Rewiring care delivery through Digital Therapeutics (DTx): a Machine Learning-Enhanced Assessment and Development (M-LEAD) framework

Abstract

Abstract Background Digital transformation has sparked profound change in the healthcare sector through the development of innovative digital technologies. Particular attention should be devoted to Digital Therapeutics, which offers an innovative approach to disease management and treatment. The result is a landscape in which care delivery is increasingly patient-centered, data-driven, and based on real-time information. These technological innovations can lead to better patient outcomes and support for healthcare professionals. Digital transformation can, moreover, support healthcare systems by offering powerful tools to cope with the ever-increasing demand for care in an environment with limited resources. As these digital technologies continue to evolve, the healthcare field must be ready to integrate them into processes to take advantage of their benefits. This study aims to develop a framework for the development and assessment of Digital Therapeutics. Methods The study was conducted relying on a mixed methodology. 338 studies about Digital Therapeutics resulting from a systematic literature review were analyzed using descriptive statistics through RStudio. The software allowed studies to be described according to 33 variables. Three machine learning algorithms (K-NN, decision trees, random forests) were applied to analyze variables and find patterns in the data. The results of these analytical analyses were summarized in a framework qualitatively tested and validated through expert opinion elicitation in the form of semi-structured interviews and focus groups. Results The research provides M-LEAD, a Machine Learning-Enhanced Assessment and Development framework that recommends best practices for developing and assessing Digital Therapeutics. The framework takes as input Digital Therapeutics characteristics, regulatory aspects, study purpose, and assessment domains. The framework produces as outputs recommendations to design the Digital Therapeutics study characteristics, particularly the sources of evidence, study type and randomization, enrolled patients, study duration, comparators and arms, and outcomes. Conclusions The proposed framework seizes an opportunity and contributes to filling a relevant gap in Digital Therapeutics product development and assessment. The framework constitutes the first step toward standardized guidelines for the development and assessment of Digital Therapeutics. The results of this study may support manufacturers and inform decision-makers of the relevant results of the Digital Therapeutics assessment.