Presentation (CTI)

OBJECTIVES: A recent clinical evaluation conducted within an NIHR-funded AI dental project demonstrated that AI assistance improved dentists’ diagnostic accuracy for detecting dental caries on radiographs. An embedded modelling study aimed to evaluate whether these improvements could translate into meaningful health and economic benefits in England.
METHODS: An early economic evaluation was conducted to assess the cost-effectiveness of an AI-assisted diagnostic pathway for dental caries versus standard care from the NHS England perspective. A decision-analytic model combined a decision tree capturing diagnosis and initial treatment allocation with a cohort state-transition model capturing long-term disease progression. The model simulated a representative English adult cohort aged 49 years over a 20-year time horizon. Diagnostic accuracy inputs were derived from the associated multi-reader, multi-case clinical evaluation, in which AI assistance improved overall caries detection sensitivity from 53.4% to 74.2% and specificity from 94.8% to 97.2%. Disease prevalence was informed by the Adult Oral Health Survey 2023, while treatment allocation, disease progression, costs, and health outcomes were informed by published literature and national data sources. Outcomes included costs, quality-adjusted life years (QALYs), quality-adjusted tooth years (QATYs), and incremental cost-effectiveness ratios (ICERs).
RESULTS: Improved diagnostic performance, particularly higher sensitivity, increased caries detection and shifted patients towards earlier and more active treatment pathways. These changes in early management influenced subsequent disease trajectories and contributed to downstream health gains. In the base-case analysis, AI-assisted diagnosis increased discounted costs by £1,071 per person and generated an additional 0.38 discounted QALYs and 0.08 discounted QATYs compared with conventional assessment. The resulting ICERs were £2,812 per QALY gained and £12,822 per QATY gained.
CONCLUSIONS: AI-assisted radiographic detection of dental caries may represent a cost-effective use of NHS resources in England. Its health and economic value depends on how improved detection influences subsequent treatment decisions and disease progression.