Cost-Effectiveness of Artificial-Intelligence Enabled Kidney Disease Risk Stratification in US Veterans With Early-Stage Diabetic Kidney Disease

Speaker(s)

Sarker J1, Abdelaziz A1, Crook J2, Nelson RE3, LaFleur J2, Lu CC2, Nyman H2, Kim K1
1University of Illinois Chicago, Chicago, IL, USA, 2University of Utah, Salt Lake City, UT, USA, 3Division of Epidemiology, Department of Internal Medicine, Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, USA

Presentation Documents

OBJECTIVES: To evaluate the cost-effectiveness and budget impact of the artificial intelligence enabled kidney disease (AIKD) risk stratification among US veterans with diabetic kidney disease (DKD).

METHODS: A decision tree was developed to compare the predictive performance of the AIKD and Standard of Care (SoC) in assessing the risk of DKD progression, outlining care pathways. This was followed by Markov state transitions across various DKD stages, where a proper renal-protective treatment given to the patients at elevated risk would reduce the rate of stage progressions. Stage specific costs of care and annual transition rates were obtained from the analysis of Veterans Health Administration (VHA) electronic healthcare records and Managerial Cost Accounting data. We estimated the incremental cost-effectiveness ratio (ICER) for AIKD versus SoC over a 5-year time horizon, applying a 3% annual discounting rate to both direct healthcare costs and quality-adjusted life years (QALYs). Sensitivity analyses were performed on model inputs to assess their influence on ICERs. A five-year budget impact of AIKD was estimated for a cohort of 42,000 patients, representing about 10% of those eligible for AIKD and in stages 1 to 3b of DKD.

RESULTS: The 5-year discounted costs for AIKD and SoC were $146,437 and $145,120, respectively, while the QALYs were 2.828 for AIKD and 2.816 for SoC. This leads to an ICER of $116,349 per QALY gained for AIKD. The model demonstrated robustness in sensitivity analyses. The implementation of AIKD to the 10% of eligible DKD patients would have a downstream budget impact of $56 million over five years.

CONCLUSIONS: For US veterans with early-stage DKD, AIKD emerges as a cost-effective strategy, given its ICER falls below the $150,000/QALY threshold. Integration of AIKD into DKD management is projected to have a manageable five-year budget impact.

Code

EE364

Topic

Economic Evaluation, Medical Technologies

Topic Subcategory

Budget Impact Analysis, Cost-comparison, Effectiveness, Utility, Benefit Analysis, Diagnostics & Imaging

Disease

Diabetes/Endocrine/Metabolic Disorders (including obesity), Urinary/Kidney Disorders