Cambios estructurales en la demanda de urgencias respiratorias en Chile (2017–2024): un análisis retrospectivo mediante modelamiento bayesiano de series temporales
European Journal of Health Research · 18.03.2026 · 12 · 1-18
DOI: 10.32457/ejhr.v12.3561 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Journal of Health Research, online erschienen am 18.03.2026. Verfasst von Eric Comincini Cantillo und 1 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective: to identify structural changes in respiratory emergency demand in Chile between 2017 and 2024 using a Bayesian time-series framework incorporating annual seasonality, day-of-week effects, and smooth change points, stratified by level of care. Materials and Methods: An ecological time-series study was conducted using official administrative records of 33,597,638 respiratory emergency visits in Chile from January 1, 2017, to December 31, 2024. Daily counts were stratified by care level: primary care emergency services (APS) and hospital emergency departments. Bayesian negative binomial regression models were fitted incorporating Fourier-based annual seasonality, day-of-week effects, and two a priori structural change points with smooth transitions aligned to key COVID-19 epidemiological milestones. Structural effects were summarized using multiplicative ratios and 90% credible intervals, and compared with classical methods including CUSUM, moving averages with Z-score thresholds, and deterministic segmentation.
Results
Of the 33,597,638 visits recorded, 24,617,474 occurred in primary care and 8,980,164 in hospital emergency departments. The series exhibited marked annual seasonality and two structural transitions consistent with the early and later pandemic phases. Both care levels showed multiplicative ratios below unity at the first change point and above unity at the second. The Bayesian approach demonstrated shorter detection latency than classical methods in retrospective comparison.
Conclusions
Respiratory emergency demand in Chile between 2017 and 2024 displayed structural changes clearly distinguishable from seasonal variation, with distinct patterns across care levels. Bayesian time-series modeling enabled more precise characterization of demand transitions, supporting its use as an analytical tool for health system surveillance in high-variability contexts.
Zitieren
APA
Cantillo, E. C., & Visbal, J. H. W. (2026). Cambios estructurales en la demanda de urgencias respiratorias en Chile (2017–2024): un análisis retrospectivo mediante modelamiento bayesiano de series temporales. European Journal of Health Research, 12, 1-18. https://doi.org/10.32457/ejhr.v12.3561
Vancouver
Cantillo EC, Visbal JHW. Cambios estructurales en la demanda de urgencias respiratorias en Chile (2017–2024): un análisis retrospectivo mediante modelamiento bayesiano de series temporales. European Journal of Health Research. 2026;12:1-18. doi:10.32457/ejhr.v12.3561
BibTeX
@article{Cantillo2026,
author = {Cantillo, Eric Comincini and Visbal, Jorge Homero Wilches},
title = {Cambios estructurales en la demanda de urgencias respiratorias en Chile (2017–2024): un análisis retrospectivo mediante modelamiento bayesiano de series temporales},
journal = {European Journal of Health Research},
year = {2026},
volume = {12},
pages = {1--18},
doi = {10.32457/ejhr.v12.3561}
}
EUVIMED