Abstract
OBJECTIVE: To systematically examine how digital health startups define and operationalize engagement in the post- coronavirus disease environment (2020-2025). PATIENTS AND METHODS: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines adapted for web-based literature, we systematically reviewed publicly available information from digital health startups founded or significantly operating between 2020-2025. We extracted engagement definitions from company websites, white papers, blog posts, and press releases. Definitions were coded by type (explicit, implicit, or nondefinition) and dimensional focus (behavioral, cognitive, affective, and social). Inter-rater reliability was assessed using Cohen's κ (κ=0.82). We conducted this systematic review from April 20, 2025, to May 21, 2025. RESULTS: We analyzed 64 engagement definitions from 30 digital health startups. Only 18.8% (n=12) were explicit definitions with clear measurement criteria, whereas 45.3% (n=29) were implicit definitions and 35.9% (n=23) were nondefinitions that mentioned engagement without defining it. The behavioral dimension dominated (64.1%, n=41), followed by social (28.1%, n=18), cognitive (21.9%, n=14), and affective dimensions (17.2%, n=11). Statistical analysis revealed significant associations between definition type and dimensional focus (P<.05). Based on our findings, we developed a taxonomy of engagement definitions and a 5-level engagement definition maturity model. CONCLUSION: Digital health startups predominantly use implicit or undefined engagement concepts with a strong behavioral focus. The proposed taxonomy and maturity model provide frameworks for standardizing engagement definitions across the digital health ecosystem, potentially improving measurement consistency, facilitating more meaningful comparisons between solutions, and establishing a baseline for evaluating effectiveness.