Development of a Resource-Adaptive Acute Care Architecture for Low-Resource Hospitals: A Yemen-Informed Clinical and Operational Framework

Background: Acute-care delivery in resource-constrained hospitals requires two distinct judgments: what the patient clinically requires and what the facility can safely deliver at the present time. Existing triage, severity, early-warning, and syndrome-specific tools address important components of this problem but do not constitute a single operational architecture for capability matching, escalation, referral, and reassessment. Objective: To develop a testable Resource-Adaptive Acute Care Architecture (RAACA) that integrates established clinical assessment functions with a time-stamped representation of facility capability while preserving the independence of clinical risk assessment. Methods: This conceptual framework was developed through targeted evidence mapping of WHO emergency-care resources and peer-reviewed literature on emergency care, triage, deterioration assessment, and prehospital systems in low- and middle-income countries. Requirements were extracted, mapped to a seven-layer patient journey, and synthesized into explicit constructs, decision rules, operational states, and a prospective validation pathway. RAACA does not introduce a composite clinical score. Instead, it proposes a four-state operational capability taxonomy (R0–R3) and a structured capability-mismatch process. Results: RAACA comprises seven linked layers: (1) acuity/triage, (2) immediate physiologic threat, (3) severity and deterioration, (4) syndrome-specific risk, (5) required and available capability, (6) operational disposition and referral, and (7) reassessment and quality feedback. The architecture separates patient state, system state, and operational decision, and treats referral/transfer as a clinical risk-control process. R0–R3 is explicitly defined as an unvalidated operational construct rather than a severity or mortality score. Conclusions: RAACA provides a testable architecture for coordinating validated clinical tools with changing facility capability. Its principal research question is whether resource-state classification can be assigned reproducibly, represent actual capability, and improve predefined process measures without altering clinical risk classification. Prospective multicentre validation is required before clinical effectiveness can be inferred.