Global, regional and national burden of chikungunya: force of infection mapping and spatial modelling study.

INTRODUCTION: Chikungunya virus, an arbovirus transmitted by Aedes mosquitoes, causes epidemics in tropical regions with potential risk in higher latitudes. Our aim is to estimate the global, regional and national burden of chikungunya across affected and environmentally suitable at-risk regions. METHODS: We used a random forest model to predict force of infection and estimate chikungunya burden at high spatial resolution (5×5 km) using covariates from climatic, socioeconomic and ecological domains. We used a focal scenario to estimate the observed burden (lower bound) and an at-risk scenario to estimate the potential burden (upper bound) of chikungunya transmission. RESULTS: We predicted global long-term average annual force of infection at 0.012 (95% UI: 0.007 to 0.019) for focal scenario and 0.013 (95% UI: 0.005 to 0.03) for at-risk scenario in 103 countries. We estimated global chikungunya burden annually of 14.4 million (95% UI: 11.0 to 17.8 million) infections and 0.96 million (95% UI: 0.56 to 1.6 million) disability-adjusted life years (DALYs) in the focal scenario, and 34.9 million infections (95% UI: 26.7 to 43.1 million) and 2.3 million DALYs (95% UI: 1.4 to 3.8 million) in the at-risk scenario for 2020. The chronic phase accounts for 54% of chikungunya burden, with relatively higher burden among 40-60-year-old population, with mortality disproportionately affecting children under 10 and adults over 80. CONCLUSION: While chikungunya transmission has high geographical uncertainty, high force of infection is not limited to tropical regions and is distributed across all continents. Our estimates of chikungunya burden are useful for prioritisation of regions and target age groups for chikungunya vaccine introduction.