KEY TAKEAWAYS
- The study aimed to evaluate the cost-effectiveness of integrating AI-based risk-stratified screening into BC programs.
- Results showed AI-guided risk-stratified screening is likely cost-effective, offering added health benefits and reduced costs.
Previous research indicated that the Mirai AI model effectively predicts short-term breast cancer (BC) risk. However, its economic impact remains unstudied.
Harry Hill and the team aimed to evaluate the cost-effectiveness of incorporating a BC AI model into risk-stratified screening.
The study developed a decision-analytic model to estimate health-related quality of life (QoL), cancer survival rates, and lifetime costs for women eligible for screening. The analysis, simulated a cohort of women aged 50 to 70 years. Mammography screening was assessed at 1 to 6-year intervals based on BC risk compared to standard care (every 3 years).
Incremental net monetary benefit was evaluated using quality-adjusted life-years (QALYs) and National Health Service (NHS) costs (in pounds sterling; convert to US dollars by multiplying by 1.28).
They found that AI-based risk-stratified screening programs were cost-saving and increased quality-adjusted life-years (QALYs) compared to the current screening approach.
A proposed schedule—every 6 years for low-risk individuals, biannually for those below average risk, triennially for those above average risk, and annually for high-risk individuals—was estimated to provide annual net monetary benefits of approximately £60.4 million (US $77.3 million) and £85.3 million (US $109.2 million) with QALY values set at £20,000 (US $25,600) and £30,000 (US $38,400), respectively.
Even if decision-makers were cautious about additional NHS resource allocation, implementing the proposed strategies at a QALY value of £1 (US $1.28) was projected to generate an annual monetary benefit of around £10.6 million (US $13.6 million).
The study concluded that integrating AI-based risk-stratified screening into BC programs was likely cost-effective, providing additional health benefits at lower costs. These findings were particularly pertinent for healthcare settings facing resource constraints. Further prospective studies to evaluate AI-guided screening were recommended.
Research was funded by grant 2019DecPR1395 from Breast Cancer Now (Dr Brentnall); grant C49757/A28689 from Cancer Research UK (Dr Brentnall); grant PR-PRU-1217-20401 from the National Institute of Health Research (NIHR) Policy Research Programme, conducted through the Policy Research Unit in Economic Methods of Evaluation in Health and Social Care Interventions (Dr Hill); and grant PR-PRU-1217-21601 from the NIHR Policy Research Programme, conducted through the Policy Research Unit in Cancer Awareness, Screening and Early Diagnosis (Dr Duffy).
Source: https://pubmed.ncbi.nlm.nih.gov/39235813/
Hill H, Roadevin C, Duffy S, et al. (2024). “Cost-Effectiveness of AI for Risk-Stratified Breast Cancer Screening.” JAMA Netw Open. 2024;7(9):e2431715. Published 2024 Sep 3. doi:10.1001/jamanetworkopen.2024.31715
