Abstract: TH-PO004
Health System Economic Burden in Patients with CKD: Insights from the Klinrisk Model
Session Information
- Augmented Intelligence for Prediction and Image Analysis
October 24, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Tangri, Navdeep, University of Manitoba, Winnipeg, Manitoba, Canada
- Singh, Rakesh, Bayer U.S. LLC, Whippany, New Jersey, United States
- Betts, Keith A., Analysis Group Inc, Los Angeles, California, United States
- Du, Yuxian, Bayer U.S. LLC, Whippany, New Jersey, United States
- Gao, Sophie, Analysis Group Inc, Los Angeles, California, United States
- Katta, Arvind, Bayer U.S. LLC, Whippany, New Jersey, United States
- Farag, Youssef MK, Bayer U.S. LLC, Whippany, New Jersey, United States
- Fatoba, Samuel T., Bayer U.S. LLC, Whippany, New Jersey, United States
- Liu, Hongjiao, Analysis Group Inc, Los Angeles, California, United States
- Chen, Jingyi, Analysis Group Inc, Los Angeles, California, United States
- Ferguson, Thomas W., University of Manitoba, Winnipeg, Manitoba, Canada
- Whitlock, Reid, University of Manitoba, Winnipeg, Manitoba, Canada
- Leon Mantilla, Silvia Juliana, University of Manitoba, Winnipeg, Manitoba, Canada
- Singh, Ajay K., Harvard Medical School, Boston, Massachusetts, United States
Background
CKD is associated with substantial economic burden. Early identification of at-risk CKD can facilitate appropriate and timely intervention and reduce CKD-related medical costs. This study assessed the association between CKD progression risk and healthcare burden.
Methods
A retrospective observational study was conducted in 1,050,552 adult patients with CKD from Optum’s electronic health records database (1/1/2007 - 9/30/2022). A previously published and validated machine learning model, Klinrisk (Tangri N, et al Clin Kidney J. 2024 Mar 6;17(4)) was applied to classify patients into 3 groups based on their risk of CKD progression (low, medium, and high). All-cause inpatient (IP) admissions, emergency room (ER) visits, outpatient (OP) visits were evaluated in each risk group during the 1 year after CKD. Average medical costs (2023 USD) were calculated as the average length of stay for IP admissions, average number of ER and OP visits multiplied by the corresponding unit costs as estimated from a prior study.
Results
Patients with higher predicted CKD progression risk had higher healthcare utilization (HRU). High-risk patients averaged 1.49 IP admissions, 0.83 ER visits, and 35.75 OP visits per year compared with 0.31 IP admissions, 0.63 ER visits, and 24.45 OP visits among low-risk patients. The total annual medical costs for low-, medium-, and high-risk patients were $16,018, $22,090, and $50,218, respectively (Figure). IP costs were the major cost driver for high-risk patients.
Conclusion
Patients at high risk of CKD progression as predicted by Klinrisk were associated with high HRU and medical costs and may benefit from early intervention with guideline-directed therapies, to reduce economic burden.
Annualized all-cause medical costs stratified by CKD progression risk