TY - JOUR
T1 - Cost-effectiveness of AI-based diabetic retinopathy screening in nationwide health checkups and diabetes management in Japan
T2 - A modeling study
AU - Akune, Yoko
AU - Kawasaki, Ryo
AU - Goto, Rei
AU - Tamura, Hiroshi
AU - Hiratsuka, Yoshimune
AU - Yamada, Masakazu
N1 - Publisher Copyright:
© 2025 The Authors
PY - 2025/3
Y1 - 2025/3
N2 - Aims: We evaluated the cost-effectiveness of artificial intelligence (AI)-based diabetic retinopathy (DR) screening in Japan. This evaluation compared the simultaneous introduction of AI in nationwide health checkups, namely “specific health check-ups in Japan” (SHC), and diabetes complication management (AI-case) with the current situation where AI is not being introduced (conventional-case) from the healthcare payer's perspective. Methods: A cost-effectiveness analysis was conducted using a new individual-based state transition model. Model parameters, including the incidence and progression of DR, health utility values, and costs of screening and treatment, were based on literature data and expert opinion. The analysis estimated quality-adjusted life years (QALYs), cumulative costs, and incremental cost-effectiveness ratios (ICER). Results: The ICER comparing the AI-case with conventional-case was estimated to be JPY 1,598,244/QALY (USD 11,375/QALY), which is below the willingness-to-pay threshold of JPY 5 million/QALY (USD 35,584/QALY). Scenario analyses revealed that ICERs for the AI-based DR screening in SHC-only condition was JPY 1,895,226/QALY (USD 13,488/QALY) and JPY 3,960,839/QALY (USD 28,189/QALY) in diabetes management-only condition. Conclusions: The introduction of AI-based DR screening for SHC and diabetes management was cost-effective compared to the current situation in Japan.
AB - Aims: We evaluated the cost-effectiveness of artificial intelligence (AI)-based diabetic retinopathy (DR) screening in Japan. This evaluation compared the simultaneous introduction of AI in nationwide health checkups, namely “specific health check-ups in Japan” (SHC), and diabetes complication management (AI-case) with the current situation where AI is not being introduced (conventional-case) from the healthcare payer's perspective. Methods: A cost-effectiveness analysis was conducted using a new individual-based state transition model. Model parameters, including the incidence and progression of DR, health utility values, and costs of screening and treatment, were based on literature data and expert opinion. The analysis estimated quality-adjusted life years (QALYs), cumulative costs, and incremental cost-effectiveness ratios (ICER). Results: The ICER comparing the AI-case with conventional-case was estimated to be JPY 1,598,244/QALY (USD 11,375/QALY), which is below the willingness-to-pay threshold of JPY 5 million/QALY (USD 35,584/QALY). Scenario analyses revealed that ICERs for the AI-based DR screening in SHC-only condition was JPY 1,895,226/QALY (USD 13,488/QALY) and JPY 3,960,839/QALY (USD 28,189/QALY) in diabetes management-only condition. Conclusions: The introduction of AI-based DR screening for SHC and diabetes management was cost-effective compared to the current situation in Japan.
KW - AI-based diabetic retinopathy screening
KW - Cost-effectiveness analysis
KW - Diabetes complication management
UR - http://www.scopus.com/inward/record.url?scp=85216809645&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85216809645&partnerID=8YFLogxK
U2 - 10.1016/j.diabres.2025.112015
DO - 10.1016/j.diabres.2025.112015
M3 - Article
AN - SCOPUS:85216809645
SN - 0168-8227
VL - 221
JO - Diabetes Research and Clinical Practice
JF - Diabetes Research and Clinical Practice
M1 - 112015
ER -
