ORIGINAL RESEARCH article
Front. Public Health
Sec. Infectious Diseases: Epidemiology and Prevention
Volume 13 - 2025 | doi: 10.3389/fpubh.2025.1620914
This article is part of the Research TopicEmerging Arboviruses in the Americas: Epidemiology, Public Health Impact, and Future PreparednessView all 7 articles
Geospatial Clustering Identifies Dengue Burden Hotspots Across Brazilian Municipalities, 2024
Provisionally accepted
- 1Mailman School of Public Health, Columbia University, New York City, United States
- 2Instituto Keizo Asami, Universidade Federal de Pernambuco, Recife, PE, Brazil
- 3Rutgers Global Health Institute, New Brunswick, NJ, United States
- 4Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, United States
- 5Instituto Keizo Asami, Departamento de Bioquímica, Universidade Federal de Pernambuco, Recife, PE, Brazil
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Dengue virus (DENV) remains a major and recurrent public health challenge in Brazil. In 2024, the country experienced its largest recorded epidemic, with more than six million probable cases and substantial pressure on hospital systems. The burden was highly heterogeneous across space, underscoring the value of municipal-scale geospatial analysis to identify actionable hotspots. We conducted a nationwide clustering analysis to delineate dengue hotspots during the 2024 epidemic using case notifications and hospitalizations from the national SINAN surveillance system, with denominator populations from the Brazilian Institute of Geography and Statistics (IBGE). We calculated standardized case and hospitalization rates per 100,000 population for all municipalities and applied a multivariate density-based spatial clustering algorithm (DBSCAN) that integrated municipality centroids with epidemiologic burden. Parameters (eps, minPts) were chosen from k-distance inspection and sensitivity analyses. We further evaluated temporal stability through monthly DBSCAN runs using a common parameter set and summarized the share of municipalities and population that remained in a cluster for at least three consecutive months. We also examined climatic associations by pairing municipal dengue indicators with gridded precipitation from CHIRPS at monthly lags 0-3. Across Brazil, DBSCAN identified 25 high-burden municipal clusters, with 5,111 municipalities (92.6%) clustered and 408 (7.4%) classified as noise. Several clusters had average case rates >20,000 per 100,000, concentrated in Minas Gerais, Paraná, and Bahia, while some high-incidence municipalities remained geographically isolated and unclustered. Rainfall correlated positively with incidence at a ~2-month lag. Hospitalization-only clustering produced similar geography, and monthly analyses showed persistence of high-burden clusters. Treating Cluster 1 as a low-burden background focuses interpretation on high-intensity clusters (ID >1). By integrating environmental and temporal dimensions into a scalable DBSCAN framework, this study provides a reproducible and operationally relevant tool for targeting vector control and outbreak response in Brazil.
Keywords: Dengue, DBSCAN, Spatial Epidemiology, Brazil, Hospitalization, Public Health Surveillance, rainfall, clustering
Received: 30 Apr 2025; Accepted: 16 Sep 2025.
Copyright: © 2025 Sena, Herrera, Martins and Lima Filho. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Brena Figueiredo Sena, brena.sena@gmail.com
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.