Hierarchical Spatio-Temporal Transformer for Coherent Emergency Department Forecasting

2026-07-29

AI Summary

This study addresses the challenge of imbalanced resource planning in emergency departments caused by inconsistent multi-level demand forecasting. To this end, the authors propose HierSTT—the first end-to-end framework integrating hierarchical structure with Transformer architecture—which models demand across hospital, regional, and national levels through a spatiotemporal encoder-decoder. A key innovation is the introduction of a consistency-aware loss function that enforces cross-level coherence during joint optimization, overcoming limitations of conventional single-level models or post-hoc reconciliation approaches. Evaluated on a dataset encompassing 81 hospitals in Portugal, HierSTT achieves a 32% average reduction in WAPE compared to the best non-hierarchical deep learning baseline and significantly outperforms classical hierarchical reconciliation methods.

Abstract

Emergency Departments (EDs) are critical access points in healthcare systems, yet they face persistent pressure from unpredictable patient demand, seasonal surges, and non-urgent visits. Effective ED planning requires forecasts at multiple decision-making levels: hospitals need local demand estimates for staffing and bed management, regions require forecasts to coordinate healthcare units, and national authorities need system-wide projections for capacity planning. However, most existing approaches forecast ED demand independently at a single level, ignoring the hierarchy linking hospitals, regions, and national systems. This can produce incoherent predictions, where hospital-level forecasts do not aggregate consistently to regional or national demand. We propose HierSTT, a hierarchical Transformer-based framework for coherent multi-level ED forecasting. HierSTT jointly predicts hospital, regional, and national level demand in a single end-to-end model. A Temporal Fusion Transformer captures national dynamics, while spatio-temporal Transformer encoder-decoder modules model regional and hospital demand conditioned on higher-level forecasts. A coherence-aware loss penalizes cross-level inconsistencies during training. We further introduce a nationwide Portuguese ED dataset covering 81 hospitals across 5 regional health administrations, with heterogeneous covariates at each level. Experiments show that HierSTT reduces average WAPE by 32\% relative to the best non-hierarchical deep learning baseline and outperforms all classical hierarchical reconciliation methods, while producing near-coherent predictions across levels. Additional resources associated with this work are available at https://github.com/FilipaLino/HierSTT.

Problem

• Emergency Department forecasting
• hierarchical forecasting
• spatio-temporal prediction
• forecast coherence
• multi-level planning

Innovation

• Hierarchical forecasting
• Spatio-temporal Transformer
• Coherent prediction
• Emergency department demand
• Temporal Fusion Transformer