Presentation (CTI)

OBJECTIVES: Colorectal cancer (CRC) is the fourth most common cancer in the UK. Artificial intelligence (AI) technologies with polyp detection and/or characterisation functions aim to support endoscopists, which may ultimately reduce the risk of CRC. The aim of this research was to assess whether the addition of specific AI-supported colonoscopy technologies to colonoscopy represents a cost-effective use of National Health Service (NHS) resources.
METHODS: A de novo economic model was developed to assess the cost-effectiveness of AI technologies compared to colonoscopy without AI. Adenoma detection rate (ADR) and diagnostic accuracy data were the key clinical outcomes included in the model. The model used a lifetime horizon and an NHS and personal social services (PSS) perspective.
RESULTS: The economic analysis suggested that the introduction of all technologies would result in increased quality-adjusted life years (QALYs) and reduced costs (with the exception of Discovery™, which led to increased costs, and an incremental cost-effectiveness ratio [ICER] of £8,670) compared to colonoscopy without AI.
CONCLUSIONS: Despite increased uncertainty for certain technologies, there is some evidence for all technologies of an improved ADR with AI, with no major concerns about impacts on procedure durations or adverse events, compared to colonoscopy without AI. The use of AI to help detect colorectal polyps is unlikely to increase costs for the NHS, and there may be benefits beyond detection. However, the use of AI to help characterise polyps requires further research.