BRIDGE Taxonomy: A Conceptual Framework for AI-Based Interventions to Reduce Diagnostic Delay in Rheumatology from Primary Care in Latin America

The BRIDGE taxonomy is an original conceptual framework for classifying artificial intelligence (AI) and clinical decision support system (CDSS) interventions by their primary clinical function along the primary care–rheumatology diagnostic continuum in Latin America. Six functional levels are defined: B (Búsqueda activa / Active Screening), R (Riesgo estratificado / Risk Stratification), I (Inicio de manejo provisional / Provisional Management Initiation), D (Docencia / Training), G (Gestión de coordinación asistencial / Care Coordination), E (Educación del paciente / Patient Self-Management Education). Each level is operationally defined by required inputs, clinical output, and a measurable expected endpoint. The taxonomy was developed through iterative conceptual analysis grounded in a structured narrative review (SANRA criteria, 72 articles selected from 347 records). It is currently in initial development phase and has not been subjected to formal Delphi or expert panel validation. A companion peer-reviewed article is under review in Global Rheumatology by PANLAR (Manuscript ID: GR-2026-78666, accepted with minor revisions).