OP345 Evaluation Of An Artificial Intelligence-assisted Service For Cardiac Monitoring As Part Of A National Institute For Health And Care Excellence (NICE) Digital Health Technology Pilot. (December 2021)

• Record Type: Journal Article
• Authors: Erskine, Jamie; Goddard, Kate; Chalkidou, Anastasia
• Is Part Of: International journal of technology assessment in health care. Volume 37(2021)Supplement 1
• Page Start: 15
• Page End: 15
• Publication Date: 2021-12
• Subjects: Medical technology -- Periodicals; Technology assessment -- Periodicals; 610.28
• DOI: 10.1017/S0266462321001021
• Languages: English
• ISSNs: 0266-4623

Abstract

Abstract : Introduction: Zio XT Service was one of five Digital Health Technologies (DHTs) to be assessed by the National Institute for Health and Care Excellence (NICE) as part of their evaluation pilot. The King's Technology Evaluation Centre (KiTEC) act as an External Assessment Centre for NICE and worked on this pilot evaluation. The service comprises a I-Lead ECG patch, an inbuilt software that makes use of artificial intelligence (AI) algorithms to record, store and analyze ECG traces, and a team of cardiac physiologists. Methods: Although the methods were based on NICE's existing Medical Technologies Guidance Process, they were modified to suit the assessment of DHTs. The process was split into two sections, with the option to discontinue the assessment if it was considered that insufficient evidence was available for the technology. Clinical experts and patients were consulted through the process and clinical, economic and technical evidence was considered. Costs for three care pathways were modelled. Results: A total of thirty relevant clinical studies were identified, with a further study being reviewed as part of a separate technical assessment, focusing on the AI component of the technology. Four of the studies were considered to be pivotal to the decision problem, one of which was a Randomized Controlled Trial. The technology was found to have a greater diagnostic yield than a standard ambulatory monitor, however diagnostic accuracy measures were absent in the literature. Three economic models were developed to represent three care pathways: patients with syncope, patients who have had a stroke or transient ischaemic attack and a third model assessing downstream costs associated with stroke treatment. Conclusions: Digital Health Technologies and Artificial Intelligence Technologies pose novel and unique challenges to health technology assessment (HTA) bodies. Zio XT Service is a diagnostic tool, with both human and AI input, making it a particularly complex technology to assess. This work serves as a case-study in the evaluation of DHTs and AI and the lessons learned may contribute to the development of guidelines for such technologies.