# The Composite Digital Therapeutic Index (cDTI): A Multidimensional Framework and Proof-of-Concept Application to FDA-Authorized Treatments

## TL;DR

This paper introduces a new framework called the Composite Digital Therapeutic Index (cDTI) to evaluate and compare the value of digital therapeutics based on multiple factors like efficacy and safety.

## Contribution

The novel contribution is the development of a multidimensional index to benchmark FDA-authorized digital therapeutics using standardized metrics.

## Key findings

- CT-132 scored significantly higher than reSET-O on the cDTI due to stronger efficacy, engagement, and evidence quality.
- The cDTI successfully differentiated therapeutic profiles of two FDA-cleared digital therapeutics across multiple domains.
- The framework highlights variability in trial quality and safety reporting among digital therapeutics with similar regulatory status.

## Abstract

Introduction

Prescription digital therapeutics (PDTs) are an emerging class of regulated software-based interventions with increasing FDA authorizations. However, there is no standardized framework to evaluate and benchmark their overall therapeutic value.

We aimed to develop and validate a Composite Digital Therapeutic Index (cDTI) framework. This proof-of-concept application integrates efficacy, engagement, quality of evidence, and safety profile from registrational trials to compare PDTs using publicly available regulatory summaries, registries, and study readouts.

Methods

We developed a scoring framework incorporating four domains (efficacy, engagement, evidence quality, and safety), each quantified and combined into a composite index. Efficacy was calculated using standardized mean difference (SMD; Hedges' g) adjusted for statistical significance. Engagement was based on the mean proportion of patients achieving therapeutic module completion. Evidence quality was graded using an adapted American Academy of Neurology Class of Evidence framework. Safety was assessed based on treatment-emergent adverse events (TEAEs) using a hierarchical penalty system. In a proof-of-concept application of the framework, we applied cDTI to two FDA-cleared first- and second-generation PDTs, reSET-O for opioid use disorder (Pear Therapeutics, Boston, MA, USA) and CT-132 for episodic migraine (Click Therapeutics, New York, NY, USA), respectively.

Results

CT-132 achieved a cDTI score of 0.296, substantially (1287%) higher than reSET-O, which scored 0.023. CT-132’s superior performance was driven by a statistically significant efficacy outcome (adjusted SMD 0.33), high engagement (89.7%), Class I evidence quality with sham and double-blind, and no treatment-related adverse events. In contrast, reSET-O showed a non-significant primary outcome (adjusted SMD 0.0895), moderate engagement (63.0%), Class III evidence quality without patient blind and usual-care control, and missing systematic safety reporting requiring a conservative penalty.

Discussion

The cDTI successfully differentiated the overall therapeutic profiles of two PDTs across multiple clinical and usability domains. This multidimensional approach highlights how digital therapeutics with similar regulatory status can differ meaningfully in trial quality, efficacy, engagement, and safety. The cDTI offers a transparent, reproducible method for comparing PDTs and may aid stakeholders including providers, payers, regulators, and patients in decision-making about digital therapeutic adoption and coverage.

Conclusion

The cDTI provides a reproducible framework to evaluate PDTs based on clinical, usability, and safety parameters. Future work aims to expand the framework across all FDA-cleared PDTs and explore incorporation of real-world effectiveness, equity, and cost-effectiveness domains.