Minding the Gap in Sentinel Surveillance Networks: An Analysis of Brazilian Indigenous Areas
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Optimizing the allocation of sentinel surveillance sites, particularly aiming early pathogen detection, remains a persistent challenge. Equally critical is ensuring the inclusion of vulnerable and often underserved populations as targets when prioritizing surveillance sites.
Objective
This study assesses the current coverage of the respiratory pathogen surveillance network in Brazil and proposes an optimized rearrangement of sentinel locations that balance the coverage of Indigenous populations while integrating country-wide human mobility patterns.
Methods
We collected the locations of Indigenous populations living in Special Indigenous Sanitary Districts (DSEIs) from the Brazilian Ministry of Health, and derived mobility route estimates using the Ford-Fulkerson algorithm applied to input air, road, and water transportation data for the whole country. To optimize sentinel city selection for sample collection, we applied a linear optimization algorithm designed to maximize two key objectives: 1) representation of Indigenous regions and 2) coverage of human mobility patterns, thereby enhancing early pathogen detection. Validation of the strategy was performed by obtaining the current list of cities in Brazil’s influenza sentinel network combined with a health attraction index from the Brazilian Institute of Geography and Statistics, which enabled us to assess the suitability and potential benefits of our optimized surveillance network.
Results
Our optimization model provides actionable recommendations for pathogen sampling locations, enhancing early detection. By selecting 199 cities, we create a more representative sentinel network, covering all DSEIs by rearranging 108 cities (58.3%), addressing gaps in 9 of 34 previously uncovered regions. This improves nationwide mobility coverage by 16.8 percentage points—from 52.4% to 69.2%—compared to Brazil’s current sentinel system. Additionally, all newly selected cities serve as hubs for medium to high-complexity healthcare, with 37.9% of DSEI cities lacking coverage in the existing flu sentinel network.
Conclusions
We propose a strategic framework for sentinel site placement that maximizes DSEI coverage and aligns with human mobility patterns across Brazil, improving the overall effectiveness of disease surveillance, particularly in these critical regions and often underserved populations.
