Technology Category
Data Mining & Knowledge Management: Anomaly/Outlier DetectionMachine Learning: Time-Series/Data StreamsSearch and Optimization: Metareasoning and Metaheuristics
DESA: An R Package for Detecting Epidemics using a School-Absenteeism Surveillance Framework
🤖 AI Summary
This study addresses the limited early-warning capability for seasonal influenza and other infectious diseases by proposing an absenteeism-based epidemiological surveillance framework. Methodologically, it integrates time-series modeling, epidemiological transmission simulation, and multi-metric early-warning evaluation into an end-to-end automated analytical pipeline. A key contribution is the development and public release of DESA—an open-source R package (available on CRAN)—which fills a critical methodological gap in R for absenteeism-driven outbreak detection. The framework supports fully automated processing from raw school absenteeism data to real-time outbreak alerts. Its timeliness and robustness are rigorously validated across diverse simulated epidemic scenarios. Results demonstrate an average lead time of 5–7 days prior to conventional case-based detection, substantially enhancing the responsiveness and efficiency of frontline public health interventions.
Technology Category
Application Category
📝 Abstract
Absenteeism of elementary school children has been shown to be effective in the early detection of an incoming influenza epidemic within a given population. This paper introduces DESA, an R package designed to: 1) model an epidemic using school absenteeism data, 2) raise an alert for an incoming epidemic using school absenteeism data, 3) evaluate the timeliness of the raised alert using different metrics, and 4) simulate community-level household populations, epidemics, and school absenteeism to facilitate research in related fields. This paper provides an overview of the functions in the package and demonstrates its complete workflow using simulated data generated within the package. DESA offers researchers and public health officials a tool for improving early detection of seasonal influenza epidemics or epidemics of other diseases. The package is available on CRAN, making it readily accessible to the R user community.
Problem
Research questions and friction points this paper is trying to address.
Modeling epidemics using school absenteeism data
Raising alerts for incoming epidemics via absenteeism
Evaluating alert timeliness with various metrics
Innovation
Methods, ideas, or system contributions that make the work stand out.
R package models epidemics using school absenteeism
Raises alerts for epidemics via absenteeism data analysis
Simulates populations and epidemics for research evaluation
🔎 Similar Papers
No similar papers found.
