Abstract
End-stage kidney disease (ESKD) disproportionately affects Hispanic and American Indian (AI) persons with prevalence rates more than double of White persons. Furthermore, all-cause mortality is significantly higher for Hispanic and AI. The increased prevalence and mortality rates can be attributed to the higher level of health disparities caused by social determinants of health (SDOH). We are developing machine learning and computational models based on patient-level ESKD data at zip code or county level paired with SDOH data to predict the risk of mortality among Hispanic and AI persons with ESKD in South Dakota who have varying socio-demographic characteristics. The primary impact of our study is the development of a fair and unbiased identifier mortality risk that will provide the opportunity for healthcare providers and policymakers to mitigate the health disparities among Hispanic and AI persons with ESKD in South Dakota that can be applied to other geographic locations in the United States using our model. The purpose of this presentation is to provide an overview of the results of our nationwide survival analysis that was completed using United States Renal Data System data, and how to apply the survival data to currently available SDOH metrics (e.g., Area Deprivation Index). We will also describe our next steps for this work by sharing how this data will be used to develop a machine learning algorithm that will generate mortality risk scores based on SDOH, non-biomedical, and biomedical indicators at an individual level.
Notes
References:
1. United States Renal Data System. 2022 USRDS Annual Data Report: Epidemiology of kidney disease in the United States. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD.2022. https://adr.usrds.org/2022
2. American College of Cardiology. Cardiometabolic initiatives. n.d. Accessed June 15, 2023. https://www.acc.org/tools-and-practice-support/quality-programs/cardiometabolic-health-alliance#:~:text=Cardiometabolic%20disorders%20represent%20a%20cluster,abdominal%20obesity%20and%20elevated%20triglycerides
3. Center for Disease Control and Prevention. Native Americans with diabetes: Better diabetes care can decrease kidney failure. 2018. Accessed September 16, 2022. https://www.cdc.gov/vitalsigns/aian-diabetes/index.html
4. Warne D, Frizzell LB. American Indian Health Policy: Historical Trends and Contemporary Issues. American Journal of Public Health. 2014;104(S3):S263-S267. doi:10.2105/ajph.2013.301682
5. Khetpal V, Roosevelt J, Adashi EY. A Federal Indian Health Insurance Plan: Fulfilling a solemn obligation to American Indians and Alaska Natives in the United States. Prev Med Rep. 2022/02/01/ 2022;25:101669. doi:10.1016/j.pmedr.2021.101669
6. Tangri N, Inker LA, Hiebert B, et al. A Dynamic Predictive Model for Progression of CKD. Am J Kidney Dis. Apr 2017;69(4):514-520. doi:10.1053/j.ajkd.2016.07.030
7. Bundy JD, Mills KT, Anderson AH, Yang W, Chen J, He J. Prediction of End-Stage Kidney Disease Using Estimated Glomerular Filtration Rate With and Without Race. Annals of Internal Medicine. 2022/03/15 2022;175(3):305-313. doi:10.7326/M21-2928
8. Segal Z, Kalifa D, Radinsky K, et al. Machine learning algorithm for early detection of end-stage renal disease. BMC Nephrology. 2020;21(1)doi:10.1186/s12882-020-02093-0
9. Tangri N. A Predictive Model for Progression of Chronic Kidney Disease to Kidney Failure. Jama. 2011;305(15):1553. doi:10.1001/jama.2011.451
Sigma Membership
Phi
Type
Presentation
Format Type
Text-based Document
Study Design/Type
N/A
Research Approach
N/A
Keywords:
Health Equity, Social Determinants of Health, End-stage Kidney Disease, ESKD
Recommended Citation
Varilek, Brandon M.; Michael, Semhar; Moradi Rekabdarkolaee, Hossein; Brooks, Patti J.; and Ngorsuraches, Surachat, "Developing a Machine Learning Method to Identify Racial Disparities in End-Stage Kidney Disease" (2025). International Nursing Research Congress (INRC). 4.
https://www.sigmarepository.org/inrc/2025/presentations_2025/4
Conference Name
36th International Nursing Research Congress
Conference Host
Sigma Theta Tau International
Conference Location
Seattle, Washington, USA
Conference Year
2025
Rights Holder
All rights reserved by the author(s) and/or publisher(s) listed in this item record unless relinquished in whole or part by a rights notation or a Creative Commons License present in this item record.
Review Type
Abstract Review Only: Reviewed by Event Host
Acquisition
Proxy-submission
Developing a Machine Learning Method to Identify Racial Disparities in End-Stage Kidney Disease
Seattle, Washington, USA
End-stage kidney disease (ESKD) disproportionately affects Hispanic and American Indian (AI) persons with prevalence rates more than double of White persons. Furthermore, all-cause mortality is significantly higher for Hispanic and AI. The increased prevalence and mortality rates can be attributed to the higher level of health disparities caused by social determinants of health (SDOH). We are developing machine learning and computational models based on patient-level ESKD data at zip code or county level paired with SDOH data to predict the risk of mortality among Hispanic and AI persons with ESKD in South Dakota who have varying socio-demographic characteristics. The primary impact of our study is the development of a fair and unbiased identifier mortality risk that will provide the opportunity for healthcare providers and policymakers to mitigate the health disparities among Hispanic and AI persons with ESKD in South Dakota that can be applied to other geographic locations in the United States using our model. The purpose of this presentation is to provide an overview of the results of our nationwide survival analysis that was completed using United States Renal Data System data, and how to apply the survival data to currently available SDOH metrics (e.g., Area Deprivation Index). We will also describe our next steps for this work by sharing how this data will be used to develop a machine learning algorithm that will generate mortality risk scores based on SDOH, non-biomedical, and biomedical indicators at an individual level.
Description
End-stage kidney disease disproportionately affects Hispanic and American Indian persons with prevalence rates more than double of White persons, and significant increases in mortality. The increased prevalence and mortality rates can be attributed to the higher level of health disparities caused by social determinants of health (SDOH). Machine learning models based on SDOH can provide clinicians and policy makers with actionable data to mitigate mortality risk based on modifiable SDOH.