Spatial regression analysis of dengue incidence rate in the Philippines from years 2017-2023

• Main Authors: Tubise, Raquel Neera Dignos, Gonzales, Jhean Neil Custodio
• 格式: text
• 語言: English
• 出版: Animo Repository 2025
• 主題: Dengue > Philippines > Statistics; Spatial analysis (Statistics); Mathematics; Public Health
• 在線閱讀: https://animorepository.dlsu.edu.ph/etdb_math/48; https://animorepository.dlsu.edu.ph/context/etdb_math/article/1052/viewcontent/2025_Tubise_Gonzales_Spatial_regression_analysis_of_dengue_incidence_rate_in_the_Phili.pdf
• 機構: De La Salle University

Dengue remains an endemic public health concern in the Philippines, affecting millions annually due to its climatic and geographical conditions. Despite various strategic interventions, the government continues to face challenges in reducing the dengue incidence rate. This study aims to address the pressing need for statistical prediction of dengue hotspots through the analysis of dengue incidence rates. A spatial analysis of dengue incidence in the Philippines from 2017 to 2023 was conducted, with a focused spatial regression analysis for the year 2023. Spatial autocorrelation was assessed using the Global Moran’s I statistic, employing the most suitable weight matrices. Additionally, cluster detection, including the identification of hotspots and coldspots, was performed using Local Moran’s I. Consistent dengue hotspots were identified in the northern and southern provinces, with the northern regions exhibiting elevated incidence throughout the entire year and the southern provinces in the earlier half of the year. Coldspots were prominently observed in the central provinces throughout the year, providing valuable insights for nationwide health policy formulation. Furthermore, spatial regression analysis, utilizing the "LeSage and Pace" method, was conducted by initially modeling the dengue incidence rate with the Spatial Durbin Error Model (SDEM). This model was validated as the best fit for the data when compared to other simpler nested models namely Spatial Lag of X Model (SLX), Spatial Error Model (SEM), and Ordinary Least Squares (OLS). The spatial regression analysis highlighted key factors influencing dengue incidence, including population dynamics, healthcare accessibility, and environmental conditions. Notably, variables such as the hospital bed population ratio, number of doctors, rainfall amount, minimum temperature, maximum temperatures, and relative humidity were found to significantly impact dengue incidence rates.