Navigating Evidence Constraints: Modeling Approaches for Digital Health Technology in NICE's Early Value Assessment

OBJECTIVES: The National Institute for Care and Health Excellence (NICE) launched the Early Value Assessment (EVA) program in June 2022, aiming to conditionally recommend digital health technologies (DHTs) for use in the National Health Service (NHS) while additional evidence is generated. EVAs face evidence constraints, therefore requiring decisions on how to robustly undertake economic analysis. This research aims to investigate the modelling methodologies and explore the logical frameworks and discussion underpinning the approaches used.

METHODS: A targeted literature review of all NICE EVA appraisals was undertaken and data were extracted to facilitate the review of modelling approaches used by the Evidence Assessment Group to evaluate cost-effectiveness of the appraised technologies.

RESULTS: Fifteen appraisals were identified. Of the 103 technologies evaluated, 57 were conditionally recommended (55%). The remaining 46 (45%) were not recommended for use in the NHS primarily due to lack of evidence. Of the appraisals included, the following approaches were adopted: cost-utility (47%), cost-effectiveness (27%), cost-comparison (13%), cost-minimization (7%), and 7% did not report the approach used. Five of the 15 appraisals (33%) adopted a decision tree, two (13%) used a Markov structure, and three (20%) took a hybrid approach using a decision tree to model short-term events followed by a Markov afterwards. Five (33%) did not clearly report the modelling structure. Evidence gaps were identified and the subsequent impact on model development was noted and discussed.

CONCLUSIONS: This research highlights the variety of approaches used in EVAs to produce suitable models despite reported evidence gaps. The simplification of complex disease and care pathways, due to insufficient data, potentially overlooks impactful costs and outcomes however, early collaboration with economic experts to identify parameters that are key drivers and signal the value of a technology, could guide evidence generation to facilitate faster adoption of innovative technologies through processes like EVA.