Using Real-World Data on Depression from EHR-based Research Networks: A Scoping Review

Depression poses a significant global public health challenge, necessitating innovative research to understand its epidemiology and management. Electronic health record (EHR) research networks offer a powerful tool to study depression at scale, yet remain underutilized. This scoping review summarizes the extent of depression research ongoing in EHR networks.

Following the Arksey and O'Malley framework and PRISMA guidelines, we searched PubMed, Scopus, EBSCOHost, and Google Scholar in September 2024, identifying 166 studies from 1211 records. Included studies used EHR networks like TriNetX, All of Us, and the Million Veteran Program (MVP) to investigate depression, defined broadly to include various depressive disorders. Covidence with custom large language model (LLM) plugin was used to aid screening and extraction processes.

Depression research in EHR networks is limited, with TriNetX (36 studies) and All of Us (24 studies) the most utilized platforms. Populations studied were predominantly from the United States (125 studies), followed by Canada (5) and European countries (15 combined). Common predictors analyzed included age (58 studies), gender/sex (56 studies), and race/ethnicity (45 studies).

EHR networks hold vast real-world data for advancing depression research, but underutilization highlights the need for better accessibility to enhance future studies.