Challenges and opportunities of using common data models in rehabilitation

As momentum builds for value-based healthcare in rehabilitation, improved characterization of functional outcomes and rehabilitation utilization is increasingly necessary to better understand and enhance the value of care.This requires large, diverse datasets from multiple healthcare systems and care settings that reflect the variability inherent in healthcare delivery and the variety of settings in which an individual may receive rehabilitation.Curating these datasets has historically posed a major challenge in rehabilitation, as these data are collected for different purposes, stored in distinct databases, and structured differently.Common data models (CDMs) are designed to overcome these challenges and facilitate the integration of healthcare data from disparate sources (e.g., the electronic health record, billing records, payor records).CDMs standardize data by transforming diverse data sources into a shared structure and vocabulary to enable harmonization across healthcare systems.These models have facilitated transformative progress in other areas of healthcare; CDMs have been used to integrate rare disease data, create distributed networks for pharmacovigilance, and standardize metadata in medical imaging.However, to date, rehabilitation data remain largely absent from CDM-based initiatives.In this Special Communication, we aim to advocate for the untapped potential of CDMs to the rehabilitation community by: 1) introducing CDMs, 2) describing the potential value of CDMs to rehabilitation, 3) identifying rehabilitation-specific barriers to CDM use, 4) outlining current applications of CDMs in rehabilitation, and 5) offering a forward-looking vision for CDMs in rehabilitation research and practice.