EUVIMEDAtklāt EUVIMED
Open access · CC BY 4.0

Joint Bayesian Nowcasting of Severe Acute Respiratory Illness and COVID‐19 Positives in Brazil

Alba Halliday iD, Oliver Stoner iD, Theo Economou iD, Leonardo Soares Bastos

Statistics in Medicine · 2026-04 · 45 (8-9)

DOI: 10.1002/sim.70529 ↗

At a glance

Journal article in Statistics in Medicine, published online on 2026-04-17. Written by Alba Halliday and 3 co-authors. Institution: MRC Centre for Global Infectious Disease Analysis, Imperial College London, London, UK. Major MEDLINE topics: COVID-19, Severe Acute Respiratory Syndrome. 26 references. The full text is freely available (open access, CC BY 4.0). Funded by: Engineering and Physical Sciences Research Council, HORIZON EUROPE Innovative Europe, FAPERJ.

AbstraktsSubjectsBibliogrāfiskie datiCiteRelated studies

Abstract (original language)

Surveillance of severe acute respiratory illness (SARI) in Brazil provides an early warning system for respiratory outbreaks, including COVID‐19, and statistical nowcasting methods are vital for correcting long and variable reporting delays that would otherwise hinder timely decision‐making. Coherent joint prediction of SARI and COVID‐positive SARI in Brazil could improve outbreak response planning and hospital resource management, but most existing nowcasting methods only target one outcome at a time. Furthermore, existing approaches usually focus solely on reconstructing recent incidence rather than forecasting future trends, which could support more proactive risk mitigation. Here, we propose a Bayesian hierarchical framework for joint nowcasting and short‐term forecasting of two outcomes, where one is a strict subset of the other. Building on the generalized‐Dirichlet‐multinomial (GDM) method for correcting delayed reporting, a new beta‐binomial component links SARI and COVID‐positive counts, while separate conditional GDM components capture their distinct delay patterns. To allow for changes over time in the level of disease and delay distributions, flexible latent effects are included in all model components. Using national surveillance data from 2021 to 2024, we conduct a 20‐date rolling prediction experiment across Brazil's 27 federative units. Compared to a well‐established Bayesian nowcasting approach, our joint model achieves about one‐third lower mean absolute error and continuous ranked probability score for contemporaneous nowcasts, with the largest gains in high‐incidence regions. Meanwhile, energy scores indicate improved calibration for joint forecasts of total and COVID‐positive SARI relative to comparable independent models.

Subjects

MeSH terms (MEDLINE)

COVID-19Severe Acute Respiratory SyndromeBayes TheoremHumansBrazilSARS-CoV-2ForecastingModels, StatisticalPandemicsIncidenceDisease Outbreaks

Author keywords

Bayesian nowcastingCOVID‐19infectious disease surveillanceprobabilistic forecastingreporting delayssevere acute respiratory illness (SARI)

Authors and institutions

  • MRC Centre for Global Infectious Disease Analysis, Imperial College London, London, UK.
  • School of Mathematics and Statistics, University of Glasgow, Glasgow, UK.
  • Mathematics and Statistics, University of Exeter, Exeter, UK.
  • Climate and Atmosphere Research Center, The Cyprus Institute, Nicosia, Cyprus.
  • Instituto de Matemática Pura e Aplicada e Tecnologia, IMPATech, Rio de Janeiro, Brazil.
  • Programa de Computação Científica, Fundação Oswaldo Cruz, Rio de Janeiro, Brazil.

ORCID profiles: Alba Halliday, Oliver Stoner, Theo Economou

Funding

  • Engineering and Physical Sciences Research Council
  • HORIZON EUROPE Innovative Europe
  • FAPERJ
  • CNPq
  • University of Glasgow
  • Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico

Cite

APA

Halliday, A., Stoner, O., Economou, T., & Bastos, L. S. (2026). Joint Bayesian Nowcasting of Severe Acute Respiratory Illness and COVID‐19 Positives in Brazil. Statistics in Medicine, 45(8-9). https://doi.org/10.1002/sim.70529

Vancouver

Halliday A, Stoner O, Economou T, Bastos LS. Joint Bayesian Nowcasting of Severe Acute Respiratory Illness and COVID‐19 Positives in Brazil. Statistics in Medicine. 2026;45(8-9). doi:10.1002/sim.70529

BibTeX
@article{Halliday2026,
  author = {Halliday, Alba and Stoner, Oliver and Economou, Theo and Bastos, Leonardo Soares},
  title = {Joint Bayesian Nowcasting of Severe Acute Respiratory Illness and COVID‐19 Positives in Brazil},
  journal = {Statistics in Medicine},
  year = {2026},
  volume = {45},
  number = {8-9},
  doi = {10.1002/sim.70529}
}
Start for free →