Umeå University's logo

Context: Dengue is a major public health threat in the Lao People’s Democratic Republic (Laos) and Thailand. Dengue transmission is ecologically complex. Concurrently identifying both climate and landscape-based risk factors for dengue virus transmission is necessary to improve dengue prevention and control efforts in Laos and Thailand. Objectives: The objective of this study was to determine how changes in climate (temperature and rainfall), and land use (e.g. built-up areas, agricultural crops, fruit orchards, rubber plantations) and land cover (e.g. evergreen and deciduous forests, permanent and temporary wetlands) affect dengue risk in four provinces in southern Laos and north-eastern Thailand during 2002 to 2019.

Methods: A conditional autoregressive Bayesian spatiotemporal modeling framework was used to analyze the risk of dengue by spatiotemporal variations in land use and land cover (LULC) and climatic parameters.

Results: The average annual temperatures in the study area increased by 0.44–0.94 °C during the study period. The model indicated that an increase of 1 °C in weekly average temperatures (up to a 29 °C threshold level) increased the average dengue risk by up to 24% in the two Lao provinces and 18.9% in the two Thai provinces. The model suggested that a rainfall increase of 1 mm up to 60 mm increased dengue risk by 1.8–3.2%. A 0.6–1.6% increase in built-up land use increased dengue risk by 1.8–6.9%. Built-up areas and rubber plantations were positively associated with dengue in the Ubon Ratchathani province of Thailand, while wetlands were negatively associated with dengue cases in the Savannakhet province of Laos.

Conclusions: Changes in dengue risk were clearly related to increases in rainfall and temperature as well as changes in LULC in both Laos and Thailand. These insights may inform community-based dengue control activities by targeting geographically localized areas (microgeographic scale) to deploy dengue control activities more effectively in these highly endemic regions.