 2020 to February
2022, fixing the false positive rate at 20%, our model can detect 40% of the positive
patients. We found that features derived from local infection information reported by
CDC are the most important predictors and vaccination status is a useful predictor
as well. Whether a patient lives in a nursing home was an effective predictor before
vaccination but became less predictive after vaccination.
Conclusion As found in our study, the dynamics of the prediction model are frequently
changing as the pandemic evolves. County-level infection information and vaccination
are crucial for the success of early COVID-19 prediction models. Our results show
that the proposed model can effectively identify SARS-CoV-2 infections during the
incubation period. Prospective studies are warranted to explore the application of
such prediction models in daily clinical practice.