ORAHS2025
Abstract Submission

197. Optimizing Healthcare Resources through AI-Enhanced Diabetic Foot Screening: A Long-Term Cost-Effectiveness Analysis

Invited abstract in session MC-1: Simulation 1, stream Sessions.

Monday, 11:00-12:30
Room: NTNU, Realfagbygget R5

Authors (first author is the speaker)

1. YAN SUN
Health Services and Outcomes Research, National Healthcare Group
2. Gary Ang
National Healthcare Group
3. Lixia Ge
National Healthcare Group
4. Zhiwen Lo
Woodlands Hospital
5. huiling Liew
Tan Tock Seng Hospital
6. Donna Tan
National Healthcare Group Polyclinics
7. Daniel Chew
Tan Tock Seng Hospital
8. John Abisheganaden
National Healthcare Group

Abstract

Background: Diabetic foot ulcers (DFUs) are a serious complication of diabetes mellitus requiring substantial healthcare resources. While regular foot screenings prevent lower extremity amputations (LEAs), conventional annual screening approaches are resource-intensive. This study evaluates how AI can help optimize screening resource allocation while maintaining care quality.

Method: We developed a Markov state-transition model simulating disease progression across five health states: diabetes, DFU, minor & major LEA, and death. The model leverages national disease registry data for transition probabilities and costs. We compared AI-enhanced screening approach against routine annual screening for 500,000 low-risk diabetic patients over 40 years. The AI model recommends screening interval tailored to individual risk profile. Effectiveness was measured in quality-adjusted life years (QALYs). Monte Carlo micro-simulation was performed for probability sensitivity analysis (PSA).

Results: The AI-enhanced strategy demonstrated significant resource optimization, eliminating 6.8 million unnecessary screenings and saving healthcare costs by S$657.5 million over 40 years with minimal QLAY loss (ICER: S$174,572/QLAY, SD:13,296), proving its cost-effectiveness.

Conclusion: AI-enhanced screening AI-enhanced DFU screening optimizes healthcare resources through personalized risk assessment, significantly reducing unnecessary screenings while maintaining clinical outcomes.

Keywords

Status: accepted


Back to the list of papers