THE EFFECTS OF DIGITAL HOSPITAL COMMAND CENTERS ON EMERGENCY DEPARTMENT PATIENT THROUGHPUT AND CROWDING, SCOPING REVIEW

Abstract

Background and objective: ED overcrowding is a global problem linked to higher mortality, delays in care, and increased costs, driven especially by throughput bottlenecks such as prolonged boarding and high occupancy. DHCCs integrate EHR data, real‑time dashboards, and predictive models to actively manage beds and patient flow. This scoping review asked: What is the evidence on the effects of DHCCs on ED patient flow and crowding among adult patients?

Methods: A scoping review following Arksey and O’Malley’s framework searched PubMed, Scopus, and Web of Science using terms related to “digital hospital command centre,” “capacity command centre,” “AI patient flow,” “ED crowding,” “throughput,” and “length of stay.” Quantitative, qualitative, and mixed‑methods studies on DHCCs or similar digital tools reporting adult ED outcomes were included; non‑ED, non‑peer‑reviewed, and duplicate studies were excluded. Two reviewers independently screened and extracted data into an Excel matrix. From 250 records, 210 unique citations were screened, 47 full texts assessed, and 15 studies (2021–2025) included.

Results: The 15 studies comprised quantitative (47%), qualitative (20%), and mixed‑methods (33%) designs, assessing real‑time dashboards, ML‑based LOS and reattendance prediction, and pre‑triage alerts versus usual protocols. Quantitative studies reported marked boarding reductions (up to about 90%), LOS reductions of around 20% in some ML‑supported pathways, and occupancy maintained below ≈85% in several implementations. Qualitative work emphasized DHCCs’ role in real‑time bottleneck resolution and bed visibility, but highlighted barriers such as data quality issues, training needs, and implementation burden. Mixed‑methods evaluations of AI command centers showed signals of improved safety (e.g., modest mortality and readmission reductions) and positive return on investment, but inconsistent effects on LOS and subgroup‑dependent benefits.

Conclusion: Overall, current evidence suggests that DHCCs underpinned by AI analytics and real‑time dashboards can substantially improve ED throughput and reduce crowding, with additional potential safety and economic benefits. Nevertheless, heterogeneity in interventions, limited high‑quality trials, implementation barriers, and lack of standardized DHCC definitions constrain generalizability. Future work should include robust multi‑site evaluations, cost‑effectiveness analyses, and standardized frameworks integrating advanced ML/NLP to support precise and resilient ED flow management.